Phonology Seminar Schedule

UCLA Department of Linguistics
Spring 2007

The Phonology Seminar (Linguistics 261) will meet Spring Quarter 2007 in Rolfe Hall 3114. Meetings are on Tuesdays from 4 to 6. There are still open slots; if you would like to give a talk, please contact the convener, Bruce Hayes.

Tuesday 4/3/2007

            Andrew Martin

            “Geminate avoidance in English morphology” (practice talk for GLOW)

            We will also arrange the rest of the schedule at this meeting.

Tuesday 4/10/2007

            Group discussion of Ying Lin, Weak representational bias and the discovery of linguistic categories from speech waveforms. 

            We will rotate around the room engaging in exegesis, please be ready to do so.

Tuesday 4/17/2007

            Kevin Ryan:  Practice talk on the phonology of half-rhyme (for Poetics Fest 2 at UCSC)

            Katie Schack Tang:  “Phonological evidence for the structure of Javanese compounds” (practice talk for AFLA).

Tuesday 4/24/2007

            Colin Wilson.   Practice talk, title TBA, for his colloquium at the University of Chicago.

Tuesday 5/1/2007

            Discussion of Steven A. Jacobson, “Siberian Yupik and Central Alaskan Yupik Prosody,” from Michael Krauss, Yupik Eskimo Prosodic Systems.  Download here.  Note that the download is 4 megabytes, but you only have to read pp. 25-46.

Tuesday 5/8/2007

            Roy Becker, “How do Hebrew speakers ‘trick or treat’?”  Please see abstract at the bottom of this page.

Tuesday 5/15/2007

            Vet a poster:  Ingvar Lofstedt, “Allomorphy as Perceptual Optimization".  We will meet in the hallway outside Bruce Hayes’s office in 2101G Campbell, where there is a large blank wall for Ingvar to post.

Tuesday 5/22/2007

            Russ Schuh, “South Kyengsang Korean Pitch Accent:  Reprise”  Download the abstract

Tuesday 5/29/2007

            Kevin Ryan, “Affix ordering in Tagalog”

Tuesday 6/5/2007

            Bruce Hayes —CANCELLED

 

Abstracts

 

How do Hebrew speakers ‘trick or treat’?

An experimental study of verb formation from Hebrew CCVC nouns.

Roy Becker-Kristal

 

Denominative verb formation encompasses a built-in conflict for Hebrew speakers, since it involves the mapping of a relatively free phonological structure (of the noun) onto one of a small set of rigid skeletal template paradigms (a verbal binyan), with inevitable faithfulness violations. Denominative verb formation from CCVC nouns constitute a particularly interesting case, because the two most productive binyanim, kitel (aka pi’el) and its reduplication variant kitlel are the least faithful, with CCVC rendition of the noun source appearing nowhere in their paradigm. Conversly, less productive binyanim, such as hiktil and katal (aka hif’il and pa’al), as well as less common variants of kitel such as iktel and ktolel, do offer skeletally-faithful alternatives.

The experimental study reported here examines the derivational strategies Hebrew speakers apply when coining a verb from a CCVC noun, in light of three different theoretical approaches: The first, ‘Semitologist’ approach, according to which Semitic morpho-phonology involves a unique mechanism of consonantal root extraction and its mapping onto skeletal templates, implying that there should be no difference between CCVC and other tri-consonantal noun sources. The second, ‘universalist phonological’ approach, assumes that Hebrew and all Semitic languages conform to the same constraints and mechanisms as other languages. As such, denominative verb formation involves direct mapping of the noun source onto the verb derivative, without the mediation of consonantal root extraction, and this should be manifested in sensitivity to both the skeletal structure and the vocalic melody of the source. The third, ‘lexicalist’ approach, maintains that novel morpho-phonological patterns are determined predominantly by the phonological distribution of strategies in the existing lexicon: the detailed phonological structure of the source is taken into account, but only in order to determine how relevant morphological strategies are for such structures – the relationship between the source and the strategy might be entirely arbitrary. A major prediction of this approach is that well-formed and faithful but lexically unattested strategies are morphologically irrelevant.

The results, based on 24 participants so far, clearly support the second approach, and show that the derivational strategies that native Hebrew speakers prefer are the ones most faithful to the skeletal and vocalic properties of the source, and by and large disregard lexical statistics. However, direct phonological source-derivative mapping fails to explain all the results, and lexical distribution and consonantal root extraction cannot be discarded altogether. The theoretical implications of these results are not surprising, since, unlike inflectional and concatenative derivational morphology, prosodic-templatic derivational morphology yields opaque structural relationships between independent lexical entries, from which few, if any, reliable source-derivative patterns are discernable.

Particular emphasis is given to the methodology employed in this experiment, and it will be claimed that this methodology forsees and overcomes many of the problems inherent to the rather unnatural linguistic task under study.