This special issue of Toronto Working Papers in Linguistics (TWPL), “Critical perspectives and new trends in Arabic sociolinguistics,” features research presented at the international workshop Arabic Sociolinguistics: New Trends and Perspectives, held virtually at the University of Toronto on November 23, 2024. The workshop brought together junior and senior scholars in Arabic sociolinguistics to foster transnational and cross-generational dialogue, amplify underrepresented voices, and highlight innovative methodological and theoretical approaches, including quantitative, qualitative, and experimental research. Reflecting the theme of the special issue, the selected papers critically engage with current sociolinguistic questions, addressing variation, language ideology, identity, and the social functions of Arabic varieties in diverse contexts. Of the eight contributions presented at the workshop, four are included in this special issue, exemplifying the field’s breadth, rigor, and evolving directions.
Focusing on dialect contact in Beirut, this study investigates whether Palestinian Arabic (PA) is converging with Lebanese Arabic (LA), the majority variety, in different linguistic components. Using the comparative variationist framework (Poplack and Tagliamonte 2001), the study combines synchronic and diachronic data sources to compare two variables in LA and PA: a phonological variable known as imala, involving the word-medial raising of /a:/ to [e:], and future temporal reference. Results reveal evidence of contact-induced change in PA in terms of word-medial imala, while the future system in PA diverges from that in LA. The findings for imala are consonant with previous claims that socially salient aspects of segmental phonology are particularly susceptible to convergent change (Trudgill 1986), because speakers’ awareness of word-medial imala and its iconic association with LA facilitate its adoption. On the other hand, the future is not subject to social assessment or normative commentary of any kind. These findings attest to the propensity of phonological and indexical variables to engage in convergent change.
Terms of address have not been explicitly or deeply explored from a stylistic perspective in the Arab world and beyond. This sociopragmatic study demonstrates how 227 tokens of love terms of address by 24 couples in six Egyptian TV series (2016–2024) construct the style of a specific social group and (re)shape societal stereotypes. A style is identified by the frequent co-occurrence of a linguistic feature with a specific social group more than others, serving social functions (Mejdell 2006). Performance is linked with styles and stereotypes by displaying linguistic variations among social groups, reflecting and shaping societal perceptions of social identities and linguistic practices (Schilling-Estes 1998). I argue that address terms are often a stylistic choice, where specific lexical meanings align with particular social groups and serve social goals beyond these meanings, and that their social use on screen is not arbitrary but governed by social rules and societal perceptions. Young upper-class men are depicted as using the style of love terms of address to manage their love relationships rather than simply express affection, reinforcing and negotiating societal and broader claims about gender.
Minority or minoritized languages are often restricted to ingroup use by their own speakers and indexically related to them. Nevertheless, minority languages and their speakers are never isolated; they are part of a multilingual ecology and included in local language hierarchies. In this paper, I focus on the occurrence and role of Siwi in the interaction between Siwi and Arabic speakers in the Siwa Oasis. In this multilingual space, a national minority language, Siwi Amazigh, is spoken by the majority of the population, and different varieties of the dominant language, Arabic, are spoken as L1 by a smaller portion of the population and as L2 by virtually everyone. Data from mixed-group conversations show that although Arabic varieties far outscore Siwi, as one would expect, everyday practices are characterized by the hybridity and heterogeneity typical of multilingual ecologies. This paper provides an accurate portrait of conversational habits in Siwa and depicts Siwi fluctuating between an identity marker value, due to its vitality, and an identity maker value, which builds on it.
This chapter examines nominal constructions in (Iranian) Persian with the special focus on the distribution of Ezafe. Ezafe is an element which links a nominal to its modifiers and arguments. Persian has nominal modifiers of different types. They can be either prenominal or postnominal. The modifiers appearing in a postnominal position, always require Ezafe. Furthermore, it is demonstrated that the use of Ezafe is not restricted only to modification, since it also links a noun to its complement. The use of Ezafe is not restricted only to noun phrases either since Ezafe can link a nominal preposition to its complement as well.
This chapter describes nominal constructions in the Central variety of Taleshi (Northwestern Iranian). In Taleshi, noun phrases and the majority of adpositional phrases are head-final. Some nominal modifiers as well as the complement of postpositions are linked to the head of the phrase by a linking element––known as Reverse Ezafe (Stilo 2004). In this chapter, we show that the Reverse Ezafe appearing with possessors and adpositions is different from the Reverse Ezafe appearing with attributive adjectives. We also address some of the phonological constraints on the occurrence of the two types of Reverse Ezafe.
This paper describes nominal constructions in Sangesari (Northwestern Iranian). In Sangesari, similar to the pattern found in Caspian languages, some nominal modifiers as well as the complement of nominal postpositions are linked to the head of the phrase by a linking elementknown as Reverse Ezafe (Stilo 2004). In this paper, we show that the Reverse Ezafe used with attributive adjectives is different from the one used with possessors and adpositions. The former has a restricted usage, limited only to certain lexical and phonological contexts, while the latter is obligatory in nominal contexts. The examination of adpositional phrases in particular illustrates a different property of Reverse Ezafe in Sangesari compared to its counterpart in other Reverse Ezafe languages examined in this volume (i.e., Balochi and Caspian languages).
This chapter provides a description of noun phrases in the Ardalāni variety of Sorani Kurdish. In Ardalāni Kurdish, the surface word order within the Noun Phrase is head initial and post-nominal modifiers are linked to the head of the phrase by a linking element called Ezafe (Windfuhr 1989). We show that the realization of Ezafe is phonologically conditioned in Ardalāni, unlike other varieties of Kurdish. However, crucially, we show that in definite contexts, Ezafe obligatorily appears irrespective of the phonological context, and it takes a special definite form.
The Kirundi augment is a vowel found before noun class prefixes. I argue that the augment is an epenthetic vowel whose surface characteristics, including deletion, retention and weight, are conditioned by prosodic structure. While the augments of different noun classes present apparent asymmetries, my account unifies all surface representations under a single underlying structure. My arguments mainly focus on three nominal structures: simple nominals (syntactically φP), locative phrases (PP) and linker phrases (KP). I draw on the argument that both nP and DP are syntactic phases that correspond to prosodic words, as well as the observation that epenthetic vowels commonly pattern as weightless to quantity-sensitive phonological processes.
In this paper, we examine the structure of Kirundi nominals and the elements that appear on their left periphery, in particulardemonstratives, the locative prefixes mu, ku and i, and the nominal linker na. We analyze these structures from the perspective of the phonology-syntax interface. We show that the Universal Spine Hypothesis, which assumes a fixed order of the functional categories inside the nominal, allows us to explain both the syntactic properties and phonological realization of nominals and demonstrative, locative and linker phrases. We conclude that all of these elements have different syntactic statuses and semantic roles in the nominal structure, despite some previous claims in the literature. Our proposal also adds to the typological picture of the structure of nominal and locative phrases and the role of the augment in Bantu.
Languages have different ways of counting nouns, involvinga variable combination of pluralisation, combination withnumbers, and quantification. Established cross-linguisticliterature in this field like that of Chierchia (2019) suggestsa tripartite typological division between number-marking(e.g., English), classifier (e.g., Mandarin), and number-neutral (e.g., English) languages. In any case, the literatureargues for certain universals irrespective of type like thedivision of nouns into number-counting (e.g., pieces of meat)and kind-counting (e.g., pork and beef). In comparison withcross-linguistic typology and neighbouring languages likeLingala, Tshiluba does show affinities with the number-marking category with categories like fluid substancesneither able to change class nor combine directly withnumerals. However, there are other affinities with number-neutral languages like in the interpretation of quantifiers forthese fluid mass nouns, in this case a buunyi which can mean“many” bottles of water or “much” water. Ultimately, thetypological system is present but the motivation for it is morediscursive in that there can be countable and uncountableiterations of words like tshi-manu “wall” as opposed tocertain words being inherently (un)countable.
The Kirundi verbal suffix -an is notably polysemous with three possible meanings: reciprocal, associative, and antipassive. The contexts in which the antipassive is ambiguous or distinct raise questions for -an’s syntactic and semantic analysis. In this paper, I examine new data which demonstrate that -an’s primary function is to add a relation between the verb and an implicit argument. These data support an analysis where -an is a specialized high applicative head (ApplREL) which adds the semantics of such a relation. The distribution of the different meanings is then accounted for by differential semantic selection and binding. This analysis thereby simplifies -an to its relational role and accounts for the ambiguous and non-ambiguous contexts that characterize the suffix. Additionally, this paper addresses various questions about -an’s antipassive status and potential issues with the applicative approach.
This paper discusses the morpheme -ik in the Bantu language Kirundi (JD62). Dom et al. (2018) put forth a Proto-Bantu 'neuter middle marker' following Kemmer's (1993) 'middle domain', from which a Kirundi -ik can clearly be derived. They find that derivations of this marker occur in anticausative, agentless passive, and passive constructions in various Bantu languages. This paper demonstrates that Kirundi -ik is an anticausative marker, distinct in use and meaning from the true passive (marked with -u). This follows from tests in Alexiadou et al. (2006) and Gluckman and Bowler (2016). After determining its anticausative nature, I will discuss -ik's syntactic properties. Using the layering approach of change-of-state verbs as posited by Alexiadou et al. (2006) and Kratzer (2005), I will argue that -ik is an overt realization of Voice, providing evidence for a null semantic Voice in the crosslinguistic account of anticausatives theorized in Alexiadou et al. (2015).
Kirundi has two principle ways of expressing negation: nti-/si- and ta-. One occurs before subject-marking in matrix clauses (primary/NEG1) and the other after subject-marking in embedded clauses or following A’-movement (secondary/NEG2) respectively. Thus, the aim of this paper is to investigate the distinction between primary and secondary negation in Kirundi. I argue that the different contexts and positions for negation are due to left peripheral phenomena. More specifically, when C0 is filled in subordinate clauses and following A’-extraction, primary negation is blocked from occurring. In these cases, secondary negation surfaces instead. This analysis accounts for their complementary distributions.
This paper provides a description and analysis of morphological causative constructions in Kirundi. I show that the causative inter- pretation of Kirundi’s two causative suffixes, -i and -ish, can be explained by a single syntactic structure. To do so, I demonstrate that the phonological and syntactic properties of -ish causatives re- quire a bimorphemic account wherein the suffix is derived from two functionally different heads: an unaccusative v head embedded by a higher Causative head. I present new empirical data collected from fieldwork in order to support these claims and fill a gap in the exist- ing documentation of the argument structure of Bantu causatives. In addition to presenting a singular account of Kirundi causative syn- tax, the proposal sheds light on cross-linguistic variation in the struc- ture of v/Voice and valency-effecting morphology, with implications for formal typologies of causatives.
