Proceedings Published by the LSA (Linguistic Society of America)
Not a member yet
4323 research outputs found
Sort by
Taiwanese complementizer kóng, sentence-final particles, and the final-over-final condition
Full text linkRecent research (Paul 2014, Erlewine 2017) has found Mandarin sentence-final particles (SFPs) to be C heads, as per Rizzi (1997). This apparently violates the proposed universal Final-over-Final Condition (FOFC; Sheehan et. al. 2017), which prohibits the projection of any head-final phrase over a head-initial one. This paper brings this issue into Taiwanese, which has the added complexity of a head-initial complementizer kóng. I use co-occurrence restrictions to argue that Taiwanese SFPs occupy multiple head positions, and show that only one of these can be embedded. The result is argued to support the generalization from Richards (2016) and Erlewine (2017) that FOFC applies only within individual phases
Stronger together: Unleashing the social impact of hate speech research
Full text linkThe advent of the internet has been both a blessing and a curse for once marginalised communities. When used well, the internet can be used to connect and establish communities crossing different intersections; however, it can also be used as a tool to alienate people and communities as well as perpetuate hate, misinformation, and disinformation especially on social media platforms. We propose steering hate speech research and researchers away from pre-existing computational solutions and consider social methods to inform social solutions to address this social problem. In a similar way linguistics research can inform language planning policy, linguists should apply what we know about language and society to mitigate some of the emergent risks and dangers of anti-social behaviour in digital spaces. We argue linguists and NLP researchers can play a principle role in unleashing the social impact potential of linguistics research working alongside communities, advocates, activists, and policymakers to enable equitable digital inclusion and to close the digital divide.
Must "big" syllables carry stress in English?
Full text linkTwo types of syllables with ambiguous stress-status are identified: open syllables with a full vowel adjacent to the primary stress (e.g. no.tá.tion) and closed syllables with a full vowel that occur non-adjacent to, but to the right of the primary stress syllable (e.g. cá.ra.van}). Both syllable types have a full vowel but need not be heavy; the latter type also has a coda consonant. These two segmental properties of syllables were separated into two "big" syllable shapes, [Cæ] and [Cəs]. Two perception studies were run in which these syllables were given the prosodic characteristics of unstressed syllables and placed in syllable strings where a listener would expect a stressed syllable given the otherwise alternating pattern of the string. Listeners also heard truly alternating strings, and strings with initial or final stress lapse without a big syllable as part of the stress lapse. It was found that unstressed open syllables with full vowels were highly confusable with a truly alternating pattern, whereas unstressed closed syllables were not. As both ambiguously-stressed syllable shapes under consideration involve a full vowel, our full vowel study gives support for the hypothesis that such syllables may not be stressed, and are confusable with stressed syllables because of their vowel quality.
Comparing form-based and meaning-based gender biases in pronoun resolution: Inferences from names and role nouns
Full text linkReferring to people using only their name (e.g. Jones came in) is known to evoke the assumption that the individual is male. The same holds with some role nouns (e.g. mechanic, boxer). We explore these effects through the lens of pronoun interpretation in English. In two sentence-completion studies, we show that both form-based (last-name-only style) and meaning-based gender biases (from role nouns) are powerful enough to eliminate otherwise robust verb semantic effects on pronoun interpretation (implicit causality). In addition, the results provide initial evidence that meaning-based biases (at least the ones tested here) can be stronger than form-based biases, which may stem from differences in the form-function mapping
The morphosemantics of incremental plurality in Hualapai (Yuman)
Full text linkHualapai is a Yuman language with a verbal morphological system, seen in various languages of the region, which poses difficulties for a compositional analysis. In particular, Hualapai verb morphology exhibits incrementality (see Baerman 2016, 2019, 2024), where there is no one-to-one mapping between forms and meanings. Instead, forms are ordered on a scale tracking morphological complexity, and more complex morphological forms are mapped, all things being equal, to meanings that are higher on some semantic scale. In the case of Hualapai, the incremental system concerns a plurality, which in this case conflates plural argument marking and plural event marking, also known as pluractionality. This paper provides the first compositional morphosemantic treatment of Hualapai verbal plurality, which in addition, is used to think about how we might handle such incremental systems in general using standard morphological and semantic tools
Tone 3 Sandhi in Mandarin: Productivity and acoustic realization in L1 and L2 speakers
Full text linkThis study investigates the productivity and acoustic realization of a highly regular tonal alternation in Mandarin, Tone 3 (T3) sandhi, among native speakers and intermediate-level second language (L2) learners across different lexical contexts. Seventeen native speakers (NSs), fourteen intermediate-low (IL) learners, and sixteen intermediate-high (IH) learners completed a production task involving both real and nonce disyllabic words. Results show that T3 sandhi is highly productive among native speakers and is consistently applied by L2 learners, with IL learners achieving application rates above 77% even in pseudo-word contexts, providing strong support for a computation-based mechanism in sandhi production. Acoustic analysis furthered revealed that IH learners produced more native-like sandhi realizations than IL learners, although differences from NSs remained under pseudo-word conditions. In addition, NSs and IH learners showed similar differences in T3 and Tone 2 (T2) production across lexical conditions, highlighting a lexical effect on tonal production. These findings contribute to our understanding of L2 phonological acquisition, demonstrating that regular phonological alternations such as T3 sandhi can be productively acquired at relatively early stages of L2 acquisition
Incremental or delayed processing? L2 learners’ active gap-fillingin sentence comprehension
Full text linkThe contrast between first language (L1) and second language (L2) sentence processing provides valuable insights into how linguistic knowledge and cognitive mechanisms shape real-time comprehension. While L1 speakers are known to engage in incremental parsing–integrating linguistic input as it becomes available–it remains debated whether adult L2 learners adopt similar strategies or rely on delayed, non-structural parsing as proposed by the Shallow Structure Hypothesis (SSH). To investigate this, the present study employed the visual world eye-tracking paradigm to examine how Japanese L2 learners of English process structurally ambiguous wh-questions. Results revealed a strong preference for main clause (MC) interpretations, mirroring native speaker behavior. Crucially, eye-tracking data showed early fixation patterns consistent with active gap-filling, providing evidence against SSH. Furthermore, L2 learners demonstrated an even stronger commitment to the MC analysis than native speakers, suggesting that while their syntactic processing is guided by incremental structural parsing, it may also reflect a greater reliance on initial interpretations
Free relatives free from silent operators
Full text linkFree relatives have long posed challenges for theories of the syntax-semantics interface: although clausal in form, they have the same distribution and interpretation as definite noun phrases. In this paper, I present a typologically novel class of free relatives from Old Avestan: those introduced by non-interrogative-based pronouns, contrasting with the standard definition which assumes an interrogative-based pronoun as the relativizer (Caponigro 2003). I also propose a new analysis of free relatives at the syntax-semantics interface that, based on the unique properties of the Old Avestan data, minimizes the need for silent operators to derive the definite reading of free relatives.
Degrees of animacy and telicity in Hindi intransitives: A corpus study
Full text linkIntransitives are known to fall into two classes, unergatives and unac-cusatives, associated with a wide range of diverging properties. It has long been argued that the semantics of a given verb determines whether it will preferentially exhibit unergative or unaccusative behavior. In particular, agentive semantics has been associated with unergativity, telic semantics with unaccusativity. In this paper, we test this claim empirically using corpus data from Hindi. For a wide range of intransitives, we compute how often they take an animate argument and how often they combine with a telic light verb. The results offer some support for the effect of agentivity and telicity on the syntax of intransitives, but are too limited in scope to allow for reliable conclusions. We end by discussing the methodological challenges of establishing the semantic correlates of split intransitivity
On Wh-Movement under Sluicing
Full text linkIn this study, I provide novel arguments for Fox and Lasnik’s (F&L) (2003) analysis of sluicing. F&L assume that in sluicing constructions, wh-movement can proceed either locally or non-locally, and whether an occurrence of wh-movement is ruled out or not depends on whether it creates a dependency that is required to satisfy Parallelism. If these authors are on the right track, we would predict that wh-movement in sluicing constructions could proceed differently from wh-movement in non-sluicing contexts. I show that this prediction is correct. The first argument is drawn from English double object constructions, and the second is drawn from Dutch R-pronouns. The supporting facts not only align with the prediction but also follow naturally from Parallelism. Although both cases have been discussed in the literature, I reexamine them from a new perspective