Proceedings Published by the LSA (Linguistic Society of America)
Not a member yet
4323 research outputs found
Sort by
On meanings of the reciprocal suffix -š- in Balkar
Get PDFThe paper is based on evidence collected from Balkar speakers during the authors’ field trips to Verkhnyaya Balkariya village, Kabardino-Balkarian Republic in August of 2021. The study addresses semantics of the suffix -š- in Balkar. This marker has three basic meanings: reciprocal, sociative, and competitive. Reciprocal denotes simultaneous mutual action by two or more participants, each (or most) performing more than one semantic role. Sociative describes collective or distributive plural actions. Competitive interpretations refer to situations when two or more people are trying to surpass each other in some activity. All of the basic meanings can be confined to verbal plurality. The suffix also manifests some scalar properties. For each of the basic meanings we reveal certain restrictions on the semantic class of the base verb. Apart from that, a range of additional meanings sometimes manifest in -š-derivates: durative, iterative, and excessive plurality
The role of relevance, competence, and priors for scalar inferences
Get PDFAlthough it is often assumed that the natural language expressions \u27some\u27 and \u27or\u27 are interpreted according to their first-order logic counterparts, in certain contexts, they receive a narrower interpretation: \u27some\u27 is strengthened to \u27some, but not all\u27, and \u27or\u27 to \u27A or B, but not both\u27. This process is typically explained as an instance of scalar inference. To test this scalar implicature hypothesis, we collect experimental evidence for the effects and interactions of three factors that should affect the robustness of the scalar inferences of \u27some\u27 and \u27or\u27: the relevance of the stronger alternative, the speaker\u27s competence about the alternative, and the prior probability that the alternative is true. We find that the interpretation of both triggers was affected by speaker competence, but only \u27some\u27 was also affected by prior probability, while relevance did not affect either trigger. Ultimately, our results suggest that the interdependence of the three factors is more complex than just the sum of their effects
Beyond Surprising: English Event Structure in the Maze
Get PDFTo what extent can we tease apart semantic representations and processes from other influences on processing such as probabilistic prediction? In this paper I detail two experiments testing the hypothesis that there are semantic complexity in the lexical representations of result verbs that influences reaction times above and beyond probabilistic distributions. This is done by replicating a self-paced reading study from Levinson & Brennan (2016) while also modelling lexical surprisal. Experiment 1 replicates the original result, but only in experiment 2 using the maze task does the effect emerge beyond surprisals. The more focal maze task results suggest that processing costs associated with bieventive result verbs should be accounted for by grammatical factors, in addition to probabilistic prediction
Representing affect information in word embeddings
Get PDFA growing body of research in natural language processing (NLP) and natural language understanding (NLU) is investigating human-like knowledge learned or encoded in the word embeddings from large language models. This is a step towards understanding what knowledge language models capture that resembles human understanding of language and communication. Here, we investigated whether and how the affect meaning of a word (i.e., valence, arousal, dominance) is encoded in word embeddings pre-trained in large neural networks. We used the human-labeled dataset (Mohammad 2018) as the ground truth and performed various correlational and classification tests on four types of word embeddings. The embeddings varied in being static or contextualized, and how much affect specific information was prioritized during the pre-training and fine-tuning phase. Our analyses show that word embedding from the vanilla BERT model (Devlin et al. 2019) did not saliently encode the affect information of English words. Only when the BERT model was fine-tuned on emotion related tasks or contained extra contextualized information from emotion-rich contexts could the corresponding embedding encode more relevant affect information
Proportions vs. cardinalities: Comparative ambiguities and the COVID pandemic
Get PDFThis paper reports two psycholinguistic experiments on quantity comparatives and superlatives that are potentially ambiguous between cardinal and proportional readings. By using statements about COVID cases and vaccination numbers as a naturalistic context with real-world relevance, this work furthers our understanding of what happens in linguistic environments where multiple measure functions are available – what modulates the choice between them? The results provide new evidence that comparatives and superlatives can refer to scales ranging over degrees of proportion, in addition to degrees of cardinality. Furthermore, this experimental evidence points to a preference for cardinal interpretations, but also show that this is not rigid and can be weakened in favor of proportional readings by semantic factors – including considerations potentially related to stage- vs. individual-level differences – and by certain linguistic forms
On the interpretation of German einige. The effect of tense and cardinality
Get PDFWe present a study investigating the effect of tense (past vs. future) on the computation of scalar implicatures in connection with the German quantifier einige ‘some’ in an interactive experiment, which included a financial incentive for participants to consider whether another speaker would share their judgment. We tested the hypothesis that scalar implicatures are less frequently drawn in future tense than in past tense. In addition, we studied to what extent sets with various cardinalities are prototypical representatives of einige + N. We hypothesized that larger cardinalities are more prototypical representatives of the quantifier einige than smaller cardinalities (relative to the cardinality of the total set). We analyzed the experimental data with probabilistic Bayesian models with a linking hypothesis between participants’ responses and readings based on utility maximization in simple decision problems. In line with the hypotheses, we found that less scalar implicatures are drawn in future tense than in past tense, which replicates the results of previous research on English some, and that with an increase in set size acceptance of statements involving einige also increases
Use It or Lose It Harmony in Komo
Get PDFThis paper discusses a case of putative dominance reversal in the Komo language (Otero 2015, 2019), which we analyze as a related, but distinct repair strategy called “Use it or Lose it” (Mullin & Pater 2015).Mullin & Pater (2015) argue that Use it or Lose it harmony is a pathological prediction of Agree for the same basic reason that “Sour Grapes” harmony (Wilson 2003, 2006; Heinz & Lai 2013) has been regarded as pathological – both Use it or Lose it and Sour Grapes harmony patterns are non-myopic (Wilson 2003, 2006). Wilson argues that unbounded spreading patterns are universally myopic, and as such, no theory should predict that the realization of some element in spreading – trigger or target – depends on downstream information. However, recent research has shown that some patterns in natural languages are, in fact, non-myopic, indicating that the predictions of Agree are not as problematic as previously thought. This paper argues that the best analysis of Komo relies on the activity of [Atr] and both regressive [+Atr] spreading and [+Atr] trigger effacement are repairs to a single marked structure in the language, *VC0[Hi, Atr]
The Phonology of the Definite Determiners in Yemeni Tihami Arabic
No full textThis paper analyzes the phonology of the definite determiner (DET) of two Yemeni Tihami Arabic dialects. The underlying Det for one dialect (the b-dialect) is /b-/: [θoor] ‘ox’- [b-θoor] ‘the ox’, and /m/ for another dialect (the OCP m-dialect): [m-θoor] ‘the ox’. Although the determiner is underlyingly different in both dialects, it fully assimilates to the following word-initial consonant if it is labial, creating a word-initial geminate. On the surface, the onset geminate behaves differently in both dialects. I present an analysis that treats them the same except for their underlying representation and the ranking of two constraints. Both dialects present further complications involving these onset geminates. In the OCP m-dialect, labial assimilation is always optional, and if the word-initial syllable is light, as in /firag/ ‘teams’, gemination is optional ([f-firag] ~ [m-firag]), but if it is heavy, gemination is resolved by epenthesis: /m-fiiraan/ → [ʔaf.fii.raan] ‘the mice’. This is explained if we assume all geminates, even word-initial ones, are moraic: in that case, epenthesis is compelled by a ban on extra-heavy syllables. In the b-dialect, labial assimilation is obligatory because of a high-ranking Ocp-Lab. However, gemination in heavy syllables is only optional: [f-faanuus] ~ [ʔif.faa.nuus], ‘the lantern’
Prosodically-conditioned relative clause extraposition in Armenian
Get PDFArmenian is an SOV language that has post-nominal finite relative clauses (RCs). These clauses are typically immediately post-nominal: N+RC. But in various contexts, the relative clause is extraposed to the right edge of the sentence: N+V+RC instead of *N+RC+V. The various contexts are united by how the modified noun is prosodically phrased with an immediately following verb. We argue that extraposition is conditioned by prosodic phrasing. A host of syntactic factors (definiteness, subject/object, valency) are indirectly involved in extraposition, but these factors are tied directly to prosodic phrasing. Exceptions to this generalization are limited and come from verb focus and possible recursive phrasing
Asymmetric Infixation
Get PDFIn this paper, I present novel evidence supporting the claim that there are no right-edge infixes. Based on a typological survey, I demonstrate that all putative right-edge infixes only surface in languages with right-edge prosodic prominences. It is therefore possible to reanalyze all right-edge infixes as prominence-oriented infixes. Infixes, as a result, are highly asymmetric: they can occur in the left edge of a stem or in a prosodically prominent position, but nowhere else. To account for this asymmetric distribution, I propose that infix subcategorization is implemented by Anchor, rather than Alignment. Anchor has been previously argued to also be asymmetric (Nelson 2003), where it can target left edges or prominent positions, but crucially never right edges alone. By contrast, Alignment does not predict this asymmetry. I therefore conclude that affixation is divided into two distinct typologies: reduplication and infixation are governed by Anchor, but generic affixation is governed by Alignment