Linguistic Variation - Volume 25, Issue 2, 2025

Ewe (Kwa, Niger-Congo) has a construction known in the literature as the nyá-construction (Ameka 1991, 2005a; Collins 1993; Duthie 1996; Adjei 2014). The logical internal argument of the construction occurs in subject position and the logical external argument is either absent on the surface or represented in the construction as a for-PP. In this paper, I consider the syntax of the Ewe nyá-construction, exploring data from the Tongugbe dialect. I show that the nyá-construction shares properties with English middles. I demonstrate that the agent or experiencer for-PP that may occur in the nyá -construction is its external argument, projected in Spec vP. Further, I argue that even if the for-PP is not overtly realized, it is syntactically projected in Spec vP, contrary to theories like Bruening 2013. The analysis I put forward provides support for the Theta-Criterion, which forces all arguments to be syntactically projected.

This paper investigates the emergence of differential object marking (DOM) in the Asia Minor Greek dialect of Pharasa (PhG) under contact with Turkish. We show that DOM in Turkish and PhG are both instances of structural accusative case and DOM can be formally modeled as context sensitive dependent case. We propose that two factors caused the emergence of DOM in PhG, namely (i) case neutralization in indefinite contexts, and (ii) an increase in the number of V-NP idioms borrowed from Turkish where the NP is in bare form. These perturbations led to a significant change in the overall data created by the community resulting in mixed input for the younger generations. Once the amount of bare NPs passed a certain threshold, a divergent grammar became inevitable. We test our proposal using an abductive generalization learning algorithm based on the Tolerance Principle and running a number of simulations. Our simulation results confirm our hypothesis.

We present a cross-linguistic dataset of force-flavor combinations in modal elements, which currently contains information on modal semantics in 24 languages and is accessible at https://github.com/EdinburghMeaning​Sciences/modals_database. We discuss theoretical motivations for constructing the dataset, the data collection methodology, as well as the design and the format of the dataset. We also present four case studies using the data: (i) assessment of cross-linguistic generalizations on force/flavor variability; (ii) exploration of generalizations in the lexicalization of negative modality; (iii) investigation of the typology of the morphological encoding of modal strength; and (iv) examination of how future contributes to modality. These case studies illustrate that the dataset supports in-depth assessment of potential cross-linguistic generalizations as well as theory-informed investigations of cross-linguistic variations in modal semantics.

We find vowel harmony systems in many non-standard varieties of modern Armenian. It has been speculated that these may have acquired vowel harmony due to contact with Turkic varieties (Scala 2018). On the basis of an exploration of the synchrony and typology of Armenian vowel harmony, consideration of historical changes that could have caused harmony to develop, and evaluation of new data bearing on the origins of backness and rounding harmony in Oghuz, we propose that the vowel harmony systems of the modern Armenian dialects show evidence of having been influenced by Turkish, but the numerous differences between Armenian and Turkish vowel harmony point against a straightforward copying of the Turkish phonological system. We theorize that vowel harmony in Armenian arose due to a combination of language-internal and ‑external factors: vowel shifts in some Armenian dialects, alongside universal analytic and channel biases, provided the necessary preconditions for the development of vowel harmony by the 11th century AD, prior to the arrival of Turkic speakers in the Armenian homeland. Extensive contact with Turkic vowel systems may then have encouraged the phonologization of this assimilation process, but in strikingly different ways than are found in Turkic languages.

This paper presents a survey of complementation strategies that are employed in Pazar and Ardesheni dialects of Laz, an endangered South Caucasian language that has been in contact with Turkish for decades. Our survey reveals strong signatures of contact in that nominalization patterns not present in cognate systems but present in Turkish seem to have been transferred into Laz. An intriguing asymmetry concerning the two dialects is that the Turkish noninalization pattern seems to have been directly copied into Pazar Laz whereas Ardesheni Laz seems to have developed a novel complementation pattern that mixes finite complementation with nominalization. Furthermore, in both dialects, complement clauses that denote propositions remain untouched by the dominant nominalization strategy in Turkish, raising questions on grammatical asymmetries in susceptibility to contact effects.

Tshila /ts^hílà/ is a Kalahari Khoe language spoken in central-southeastern Botswana by approximately 300 people (Schwab & Collins 2024). It is an SOV language. Tshila has a copular predicate nominal construction with the word order [Noun Noun COP]. The nouns in this language can vary in size according to factors such as the projection of phi-feature morphology and the presence of constituent negation and focus. Also possible in Tshila is predicate inversion. The interaction of the size of the DPs along with the presence of inversion shows interesting constraints on nominative Case assignment and phi-agreement. The analysis of this construction supports a universal hierarchy of phi-features (Harley & Ritter 2002; Béjar 2003), not only within nominals but also on phi-probes. This analysis unifies Case assignment and phi-agreement as a single operation (Chomsky 2000; 2001; Platzack 2006; Legate 2008; Georgi 2017), which supports the view that Case assignment is not postsyntactic.