Layout

You are using an outdated browser. Please upgrade your browser to improve your experience.

OVERABUNDANCE IN THE ARABIC DIALECT OF TUNIS:
A DIACHRONIC STUDY OF PLURAL FORMATION

INES DALLAJI

INES GABSI

Department of Oriental Studies (University of Vienna)

Austrian Centre for Digital Humanities (Austrian Academy of Sciences)

Abstract: During fieldwork in Tunis in the framework of the Vienna-based project Linguistic Dynamics in the Greater Tunis Area: A Corpus-based Approach (TUNICO), we conducted research on plural formation in the dialect of Tunisia's capital. The main purpose was to collect data absent in historical sources and in the corpus which we are currently compiling in order to add this data to the dictionary of Tunis Arabic which we are about to create. As the historical sources (e.g. Singer's grammar of the dialect of the Medina of Tunis, the Takroûna glossary by Marçais and Guîga) list some nouns with two or more plural forms, another purpose of our research was to find out which of the listed forms are still used. The research was based on a list of approximately 200 nouns and conducted with the help of almost 30 informants, most of whom were younger than 30. This is due to the fact that the focus of the TUNICO project lies on producing a corpus of spoken youth language and on drawing conclusions about the current linguistic situation in Tunis and its suburbs. This paper is a first approach to study overabundance in the Arabic dialect of the greater Tunis area. It contains all the examples for nouns with more than one possible plural form which we could gather during our fieldwork. The comparison of the collected data with forms listed in historical sources enables diachronic research and allows drawing conclusions about processes of linguistic dynamics.

Keywords: TUNICO project, Tunis Arabic, plural formation, overabundance, diachronic study.

1. Introduction

This paper presents the preliminary results of our research on plural formation and overabundance in Tunis Arabic, which we conduct within the framework of a project with the title Linguistic Dynamics in the Greater Tunis Area: A Corpus-based Approach. TUNICO, as this project funded by a three-year grant of the Austrian Science Fund is called for short, is based at the Department of Oriental Studies of the University of Vienna and the Austrian Centre for Digital Humanities (ACDH) at the Austrian Academy of Sciences, and is headed by Stephan Procházka and Karlheinz Mörth. It has grown out of an ongoing joint initiative of the above mentioned institutions that was initiated several years ago and goes by the name Vienna Corpus of Arabic Varieties (VICAV). VICAV aims at the collection of digital language resources documenting varieties of spoken Arabic1.

Within the framework of the TUNICO project, two digital language resources are produced. The first one is a corpus of transcribed texts in the Arabic dialect of Tunis. To be more precise, we create a corpus of spoken youth language, by which we understand the speech of the generation below 35. The majority of dialogues and narratives which we are in the process of transcribing were recorded during our first fieldwork in Tunis in September 2013. The second language resource which the TUNICO project is about to produce is an online dictionary that is fed with data from the corpus as well as some other sources which will be mentioned below. The tool that is used for creating this dictionary is called Viennese Lexicographic Editor (VLE).

The corpus and other sources do not only function as providers of lexicographic data. They are also used to investigate a number of selected topics dealing with the morphology and syntax of

contemporary Tunis Arabic like, for example, plural formation and overabundance, by which we understand the coexistence of more than one plural form in the nominal system (see section 4). The fact that most of the relevant literature dates back to the last century enables the compilation of a micro-diachronic dictionary and explains why this study of plural formation in Tunis Arabic is diachronic too.

The paper is organized as follows. In section 2, we give an overview of the sources on which the dictionary entries are based and of the historical sources that are used for a diachronic study of plural formation. Section 3 provides information on our research on plural formation in the greater Tunis area in September 2014. Section 4 summarizes general information on the formation of Arabic plurals with a special focus on Tunisian Arabic and includes a definition of the term “overabundance”. Section 5 contains all the examples for nouns with more than one possible plural form we could gather during our fieldwork. The collected data are compared with forms listed in historical sources in order to draw conclusions about processes of linguistic dynamics. The historical developments this diachronic study reveals are summed up in section 6.

2. Sources

Besides the corpus which we are currently compiling, one of the contemporary sources on which the dictionary entries are based is a textbook by Veronika Ritt-Benmimoun. This textbook serves as teaching material for the Tunisian Arabic class at the University of Vienna but has not been published yet. The second contemporary source is an amateur glossary from 2010 by Karim Abdellatif called Dictionnarie «le Karmous» du Tunisien. Unfortunately, it lacks plural forms, which is why it is unhelpful for a study of plural formation in Tunis Arabic.

The main historical source is the 800-page grammar of the Arabic dialect of the Medina of Tunis by Hans-Rudolf Singer, which was published in 1984. One of the objectives of the TUNICO project is to feed the micro-diachronic dictionary with the rich lexical material which this grammar contains.

Besides this grammar by Singer, the historical sources which we use for this diachronic study are the Peace Corps English – Tunisian Arabic Dictionary by Rached Ben Abdelkader, Abdeljelil Ayed and Aziza Naouar published in 1977 and The Morphology of the Arabic Dialect of Tunis, a PhD thesis by Ferid Chekili from 1982. The oldest historical source is Hans Stumme's Grammatik des Tunisischen Arabisch, which dates back to 1896. The eight-volume glossary of the Textes arabes de Takroûna compiled by Marçais and Guîga (1958-61) is very useful for comparison purposes, specifically to check if nouns which, according to our study, are prone to multiple pluralization are listed with the same possible plural forms in this glossary as well.

3. The fieldwork

In September 2014, we conducted fieldwork in Tunis in order to gather plural forms absent in the textbook, the corpus, and the historical sources which we are using for creating our dictionary (see section 2), and to find out whether or not certain plural forms listed in historical sources are still in use. During a two-week stay, we asked 30 informants to form plurals of approximately 200 nouns. Most of the informants were between 18 and 30 years old and were either born in Tunis or in one of its suburbs, or have lived in the greater Tunis area most of their lives. The nouns with more than one possible plural form this paper contains were identified and gathered during the evaluation of the results of our fieldwork. The examples listed below are supplemented by forms occurring in the TUNICO corpus and in the historical sources we use for a diachronic study of plural formation (see section 2). In brackets it is added by how many people a certain plural form was given and in which historical source or sources it was found. As one may notice, the total number of plural forms we received varies from example to example. This is due to the fact that sometimes we asked small groups of people to form the plurals of the nouns on which our research was based and we did not always receive an answer by each member of these groups.

4. Plural formation

Laks (2014: 5) outlines that Arabic has two types of plurals: a suffix-based sound plural and a template-based broken plural. To put it in other words, “plural morphology in Arabic involves two ways of deriving words, concatenative and nonconcatenative” (Albirini & Benmamoun 2014: 855).

Brustad (2000: 52) writes that “morphologically, Arabic distinguishes sound and broken plurals, and collective and distributive ones”. Since 20 of the approximately 200 nouns on which our fieldwork in Tunis was based were collective and singulative nouns, it could be of interest to conduct a separate study of collective and distributive plurals in Tunis Arabic, similar to what Brustad (2008) investigated in Levantine Arabic.

As far as concatenative plural formation in Tunisian Arabic is concerned, there are /-īn/-, /-āt/-, and /-a/-suffixed plurals (Singer 1984: 455, Chekili 1982: 186). Chekili (1982: 186) writes that the most common plural forms in Tunis Arabic are /-āt/-suffixed plurals2.

As regards nonconcatenative formation, Nada Tomiche (1964: 172) and Manwel Mifsud (1994: 92) mention that all Arabic dialects have reduced the number of broken plural patterns from that of Old Arabic. Without considering vowel qualities or distinguishing whether a four-consonant noun consists of four radicals or of three radicals and a prefix, we have counted 14 broken plural patterns in Singer's Grammatik der arabischen Mundart der Medina von Tunis (1984: 576-607).

As the title of this paper reveals, the focus of our study of plural formation in Tunis Arabic lies on overabundance. According to Mörth and Dressler (2014: 249), who quote Anna M. Thornton (2011: 360, 2012: 183sq.), overabundance refers to the “coexistence and rivalry of two or more cellmates within the same cell of the same paradigm”. The reason why Thornton (2011: 360 & 362, 2012: 183sq.) prefers the term “cell-mates” to the term “doublets” is that it can be more than two forms that realize the same cell.

With respect to Arabic plural forms, overabundance is still a little-studied phenomenon. Both the fact that the glossary compiled by Marçais and Guîga (1958-61) contains ample evidence for the coexistence of more than one plural form in the nominal system of the dialect spoken in Takroûna and the results of our fieldwork suggest that overabundance in nominal plural forms can be expected of contemporary Tunis Arabic too.

5. Overabundance in Tunis Arabic

The results of our research indicate that nouns which trace back to the Old Arabic patterns CvCC or CvCvC tend to exhibit overabundance:

  1. bḥaṛ (“sea”): bḥūṛāt (22, 1 C3), bḥūṛa (2, TATk II/14: 234), bḥaṛṛāt (1), bḥūṛ (0, Peace Corps5: 340, TATk II/1: 234)
  2. bdan (“body”): bdūnāt (15), bdannāt (10), (a)bdān (0, Peace Corps: 51, TATk II/1: 250)
  3. qbaṛ (“grave”): qbūṛāt (17, Peace Corps: 174, Chekili 1982: 198-qburāt), qbūṛa (7), uqbṛa (4), qbūṛ (0, Stumme 1896: 83, TATk II/6: 3090)
  4. qlam (“pencil”): qlammāt (18, Singer 1984: 460, Chekili 1982: 187, Peace Corps: 284), (a)qlām
    (6, TATk II/6: 3291sq.), uqᵊlma /aqᵊlma (2, TATk II/6: 3291sq.-uqulma), qlūmāt (1)
  5. ʕirs (“wedding”): ʕṛūsāt (24, Singer 1984: 463, Peace Corps: 440), (a)ʕṛās/(a)ʕrās (5, 1 C)
  6. nahž (“alley, lane”): unᵊhža (22, Chekili 1982: 198), nhūžāt (2, Singer 1984: 463), nhūža (1)
  7. yūm (“day”): ayyām (14, Singer 1984: 582-āyām, Peace Corps: 100, TATk II/8: 4428-ayām), ayyāmāt (13, Singer 1984: 463-āyāmāt, TATk II/8: 4428-ayāmāt)

Chekili (1982: 187) writes that the final consonant of nouns with the singular pattern CCvC is geminated immediately before the plural suffix /-āt/ as seen in bhaṛṛāt, bdannāt and qlammāt.

Only in the case of qlam, the attachment of /-āt/ to the singular pattern is the most frequently mentioned formation. The plural form qlammāt is also listed in three historical sources.

As far as bḥaṛ, bdan and qbaṛ are concerned, the plural forms that were mentioned by the largest number of informants are bḥūṛāt, bdūnāt and qbūṛat, which are a combination of templatic variation and concatenative formation. The forms bḥūṛāt and bdūnāt are not listed in any of our historical sources. The broken plurals bḥūṛ, (a)bdān and qbūṛ were not formed by any of our informants, but are mentioned in some of the historical sources.

As the example ʕirs shows, the majority of informants combined concatenative and nonconcatenative formation (ʕṛūsāt), whereas in the case of nahž, the most frequently mentioned formation was nonconcatenative. Singer (1984: 463), however, only lists the so-called double plurals ʕṛūsāt and nhūžāt.

As regards yūm, the number of people who formed the irregular plural ayyām and the number of those who mentioned the double plural ayyāmāt is almost the same. Both forms are listed by Singer (1984: 463 & 582) and in the Takroûna glossary (TATk II/8: 4428), but with different spelling.

The following examples show that some nouns with the singular pattern CvCCa are also prone to overabundance:

  1. xaṭwa (“step”): xaṭwāt (11, TATk II/3: 1131), xuṭwāt (11), xṭāwi (4, Singer 1984: 588, TATk II/3: 1131)
  2. ṛabṭa (“bunch; knot”): ṛabṭāt/ṛabtāt (14, TATk II/3: 1420sq.), ṛbāyṭ/ ṛbāyiṭ (7, Chekili 1982: 185), ṛbuṭ (1) ṛbāṭi (0, Singer 1984: 588, TATk II/3: 1420sq.)
  3. ṛukba (“knee”): rkāyb/rkāyib (16, Singer 1984: 93, Peace Corps: 211, Chekili 1982: 192, Stumme 1896: 85, TATk II/3, p. 1570), ṛukbāt (7, TATk II/3, p. 1570), rkub (4), rkābi (0, Singer 1984: 588)

In the case of xaṭwa and ṛabṭa, the majority of informants formed the sound feminine plural. As far as ṛukba is concerned, more than half of the informants mentioned the broken plural rkāyb/rkāyib. A broken plural with the same pattern also exists for ṛabṭa, but only seven informants formed it. The broken plurals ṛbāṭi and rkābi, which are listed by Singer (1984: 588), were not mentioned by any of our informants. For ṛukba, the grammar by Singer (1984: 93 & 588) even contains two possible plural forms: rkāyb and rkābi6. What must be noted is that only the four informants who were older than 30 formed the broken plural xṭāwi, which is listed by Singer (1984: 588) as well as in the Takroûna glossary (TATk II/3: 1131). This may indicate that the generation below 30 does not use this broken plural of xaṭwa anymore. The fact that the majority of informants formed the sound feminine plurals xaṭwāt and ṛabṭāt, while Singer (1984: 93, 588) only mentions nonconcatenative formation, may suggest that the younger generation of speakers of Tunis Arabic tend to employ regular patterns when forming the plural of feminine nouns.

The following example also supports the theory of a tendency towards regular patterns:

  1. ʕāda (“habit, custom”): ʕādāt (18, Peace Corps: 98, TATk II/5: 2737), ʕwāyd/ʕwāyid (10, Peace Corps: 98, TATk II/5: 2737)

The plural forms ʕādāt and ʕwāyd/ʕwāyid are both listed in historical sources, but the majority of informants formed the sound feminine plural ʕādāt.

Nouns with the singular pattern CCā form their plural by inserting /w/ or /y/ between the final vowel and the plural suffix /-āt/ (Chekili 1982: 188). As far as the formation of the plural of ġṭā is concerned, not only Chekili (1982: 188) but also the majority of informants mentioned the sound feminine plural ġṭāwāt, although the majority of the consulted historical sources list the broken plural uġṭya:

  1. ġṭā (“lid, cap, cover”): ġṭāwāt (20, Chekili 1982: 188, TATk II/6: 2831), uġṭya (5, Singer 1984: 239, Stumme 1896: 84, Peace Corps: 222, TATk II/6: 2831)

This again could indicate a tendency towards regular patterns.

The following two examples with the singular pattern CvCCv̄ Ca, however, show that not all

young speakers of Tunis Arabic tend to form the plural of feminine nouns by attaching the suffix /-āt/:

  1. ẓaṛbīya (“carpet”): ẓṛābi (18, Peace Corps: 67, Stumme 1896: 89, TATk II/4: 1659), ẓṛāba (7), aṛbīyāt (2, TATk II/4: 1659)
  2. kuṛṛāsa (“notebook, exercise book”): kṛāṛis (22, Peace Corps: 52), kuṛṛāsāt (6)

The fact that kuṛṛāsāt has not been found in any of our historical sources but was formed by six informants could at least indicate that a tendency towards regular pluralization is about to emerge.

As far as intensive nouns with the singular pattern CvCCv̄ C are concerned, Cohen (1975: 189) writes that the attachment of /-īn/ predominates in the dialect of the Muslims of Tunis, whereas in the dialect of Jews, there is a slight tendency to plural formation with /-a/.

According to Chekili (1982: 191) and Singer (1984: 461), however, nouns with the singular pattern CvCCv̄ C form their plural by attaching the suffix /-a/7. Stumme (1986: 76 & 79sq.) points out

that occupational nouns with the pattern CvCCv̄ C can form a plural by suffixing /-a/, but a large number of these nouns form their plural with the masculine plural ending /-īn/. In the Bedouin dialect of the southern Tunisian region of Douz, intensive nouns can form the sound plural by attaching the suffix /-a/ as well as by attaching /-īn/ (Ritt-Benmimoun 2014: 214).

The following examples show that with the exception of ṛassām, the majority of our informants formed the plural with /-a/. The first two examples, ḥažžām and ṛassām have /-a/- and /-īn/-suffixed plurals:

  1. ṛassām (“painter”): ṛassāmīn (14), ṛassāma (12)
  2. ḥažžām (“hairdresser”): ḥažžāma (14, Chekili 1982: 191), ḥažžāmīn (7)
  3. ḥawwāt (“fish seller”): ḥawwāta (23), ḥawwātīn (0)
  4. xabbāz (“baker”): xabbāza (21, Stumme 1896: 80), xabbāzīn (0)

The only noun among the singular nouns on which our research was based that forms two semantically distinct plurals is ʕīn, which itself has two different meanings, “source” and “eye”. The plural of “source” is the broken plural ʕyūn. For “eyes”, ʕīnīn, a dual noun with the value of a plural noun, is formed.

6. Historical developments

This section outlines the processes of language change which we have observed during our small diachronic study of plural formation and overabundance in Tunis Arabic.

Stumme (1896: 83) writes that nouns that form broken plurals with the pattern CCūC, which in most cases trace back to the Old Arabic singular pattern CvCC, are very frequent. One of the examples he lists is qbūṛ (“graves”), which was not mentioned by any of our informants (see section 5, example 3). He further writes that in a few cases, the plural forms of nouns which trace back to the Old Arabic singular patterns CvCC or CvCvC can be a combination of broken and sound plural. He gives the example faṛš (“bed”), for which he mentions the plurals fṛūš, fṛūša and fṛūšāt (Stumme 1896: 83). Singer (1984: 463) also writes that the suffixes /-āt/ and /-a/ are attached to some broken plural forms.

In addition to the above mentioned forms ʕṛūsāt, nhūžāt and āyāmāt (see section 5, examples 5, 6 & 7), he notes the example fṛūšāt but, unlike Stumme, he neither mentions fṛūš nor fṛūša. In the EnglishTunisian Arabic dictionary published by Ben Abdelkader et al. (Peace Corps: 45), fṛūšāt is also the only form mentioned. As far as qbaṛ is concerned, to our knowledge, Singer's grammar does not contain any plural form. According to Ben Abdelkader et al. (Peace Corps: 174) and the majority of our informants, the plural of qbaṛ is qbūṛāt (see section 5, example 3). The fact that qbūṛ or fṛūš are only listed by Stumme implies that these broken plurals with the pattern CCūC are not in use any more. In the case of bḥaṛ and bdan, none of our informants mentioned the broken plurals bḥūṛ and (a)bdān, although they are listed in two historical sources (see section 5, examples 1 & 2). The most frequently mentioned forms were the double plurals bḥūṛāt and bdūnāt, which do not occur in any of our historical sources.

To summarize the above mentioned observations, most of the broken plurals which the consulted historical sources list seem to be no longer in use in contemporary Tunis Arabic, as spoken by the younger generation. A shift towards a combination of concatenative and nonconcatenative formation, especially towards attaching the suffix /-āt/ to broken plurals, can be observed. Stumme (1896: 90) remarks that double plurals are not as common in Tunis Arabic as in more Western dialects, for example in Moroccan Arabic. We do not have enough evidence to support or refute this theory, since we have not compared plural forms used in Tunis Arabic with plural forms in Moroccan dialects. But the results of this diachronic study lead us to assume that the use of double plurals in Tunis Arabic has become more frequent over the past few decades than it was at the end of the 19th century.

The examples 8, 9, 11 and 12 in section 5 (xaṭwa, ṛabṭa, ʕāda and ġṭā) show that young speakers of Tunis Arabic tend to regular pluralization, whereas historical sources either list the broken plural only or reveal that both forms, the sound and the broken plural, are used.

As regards intensive nouns with the singular pattern CvCCv̄ C, the opinions in the historical sources are divided. The four examples in section 5 (ṛassām, ḥažžām, ḥawwāt and xabbāz) hint at a tendency to pluralization by attaching the suffix /-a/, but the number of examples is too small to answer the question as to whether the formation with /-īn/ is less frequent than the formation with /-a/.

7. Conclusion

Karlheinz Mörth and Wolfgang Dressler (2014: 250) write in their article on German plural doublets that “the existence of doublets can often be seen as a particular stage in a process of language change”. The results of our diachronic study already hint at a language change in terms of plural formation, as they reveal a decline of nonconcatenative formation and a development towards double plurals and regular patterns. In order to make more profound remarks regarding plural formation in contemporary Tunis Arabic, and to answer the question as to whether or not plural forms listed in historical sources are still in use and which plural forms are prone to overabundance, we will not only have to continue analyzing our corpus, which is still in progress, but we will also have to conduct further fieldwork in the greater Tunis area.

References

Abdellatif, Karim. 2010. Dictionnaire “le Karmous” du Tunisien. http://www.fichier-pdf.fr/2010/08/31/m14401m/.

Albirini, Abdulkafi, & Benmamoun, Elabbas. 2014. “Concatenative and Nonconcatenative Plural Formation in L1, L2, and Heritage Speakers of Arabic”, The Modern Language Journal 98 (3). 854-871.

Ben Abdelkader, Rached, Ayed, Abdeljelil, & Naouar, Aziza. 1977. Peace Corps English-Tunisian Arabic Dictionary. http://www.eric.ed.gov/PDFS/ED183017.pdf. (Peace Corps)

Brustad, Kristen. 2000. The Syntax of Spoken Arabic: A Comparative Study of Moroccan, Egyptian, Syrian and Kuwaiti Dialects. Washington, DC: Georgetown University Press.

Brustad, Kristen. 2008. “Drink Your Milks! -āt as Individuation Marker in Levantine Arabic”, Gruendler, Beatrice, & Cooperson, Michael (eds.), Classical Arabic Humanities in Their Own Terms: Festschrift for Wolfhart Heinrichs on his 65th Birthday Presented by his Students and Colleagues, Leiden-Boston: Brill.

Chekili, Ferid. 1982. The Morphology of the Arabic Dialect of Tunis. PhD Thesis. London: University College.

Cohen, David. 1975. Le parler arabe des juifs de Tunis. Tome II. Étude linguistique. Paris-La Haye: Mouton.

Laks, Lior. 2014. “The Cost of Change: Plural Formation of Loan Words in Arabic”, Zeitschrift für Arabische Linguistik 60. Wiesbaden: Harrassowitz. 5-34.

Mifsud, Manwel. 1994. “Internal Pluralization in Maltese: Continuity and Innovation”, Caubet, Dominique, & Vanhove, Martine (eds.), Actes des premières journées internationales de dialectologie arabe de Paris. Paris: INALCO. 91-105.

Marçais, William, & Guîga, Abderrahmān. 1958-61. Textes arabes de Takroûna. II. Glossaire. 8 vol. Paris: Centre Nationale de la Recherche Scientifique. (TATk II)

Mörth, Karlheinz, & Dressler, Wolfgang U. 2014. “German Plural Doublets with and without Meaning Differentiation”, Rainer, Franz, Gardani, Francesco, Luschützky, Hans Christian, & Dressler, Wolfgang U. (eds.), Morphology and Meaning. Selected Papers from the 15th International Morphology Meeting. Amsterdam: John Benjamins. 249-259.

Mörth, Karlheinz, Procházka, Stephan, & Dallaji, Ines. 2014. “Laying the Foundations for a Diachronic Dictionary of Tunis Arabic. A First Glance at an Evolving New Language Resource”, Abel, Andrea, Vettori, Chiara, & Ralli, Natascia (eds.), Proceedings of the XVI EURALEX International Congress: The User in Focus. 15-19 July 2014, Bolzano/Bozen. Bolzano/Bozen: EURAC research. 377-387.

Ritt-Benmimoun, Veronika. 2014. Grammatik des arabischen Beduinendialekts der Region Douz (Südtunesien). Wiesbaden: Harrassowitz.

Singer, Hans-Rudolf. 1984. Grammatik der arabischen Mundart der Medina von Tunis. Berlin, New York: de Gruyter.

Stumme, Hans. 1896. Grammatik des tunisischen Arabisch nebst Glossar. Leipzig: Hinrichs.

Thornton, Anna M. 2011. “Overabundance (Multiple Forms Realizing the Same Cell): A Noncanonical Phenomenon in Italian Verb Morphology”, Maiden, Martin, Smith, John Charles, Goldbach, Maria, & Hinzeli, Marc-Olivier (eds.), Morphological Autonomy: Perspectives from Romance Inflectional Morphology. Oxford: Oxford University Press. 358-381.

Thornton, Anna M. 2012. “Reduction and Maintenance of Overabundance. A Case Study on Italian Verb Paradigms”, Word Structure 5 (2), 183-207.

Tomiche, Nada. 1964. Le parler arabe du Caire. Paris-La Haye: Mouton.

Notes
1.

For further information on VICAV and TUNICO see Mörth, Procházka & Dallaji 2014 as well as https://acdh.oeaw.ac.at/vicav/ and https://acdh.oeaw.ac.at/tunico/ [last accessed 15 September 2015].

2.

Out of 818 singular nouns on which Chekili's study was based, 338 form their plural by attaching the suffix /-āt/ (Chekili 1982: 186).

3.

1 C (< corpus) means that this form was used by one of the young Tunisians whose speech was recorded and transcribed for the TUNICO corpus.

4.

TATk II is the abbreviation for the eight-volume glossary of the Textes arabes de Takroûna (see references).

5.

Peace Corps refers to the Peace Corps English-Tunisian Arabic Dictionary published in 1977 by Ben Abdelkader et al. (see references).

6.

By adding the example rkāybi “meine Knie” in a footnote, Singer (1984: 588) points out that suffixes can only be attached to the form rkāyb.

7.

Out of 60 nouns with this singular pattern which were part of Chekili's study of plural formation, only three have /-īn/suffixed plurals (Chekili 1982: 189).