Keio x ICU LINC

Season 3 of the Keio x ICU LINC series has been completed. The videos of the talks can be viewed at:

https://www.youtube.com/channel/UCENIXAFDRt4Jy9UlTEw8AWQ/videos

Season 4

Part 1: 10 am – 12 pm, Monday, April 26, 2021 [JST: Japan Standard Time]

Caitlin Smith (Johns Hopkins University)

"Learning Derivationally Opaque Patterns in the Gestural Harmony Model"

Amber Lacey Lubera (University of Arizona) and Ryan Walter Smith (University of Arizona)

"Morphophonological evidence for onset-sensitive stress in Iron Ossetian"

Part 2: 10 am - 12 pm, Monday, May 10, 2021 [JST]

Jenny Bellik (University of California, Santa Cruz)

"Size effects in prosody: Counting branches, counting leaves"

Hannah Sande (University of California, Berkeley)

”Cophonologies by Phase: Phases as the domain of phonological evaluation"

Part 3: 10 am - 12 pm, Monday, May 24, 2021 [JST]

Sara Finley (Pacific Lutheran University)

"Testing predictions from generative phonology using artificial language learning experiments"

Hayeun Jang (Busan University of Foreign Studies)

"Simulation of Tongue Muscle for Phonological Research: a case study of coronal palatalization”

Part 4: 10 am - 12 pm, Monday, June 7, 2021 [JST]

Jonah Katz (West Virginia University)

"Campidanese Sardinian lenition is phonetic and conditioned by duration"

Rebecca Starr (National University of Singapore)

"Language attitudes and exposure as predictors of phonological variation among local and expatriate children in Singapore"

Part 1: 10 am – 12 pm, Monday, April 26, 2021 [JST: Japan Standard Time]

Caitlin Smith (Johns Hopkins University)

"Learning Derivationally Opaque Patterns in the Gestural Harmony Model"

In this talk, we examine the learnability of two apparently derivationally opaque vowel harmony patterns: attested chain-shifting height harmony and unattested saltatory height harmony. We analyze these patterns within the Gestural Harmony Model (Smith 2018) and introduce a learning algorithm for setting the gestural parameters that generate these harmony patterns. Results of the learning model indicate a learning bias in favor of the attested chain-shifting pattern and against the unattested saltation pattern, providing a potential explanation for the differences in cross-linguistic attestation between these two derivationally opaque patterns. Furthermore, we show that feature-based learning models of these patterns indicate a learning bias in favor of unattested saltatory height harmony, contra the typological facts.

Amber Lacey Lubera (University of Arizona) and Ryan Walter Smith (University of Arizona)

"Morphophonological evidence for onset-sensitive stress in Iron Ossetian"

Onset-sensitive stress systems have received much attention and scrutiny in recent phonological theory (Topintzi 2006; Hyde 2007; Topintz & Nevins 2017). Topintzi (2006) in particular argues that onset-sensitive systems come in two general varieties: i) those in which some feature of the onset contributes weight (e.g. voiced vs. voiceless onsets in Pirahã, Everett & Everett 1984), and ii) those in which onset presence contributes weight (e.g. Banawá, Hyde 2007). While additive systems are allowed, systems in which only syllables with complex onsets count as heavy, while those with simplex or no onsets are light are intentionally excluded. In this talk, we present evidence that the system excluded from Topintzi’s typology is instantiated in Iron Ossetian (Eastern Iranian; Russia, Georgia; henceforth Iron). In Iron, syllable weight is determined by the quality of the vowel in the nucleus and by onset complexity. Syllables are considered heavy (and therefore stressed) if the nucleus contains one of the “strong” vowels (/a/, /e/, /i/, /o/, /u/), or a “weak” vowel (/ə/, /ɨ/) with a complex onset. Crucially, syllables with weak vowel nuclei and simplex or empty onsets do not attract stress. Additionally, evidence from productive morphological processes and phonotactic constraints rules out an analysis of this system in which initial clusters are heterosyllabic. Iron, therefore, exemplifies a typologically rare stress system that does not fit into current typological paradigms and presents a unique challenge for modeling onset-sensitive stress.

Part 2: 10 am - 12 pm, Monday, May 10, 2021 [JST]

Jenny Bellik (University of California, Santa Cruz)

"Size effects in prosody: Counting branches, counting leaves"

Binarity—the pressure for phonological constituents to contain exactly two elements—is widely recognized as a factor that shapes prosody. This talk investigates two contrasting conceptions of phrasal binarity, and their divergent consequences. Branch-counting binarity is a structural constraint that counts a phonological phrase's branches. Branch-counting binarity prioritizes matching larger, branching XPs, and works in tandem with Match(XP) to produce recursive prosodic structure. In contrast, leaf-counting binarity is a length constraint that counts the words that a phonological phrase contains. Word-counting binarity prefers not to match larger XPs, and conflicts with Match(XP). It predicts a more complex typology than branch-counting binarity. Both branch-counting and leaf-counting binarity constraints are needed to account for the range of observed size effects cross-linguistically, as seen in examples from phrasing in Irish (Elfner 2012) and Italian (Van Handel to appear).

Hannah Sande (University of California, Berkeley)

”Cophonologies by Phase: Phases as the domain of phonological evaluation"

This talk explores the domain of phonological evaluation by looking at the amount of structure affected by morpheme-specific phonology. In some cases, morpheme-specific phonology affects a word-sized domain, but more common are morpheme-specific effects that affect a sub-word or cross-word (phrasal) domain. What, then, determines the amount of structure affected by a morpheme-specific phonological processes? Two case studies are considered here: 1) harmony in Guébie (Kru) which affects a domain smaller than a word, and 2) tone placement and spreading in Kuria (Bantu) across multiple words. Both processes are shown to correspond to syntactically defined phase domains. An analysis is provided in Cophonologies by Phase (Sande 2019, Sande et al. 2020), which assumes that morphemes can be associated with morpheme-specific sub-grammars that apply to the domain in which they are spelled out. Spell-out is assumed to apply at syntactic phase boundaries. This model makes strong predictions about how much structure is expected to be subject to morpheme-specific processes, and accounts for a wide range of morphophonological phenomena that are difficult to account for in other models.

Part 3: 10 am - 12 pm, Monday, May 24, 2021 [JST]

Sara Finley (Pacific Lutheran University)

"Testing predictions from generative phonology using artificial language learning experiments"

Generative models of phonological patterns typically produce highly abstract representations that can be difficult to verify through external, or experimental evidence. One promising approach to test the theoretical predictions found in generative phonology is the use of artificial grammar learning experiments, where participants are trained on a novel phonological pattern and then tested for learning and generalization of that pattern. In this talk, I will argue for the use of artificial language learning as a way to understand predictions from generative phonology. Specifically, I will provide examples from learnability of exceptionality in vowel harmony, and discuss suggestions for developing hypotheses from generative models.

Hayeun Jang (Busan University of Foreign Studies)

"Simulation of Tongue Muscle for Phonological Research: a case study of coronal palatalization”

I believe that understanding the principles of tongue movement is necessary to understand phonological typological patterns. However, it is not easy to study the function and interaction of tongue musculature, which is the basis of the principle of tongue movement. In this talk, I will propose that simulations using a biomechanical 3D tongue model of Artisynth (Lloyd et al. 2012; artisynth.org) can be used in phonological research. In the simulations, we can manipulate each tongue muscle's activation level and timing. I will show that a simulation using the Static Jaw Hyoid model of Artisynth helps us understand the implicit relationship of vowels triggering coronal palatalization that is problematic in the traditional approaches.

Part 4: 10 am - 12 pm, Monday, June 7, 2021 [JST]

Jonah Katz (West Virginia University)

"Campidanese Sardinian lenition is phonetic and conditioned by duration"

Campidanese Sardinian displays a complex system of obstruent lenition that has received much attention in theoretical phonology. In the Campidanese system, a voiced and voiceless UR series of stops (referred to as /D/ and /T/, respectively) contrast in post-pausal or utterance-initial position. The /T/ series lenite to voiced continuants following a vowel within a phrase; the /D/ series lenite less often and less radically. This pattern is phonologically problematic because /T/ undergoes a relatively radical change ([voi] and [cont]), while /D/ sometimes fails to undergo a less radical change (just [cont]). This talk argues that none of the phonological devices mentioned above are necessary or sufficient for describing Campidanese consonant lenition. Instead, I propose a model that derives manner-related lenition and fortition from prosodically-conditioned changes in duration, without changing phonological features at all. This phonetic approach captures several core facts about the consonant system: (1) rates of prosodically-conditioned lenition and associated changes in intensity are predictable from prosodically-conditioned differences in duration; (2) manner and intensity differences between different UR consonant series are not predictable from duration alone; and (3) duration- and intensity-based lenition and fortition affect all consonants, even extending to vowel-vowel transitions in hiatus. I will present the phonetic model, show how it derives the core properties described above, and speculate on the cross-linguistic uniformity or lack thereof in such intervocalic lenition processes.

Rebecca Starr (National University of Singapore)

"Language attitudes and exposure as predictors of phonological variation among local and expatriate children in Singapore"

Although orientation towards local norms is increasing in Singapore, Singapore English (SgE) is still perceived by some as a nonnative variety. Variation in attitudes towards SgE may shape acquisition of SgE features by both Singaporean and expatriate children, who increasingly attend government schools. The present study investigates how the -t/d deletion patterns of 60 children reflect their attitudes and school setting. Significant correlations are observed between deletion rate, attitude towards SgE, and accent self-perception among Singaporean children, highlighting that this variety is undergoing endonormative stabilization. However, while some expatriates in local schools delete more than peers in international schools, expatriate children generally do not acquire local -t/d deletion rates or constraints, regardless of familiarity with SgE or attitudes towards the variety. This gap between locals and expatriates reflects the persistence of ideologies that delegitimize SgE, as well as the growing prominence of SgE as a marker of local identity.