Part 1: 10 am - 12 pm, (Wed) June 10, 2020 [Japan Standard Time]*
Canaan Breiss (UCLA) "Phonological markedness effects in sentence formation"
Earlier research has found that phonological markedness constraints (for example, against stress clash or sibilant sequences) statistically influence speakers’ choices between particular syntactic constructions and between synonymous words. I will present results from collaborative work with Bruce Hayes and more recently Tim Hunter (both UCLA) demonstrating that phonological constraints exert an influence on sentence formation not just in particular cases, but across the board. We employ a novel method that statistically models the distribution of word bigrams (consecutive two-word sequences) and how this distribution is influenced by phonological constraints. Our study of multiple corpora indicates that several phonological constraints do indeed play a statistically significant role in English sentence formation. We also show that by examining particular subsets of the corpora we can diagnose the mechanisms whereby phonologically marked sequences come to be underrepresented. We conclude by discussing modes of grammatical organization compatible with our findings, and prospects for more fully integrated and implemented models of syntax-phonology interaction in the grammar.
Ryan Bennett (UCSC) "Vowel deletion as phonologically-controlled gestural overlap in Uspanteko"
Uspanteko is a K'ichean-branch Mayan language spoken by 1500-4000 people in the central highlands of Guatemala. Unstressed vowels in Uspanteko often delete, though deletion is variable within and across speakers. Deletion appears to be phonological, being sensitive to phonotactics, foot structure and morphology, and being insensitive to speech style. But deletion also appears to be phonetic, being variable, gradient, insensitive to certain phonotactics, and opaque with respect to accent placement. Electroglottography data from one speaker suggests that even 'deleted' vowels may contribute voicing to [C(V)C] intervals. We thus analyze deletion as phonologically-controlled gestural overlap (e.g. Gafos 2002).
Part 2: 10 am - 12 pm, (Wed) June 24, 2020 [Japan Standard Time]*
Taylor Lampton Miller (SUNY Oswego) "In Search of True Word-Level Recursion"
Recent work at the phonology-syntax interface has revived the discussion on whether recursion is present in phonological structure, and – if so – where is it permitted (e.g. Selkirk 2011, Vogel 2019). Adopting a definition of recursion which requires the nested constituents to be of the same type and demonstrate the same properties, Vogel 2019 and Miller 2020 argue that true recursion is only permitted at the Phonological Phrase and above. As recursive word analyses become more commonplace (Guekguezian 2017, Bennet 2019), I have begun collaborative work with Hannah Sande (Georgetown), in which we search in earnest for true recursion at the word-level. In this presentation, I outline our assumptions and conclusions that even very convincing cases of word-level recursion in Kaqchikel and Greek are best re-analyzed. I also discuss the ramifications of such findings for future theoretical analyses and our understanding of the interface itself.
Harim Kwon (George Mason U) "Beyond phonotactics: Perception of Nonnative consonant clusters"
Abstract: In this talk, I will present a study that examines the role of native phonetic patterns in the perception of word-initial consonant clusters. Specifically, we ask how native listeners of Georgian and French perceive onset CC clusters produced with atypical inter-consonant timing patterns. Georgian permits more varied onset clusters than French, and Georgian onset clusters are produced with a longer inter-consonant lag than those in French. The listeners heard the onset clusters in CCV-CVCV pairs that were produced by a Georgian native speaker and a French native speaker with their own respective native phonetic patterns. Their task was to determine whether the two sequences (e.g., CCV and CVCV) were the same or different. The results show that (1) French listeners confused Georgian CCV with Georgian CVCV even when the clusters were phonotactically licit in French, and (2) some Georgian listeners confused French CCV with French CVCV although all French clusters were licit in Georgian. These findings suggest the perception of nonnative consonant clusters is guided not only by native phonotactics but also by preferred inter-consonant timing patterns of the listeners' native language.
Plenary: 9 am - 12 pm, (Sat) July 4, 2020 [Japan Standard Time]*
Prof. Barbara Partee (UMass Amherst) "The Intertwining Influences of Linguistics, Logic, and Philosophy in the History of Formal Semantics)"
Formal semantics as it has developed over the last 50 years has been shaped by fruitful interdisciplinary collaboration among linguists, philosophers, and logicians, involving developments in linguistic theory, philosophy, cognitive science, and computational linguistics. Before the birth of formal semantics in the late 1960’s, linguists and logicians were mostly agreed, for different reasons, that logical tools could not be seriously applied to natural languages. Philosophers and logicians considered natural language too unruly, while linguists like Chomsky considered the work of logicians to be irrelevant. The story of formal semantics starting from the 1970’s and beyond is the story of how linguists, philosophers, and logicians learned to appreciate, use, and synthesize the advances in each other’s fields. In this talk, I’ll trace the background and history of these developments, describing(without presupposing any specific background) some of the pivotal advances and controversies that have shaped the field. Two founding revolutions laid the groundwork-the Chomskyan revolution that established formal syntax, and the Montogovian revolution that extendedlogicians’ formal semantics to natural languages. I’ll talk about the context in which Montague’s work occurred and other contributors to it, and how joint work by linguists, logicians, and philosophers after Montague’s death led to formal semantics becoming part of linguistics. A central theme is the challenge of developing a compositional semantic theory for natural languages; that challenge has taken on different forms with different syntactic and semantic theories. And quantifiers have played a major role at some crucial turning points in the history of semantics, from the linguistic wars of the 1960s that brought an end to the “Garden of Eden” period of Chomsky’s Aspects of the Theory of Syntax, to the advances of “generalized quantifiers”, to the Kamp-Heim innovations of dynamic semantics. I’ll close with a very brief discussion of some very big foundational issues concerning the nature of semantic competence, issues that are crucial to the place of formal semantics in cognitive science.
[Registration of the plenary talk: https://forms.gle/cSXf8Pp94WTLveZx8]
Part 3: 10 am - 12 pm, (Wed) July 15, 2020 [Japan Standard Time]*
Misaki Kato (U Oregon) "Acoustic characteristics of foreign accent in L2 Japanese: A cross-sectional study"
Previous studies report that adult second language (L2) learners retain a discernible foreign accent. The presence of a foreign accent can lead listeners to think that the speaker is not understandable and/or less credible, even when the message is accurately conveyed. However, it is much less well-understood what the acoustic sources of a foreign accent are. In the current study, we report an exploratory analysis, examining which acoustic features impact perception of a foreign accent the most, and whether they vary for the speech produced by learners of different proficiency levels. Japanese speech samples were collected from native Japanese speakers and native English learners of Japanese across different instructional levels and learning backgrounds. We measured 27 acoustic variables in the samples, including segmental features, rhythm features, fluency features, and tonal features. Native Japanese listeners rated the speech samples for foreign accentedness. We used random forests, a data mining procedure, to explore which acoustic variables are most important in predicting accent rating. While pitch accent, articulation rate, and vowel duration influenced perceived accentedness of the speech samples in general, the relative importance of these acoustic features varied across speaker groups. Specifically, prosodic features were most predictive of beginning to intermediate late learners’ accents, whereas vowel features were most predictive of early bilinguals. The current results shed light on issues related to the development of second language speech, and the perceptual relevance of the development as perceived by listeners.
Claire Moore-Cantwell (UCLA) "Learning a "crazy rule": Final vowels and stress in English"
English speakers exhibit productive phonological knowledge of a complex, phonetically unmotivated trend in the lexicon of English. Most long words of English vary in stress, taking either penultimate ('vanílla') or antepenultimate ('cínema'), but words ending in the high front vowel [i] strongly tend to take antepenultimate stress. Words like 'cánopy' are much more common than words like 'spaghétti' (Moore-Cantwell, 2016; Moore-Cantwell and Sanders, 2018, Hayes 1984). In this talk I will discuss how this 'crazy' pattern should be analyzed, and implications of that analysis for the acquisition process. I will discuss two main possibilities: (1) a constraint cloning analysis, in which learners can clone existing constraints (in this case NonFinality) during the acquisition process to account for patterns their universal constraint set may not be able to capture, and (2) a parochial constraint analysis, in which learners can induce simple constraints during the acquisition process. These two analyses make different predictions for a type of word not found in the lexicon of English: words with heavy penultimate syllables that also end in [i]. I present wug-test results including these type of words, demonstrating that they align with the parochial constraint analysis rather than the cloning analysis.
Part 4: 10 am - 12 pm, (Wed) July 29, 2020 [Japan Standard Time]*
Jiseung Kim (U Michigan) "Individual differences in the production and perception of prosodic boundaries in American English"
We investigate the hypothesis that individual participants vary in their production and perception of prosodic boundaries, and that the acoustic properties they use to encode prosodic contrasts are closely related to the properties used to perceive those contrasts. An acoustic study examined 32 native speakers’ production of sentences containing Intonational Phrase (IP) and word boundaries. Twenty participants returned and participated in an eye-tracking study where they listened to stimuli that were manipulated to include different combinations of the acoustic properties associated with IP boundaries. The results indicate large variability in both production and perception, and provide limited evidence for production of the boundary cues influencing the same individuals’ perception.
Sang-Im Lee-Kim (National Chiao Tung U) "Interaction between linguistic and extra-grammatical factors in merger-in-progress: the case of sibilant merger in Taiwan Mandarin"
Sound change led by men is scarce and often limited to relatively isolated patterns (Labov 1990, 1994). It is not yet well understood how linguistic and extra-grammatical factors interact to drive such changes-in-progress. In this talk, I present empirical evidence from production, perception, and imitation paradigms to shed light on the nature of the alveolar-retroflex sibilant merger in Taiwan Mandarin (TM). This sibilant merger in TM is implemented through deretroflexion of the retroflex category and is socially stigmatized to some degree. I will first show that young men employ the merger more often than age-matched women in a reading task, but both groups carry clear contrasts when disambiguation of lexical contrasts are necessary in a conversational setting. A perception study will show how speakers with the merger differ from non-mergers in terms of the extent to which social cues are incorporated into perceptual judgments. Specifically, speakers of the merger associated particular TM dialects with deretroflexion more strongly than non-mergers. A spontaneous imitation study provides further evidence for an association between deretroflexion and Taiwanese identity; mergers were more resistant to imitating a model talker carrying extreme sibilant contrasts. The bigger picture arising from these experiments is that deretroflexion has newly emerged among young TM speakers as a means of expressing positive orientation toward their speech community.