Proceedings Published by the LSA (Linguistic Society of America)
Not a member yet
4323 research outputs found
Sort by
Disambiguating quantity judgements: mass/count and extra-grammatical cues
Full text linkComparative quantity judgements are a useful probe into the semantics of the mass/count distinction, where count nouns usually trigger cardinal comparisons (more dogs), and mass nouns trigger non-cardinal measurement (more rice). However, exceptions like ‘object’ mass nouns (furniture) and ‘mixed’ comparatives (more gold than diamonds) complicate this pattern. In such cases there is often a mismatch between the mass/count status of the noun and the criterion for comparison, which challenges our understanding of the mass/count distinction and how it affects quantity judgements. We propose that these mismatches reflect a systematic ambiguity, where the mass/count distinction is one of the factors influencing disambiguation. Using a new experimental method focused on ambiguity judgements instead of truth-value judgements, the results support the traditional semantic encoding of the mass/count distinction, with operations of ‘packaging’ and ‘grinding’ triggered by extra-grammatical factors
Semantics of finite complement clauses and scope islandhood
Full text linkThis paper investigates the correspondence between the semantics of a finite complement clause and its scope islandhood. Via comparison of the semantics of canonical attitude verbs, e.g. believe and claim, with that of clause-embedding verbs like ensure and prove, whose complement clauses are not scope islands (Farkas and Giannakidou 1996, Barker 2022, Palucci 2024, a.o.), this paper argues for two claims. First, while complement clauses of attitude verbs have been argued to denote predicates of individuals with propositional content (Kratzer 2006, Moulton 2009, 2015, Elliott 2020, a.o.), those of ensure-type verbs denote predicates of events (without propositional content). Second, finite complement clauses that denote predicates of events are not scope islands
The ‘queen bee’ of the highlands: An etymological analysis of the Albanian xhubleta
Full text linkThe Albanian xhubleta (indefinite xhublete) is a traditional dress worn by women in the isolated northern highland regions of Albanian and Kosovo. It is considered to be one of the oldest traditional folk costumes in the Balkans. Many scholars have noted the ancient features of the dress (Coon 1950: 15-16; Stipcevic 1977: 90; Belkaid 2013; Gjokaj 2021: 81-2, vol. I). The only proposed etymology for the word xhubleta is found in Topalli (2017: 1604-5), which posits it as a loan from Turkish. Difficult from both the phonological and morphological perspective, I posit a different etymology, with regular phonological change from the Indo-European starting point of *h1su-melit \u27having good honey\u27, a possessive adjectival compound type found elsewhere within the Indo-European language family. Relevant to this proposal is the specialized status of the bee in the Albanian law code, the Kanun, the aesthetically similar appearance of the xhubleta with the Cretan skirt (Belkaid 2013: 12), and the Greek myth of the goddess of \u27bees and honey\u27 Melissa, who nourished the infant Zeus on the island of Crete. All of these may indicate an ancient origin of the xhubleta.
Basic clause negator in Sadat Tawaher Sign Language
Full text linkSign languages (SLs) generally have several manual signs to negate sentences, usually with one sign serving as the basic clause negator and with its function being only to reverse the polarity of a clause without adding any additional semantic content. We identify the basic clause negator in Sadat Tawaher Sign Language (STSL), a SL that emerged in a single household in a small Iranian village around sixty years ago. While STSL has several manual negators, all of which may serve as sentential negators, we argue that one negator, NEGbasic, is the basic clause negator. The data includes both isolated sentence productions and story-telling elicited from native STSL signers. The evidence for NEGbasic comes from distributional frequency, semantic function, negative concord, and negative responses
Minimal wordlist size for a phonological profile: New evidence from Kra-Dai languages
Full text linkThis study replicates, and extends to Kra-Dai languages, earlier work on minimal wordlist size needed to make a phonological profile of a language. Previous work on Australian languages recommended approximately 400 randomly sampled words to comprise a minimally complete profile in terms of reliably capturing every phoneme, and with accurate distribution. We survey 55 Kra-Dai languages to show that a longer minimal list is necessary, which we attribute to typological differences like larger consonant/vowel inventories. Given the widespread use of short wordlists in fieldwork, these results hold significance for designing language documentation surveys as well as projects that use legacy wordlist data
Revisit de re presuppositions
Full text linkThis study investigates the availability of de re interpretations for presuppositions embedded under attitude predicates in Mandarin Chinese. While previous work has proposed substitution-based mechanisms to account for de re readings, I argue instead for a unified account grounded in accommodation. A picture-assisted acceptability judgment experiment tested four presupposition triggers in Mandarin: ye (‘also’), buzai (‘no longer’), jiu (‘only’), and yishidao (‘realize’) under vary-ing belief contexts. The results show that only ye and buzai allow de re readings in contexts where the subject either denies or is ignorant of the presupposed content. These two triggers are also used flexibly in contexts that do not fully entail their presuppositions. Together, these findings challenge substitution-based accounts and instead support a two-step accommodation process: an initial Weak Commitment Assumption (WCA), followed by a context-sensitive strengthening. I argue that de re presupposition is best understood as a function of contextual accommodation, rather than substitution
Tonal alternations in attributive constructions in Mwaghavul
Full text linkMwaghavul is an underdocumented Chadic language spoken in Plateau State, Nigeria, by approximately 150,000 people (Blench 2011). Mwaghavul has tonal lowering in associative constructions, where the first nominal in the construction surfaces with low tone, regardless of its tone in isolation (Arokoyo & Fwangwar 2019). However, tonal lowering is not fully predictable, as some high tone nominals surface as mid tone in associative constructions, instead of low. Number of syllables, vowel length and quality are not consistent predictors, as there are minimal pairs for high tone alternations. We investigate the phonetics of these high tones, to determine whether two phonetically distinct high tones have been incorrectly documented as one, or whether one phonetic high tone has two phonological behaviours. The f0 of 77 tokens in isolation and 561 tokens in associative constructions was extracted at 10 points using Prosody Pro (Xu 2013). In isolation, high tones that become mid in associative are visually distinct from those that become L, with approximately 15-20Hz difference throughout the tone duration. Linear mixed effects models confirm this difference is statistically significant. The presence of separate high and superhigh tones in Mwaghavul indicates that the phonetic implementation of the floating low tone is realized differently depending on the pitch of the original tone. This suggests that the original tone is not deleted, but rather dissociated and present, affecting the realization of the tonomorpheme in an unusual pattern that is not commonly attested.
An analysis of “causative” forms in Okinawan: A benefactive view
Full text linkOkinawan is said to have two causative forms: -(r)as and -(r)ashimi, with the former primarily expressing forcible causation and the latter, permission causation. However, this is just a tendency, not an absolute rule. This paper adopts Miyara’s (2015a) view that -(r)ashimi consists of two morphemes -(r)as and -imi but proposes that -imi functions ambiguously either as a causative morpheme or as a benefactive morpheme. This makes it possible to understand some complex interpretational facts observed in causative constructions in Okinawan. A theoretical analysis of the causative suffixes is also offered in the framework of Distributed Morphology
Constructing dependencies with optional elements: Insights from Vietnamese
Full text linkThe cognitive mechanisms that underlie the phenomenon of similarity-based interference during language processing remain a controversial issue. One well-known approach attributes the source of interference to activation patterns during the retrieval process. We report research on Vietnamese that examines the possibility of similarity-based interference effects during the processing of an optional dependency between a wh-phrase and a Question-particle (Q-particle). In this dependency, the presence of the second element is optional. Using acceptability ratings and self-paced reading methods, we provide evidence for a retrieval-based account. We also provide evidence that a sub-part of a word – specifically, the wh-morpheme inside existential expressions in Vietnamese – can result in similarity-based interference. Overall, our study contributes to understanding of similarity-based interference using a previously-under-researched type of dependency, namely one where the second element is optional and thus not wholly predictable
Learning new neopronouns or task adaptation?
Full text linkEnglish neopronouns have been reported difficult to learn or use because of their low frequency, and because pronouns are a closed category, and are rated below ceiling in acceptability surveys. Hekanaho (2022) reports that metalinguistic commentary about neopronouns includes themes of prescriptivism, unfamiliarity, ‘weirdness,’ and confusion. In this study, we examine whether exposure to neopronouns during a survey increased acceptability ratings. Neopronouns were rated higher when seen later in the survey. We analyze metalinguistic comments from the survey in which participants report their experiences in the survey as educational or instructive, suggesting that some order effects may be reflective of actual learning