Seun-Eun Kim presented a paper titled "The prosody of contrastive topic in Korean" at the 27th Japanese/Korean Linguistics Conference, held Oct 18-20 2019 in Seoul, Korea

Paper Abstract:

This study investigates the prosody of Contrastive Topic (CT) in Korean, with reference to findings on CT prosody in Japanese. CT is defined as a phrase denoting what the question being addressed is about but at the same time implies other questions about different topics (Constant 2014). According to Lee (2006), CT is marked morphologically and prosodically in Korean: CT is followed by the particle -NUN, which also marks non-contrastive Topic in Korean, and is realized with a high accent ((L)H* or L+H*L-H%). CT has been discussed in relation to non-contrastive Topic (T) and Focus (F); Lee (2006) argues that Korean CT has a prosodic pattern distinct from T and F.

The present study conducted an experiment to examine these observations about CT in Korean. It specifically investigated the prosody of CT not only on the target word, but in the global pitch contour. The results showed that on the target word, CT is prosodically distinct from both F and T, but showed a bigger difference with F than T. The global pitch contour of CT and F were similar in that both showed peak compression on the following prosodic word (pwd), consistent with Tomioka’s (2010) finding that both CT and F exhibit post-peak compression in Japanese.

Draga Zec and Dr. Martin Kramer (The Arctic University of Norway)  presented a paper titled "The gradient categorical vocalic behavior of syllabic consonants" at the  7th Annual Meeting on Phonology (AMP 2019) at Stony Brook University, Stony Brook, NY, held October 11-13, 2019

Rachel Vogel presented a paper titled "Prosodic prominence effects on laryngeal and supralaryngeal properties of Nepali stops" at the Workshop on Segmental Processes in Interaction with Prosodic STructure (SPIPS), held Sep 19-20, 2019. 

Paper Abstract:

Research over the last several decades has shown that prosodic structure affects the articulation of segments, with initial edges of prosodic domains exhibiting relatively strong articulations with respect to both laryngeal and supralaryngeal properties (e.g., Fougeron and Keating 1997; Keating et al. 2003).

Domains that exhibit these effects include the syllable, the word, and the phrase. Strengthening is said to be cumulative, with increasingly strong effects at higher domains. Recent work on phonetic and phonological effects of focus have also found that narrow focus can have hyperarticulatory effects on segments (e.g., Avesani et al. 2007; Müche and Grice 2014).

This paper investigates effects of prosodic prominence on Nepali stops, which have a four-way contrast in voicing and aspiration (voiceless unaspirated, voiced unaspirated, voiceless aspirated, and voiced aspirated). Specifically, I examine both boundary and focus effects on laryngeal properties (degree of aspiration and deaspiration of phonemically aspirated stops), and supralaryngeal properties (sporadic spirantization).

Forrest Davis and Dr. Abby Cohn presented a poster titled  "Effects of lexical frequency and compositionality on phonological reduction in English compounds" at the 25th Architectures and Mechanisms of Language Processing Conference (AMLaP 2019), held Sep 6-8, 2019

Poster Abstract:

This work investigates the interplay between phonological reduction, lexical frequency effects, and relative compositionality of compounds. That is, are more opaque compounds (​cupboard)​ different from more transparent ones (​blueberry)​? ​

We establish a gradient measure for compositionality of a compound by conducting a survey of 24 native American English speakers. As a measure of phonological reduction, we utilize the duration of the final rime of the compound in the Buckeye Corpus compared to the duration of the same rime in monosyllabic nouns. For a measure of lexical frequency, we calculate the Pointwise mutual information (PMI) score for each compound using the frequency data from the Corpus of Contemporary American English.

Using a two-sample t-test we show that compounds with lower ratings of compositionality are significantly shorter in duration than would be predicted given the expected duration of their rimes. Additionally, we show that after controlling for a number of factors, compositionality in 21 compounds is a significant predictor of degree of phonological reduction in a linear regression model. Specifically, the degree of compositionality attributed to a nominal compound in English is a statistically significant factor in predicting the duration of the final rime of a compound. This yields a positive correlation, where the less compositional a compound is the shorter its final rime.

These results provide evidence that semantic opacity in compounds has reflexes in phonological form. We argue that phonological reduction needs to be addressed by both theoretical and empirical aspects of compound representation.