Proceedings Published by the LSA (Linguistic Society of America)
Not a member yet
4323 research outputs found
Sort by
Event plurality & quantifier scope across clause boundaries
Get PDFLegend has it that quantifiers cannot scope out of finite clauses. But whileislands for quantifier raising might exist, finite clauses are not that: We identifya novel environment which productively facilitates scoping universal quantifiersout of embedded clauses, involving the manipulation of event structure. With thehelp of the perfect on an embedding verb and certain adverbials that presuppose abuildup towards a result state (by noon, eventually, at long last), embedded universalquantifiers can more readily take extrawide scope. We describe, account for, anddiscuss restrictions to this effect, and conclude that scoping quantifiers out of finiteclauses is not banned by syntactic constraints, although context or processing mightfavor narrow scope readings
Dorsal harmony in Squliq Atayal: An Acoustic and Ultrasound Study
Get PDFDorsal harmony, understood as a process of changing from /*k/ into /q/ when followed by a uvular /q/ or a pharyngeal /ħ/, has been reported as a historical sound change in Atayal dialects, such as Squliq and Skikun. However, it is unattested if the observed dorsal harmony also occurs in those Atayal dialects as a synchronic morpho-phonological process. In the present study, we examined dorsal harmony in Squliq Atayal through ultrasonography and acoustic measurement (onset F3 – F2). Generalized additive mixed models (GAMMs), principal component analysis (PCA) and linear discriminant analysis (LDA) were applied to fit the tongue contours and compare the similarities of tongue images across different conditions. One participant showed that all variants of [k] were deviated from Baseline [k], and the linear discriminant (LD) values of all variants of [k] fell between Baseline [k] and [q]. The other participant exhibited consistent tongue postures in Short distance [k] and Cross-boundary [q]-[k]. The observed results suggest that dorsal harmony occurs in Squliq Atayal as a morpho-phonological process. Moreover, dorsal harmony is physiologically driven and best accounted for by coarticulatory aggressiveness, though with some individual variations. These variations nevertheless may support a coarticulatory path to sound change
Coordination, coherence and A’ingae clause linkage
Get PDFThis paper examines a particular type of clause linkage (‘bridging’) in A’ingae, an en-dangered isolate spoken in Amazonian Ecuador and Colombia.We propose a formalcharacterization of its meaning (to our knowledge the first formal account for any language)that relies crucially on two SDRT coherence relations: NARRATION and BACKGROUND.We motivate this characterization with textual data and elicited data from context-relativefelicity judgments, and propose to derive it from independently observable facts aboutprosody, coordination, and anaphora in the languag
Universal force from exhaustification: Farsi hame -i DPs
Get PDFPolarity items have been analyzed as existential DPs that introduce into the semantic derivation two types of alternatives: domain alternatives (corresponding to possible restrictions of the domain of quantification) and scalar alternatives (corresponding to stronger quantificational forces.) This approach has led to the development of a typology of polarity items that is based on the types of alternatives that these items introduce (Chierchia 2013). What are the possible dimensions of variation? Bar-Lev & Margulis (2014) argue that the Hebrew determiner kol introduces domain, but not scalar alternatives. This paper shows that a class of Farsi DPs, which we call ‘hame -i DPs’, do too
Measurements from "per" without complex dimensions
Get PDFTo what extent is the compositional structure of quantity terms in natural language aligned with the structure of the quantity calculus commonly used in scientific practice, a calculus that critically relies on mathematical operations like division and the computation of quotients? In pioneering work, Coppock (2021) addresses this general question through a case study on the English preposition "per", as in "0.9 grams per milliliter". Coppock proposes that "per" expresses the operation of quantity division, an operation that forms quantities like 0.9g/mL by using ratios of measurements from different dimensions. Here we show that this “division theory” of "per" makes the wrong prediction with respect to statements about measures of density and concentration. We argue that these types of expressions call for an “anaphoric theory” ofper. On this analysis, anaphora allows for the composition to invoke multiple measurements in basic dimensions, creating the appearance of reference to complex quantities like 0.9g/mL, even though no such quantities are actually composed nor denoted in the formal semantics
On the derivation of three-verb clusters in Old English
Get PDFWord order in Old English has some properties in common with those in other Germanic languages, such as V2 in a main clause and Vf-final order in a subordinate clause. However, it had idiosyncratic traits that led to word order changes in the later stages of English. Focusing on the word order of three-verb clusters (modal+have/be+participle) in Old English, this paper argues that the rise of functional projections with the head in its initial position shaped changes in word order in the history of English
Grandfather effects in Laoling disyllabic tone sandhi
Get PDFIn Optimality Theory, grandfather effects refer to the phenomena in which a structure that already exist in the input is allowed to surface while the same structure that is derived (i.e. not in the input) is forbidden. Most studies on this issue focus on segmental processes, and only few on tonal processes. This study presents the data of Laoling regular disyllabic tone sandhi, which demonstrates a grandfathering effect where some tones are allowed in non-derived contexts but banned in derived contexts. We propose a constraint-based analysis adopting Comparative Markedness (McCarthy 2003) constraints
“Non-linguistic” Browning: Meter and music in “Pietro of Abano”
Get PDFSapir (1921) suggests that Robert Browning’s poetry is “non-linguistic” in the way he uses the English language to convey thought. We explore the meter of a much criticized example of his poetry, his “Pietro of Abano” (1880), and conclude otherwise. The poem is in trochaic metre with extensive and systematic catalexis, i.e. unrealized metrical constituents possible only at phrase boundaries. We show that Browning locates phrase and even word boundaries so that they work sometimes for, but also sometimes conspicuously against these requirements, thus drawing attention to them. We also look at one of the poem’s most curious features, a musical score where the poem’s final lines would be expected to be, and note that rather than resolve these issues, it likewise accentuates them. We conclude that Browning’s poem draws attention to linguistic form, thus allying his practice more with Jakobson’s (1960) approach to poetry than Sapir’s
Modeling language change in English first names
Get PDFSome lexical classes are more susceptible to the effects of sound symbolism, a hypothesized relationship where speech sounds represent non-phonetic properties. Sound symbolic principles are manifested in male and female personal names in English. While previous research found distinct differences between female and male names in English, the current study fills a gap in existing research by adopting a diachronic perspective: male and female names are compared to each other as well as to themselves across time. Statistical analysis was conducted via a generalized additive model in R, using a data of 5600 names sourced from the United States Social Security Administration. The study finds that female names are longer, more likely to end in a vowel, and less likely to have initial primary stress, and that names for both sexes exhibit change over time, shifting towards a pattern previously associated with female names.
Syntax of reduplication and negative-polarity items in Buli
Get PDFNegative-polarity items in Buli, a Mabia (Gur) language spoken in Ghana, exhibit a mixed behavior between NCIs and NPIs (or between strong NPIs and weak NPIs). Thus, Buli presents a counterexample to Vallduvı\u27s generalization that negative-polarity items come in two kinds. Adopting and extending the framework of Collins & Postal (2014) and Collins et al. (2017), we will show that the apparently mysterious mixed behavior of negative-polarity items in Buli can be explained by articulating an unary-NEG structure and syntax of reduplication