[PLing] Phonology of Japanese: Talk by Connor Youngberg (CNRS/UMR-6310 & Université de Nantes)

[Markus Poechtrager]

Dear all,

I would like to extend a cordial invitation to a talk by Dr. Connor 
Youngberg (CNRS/UMR-6310 & Université de Nantes) on

"Defining deficient moras in Japanese: evidence from tone"

Monday, December 10, 2018
15:15–16:45
Seminarraum 7, Sensengasse 3a, 1090 Vienna

For further details see abstract below.

Looking forward to seeing many of you,
Markus Pöchtrager

——


Connor Youngberg (CNRS/UMR-6310 & Université de Nantes)
Defining deficient moras in Japanese: evidence from tone

Abstract: In this talk, I discuss the role special or deficient moras in 
Japanese with reference to pitch accent, here analysed as privative 
tone. I focus on Tokyo Japanese, which exhibits only CV, CVV and CVC 
syllables.

Since at least Bloch (1954), the role of final positions or moras within 
the Japanese syllable has been considered a deficient one (see also 
McCawley 1968, Haraguchi 1977, Itoh 1987, Vance 1987, Labrune 2012 and 
Kawahara 2015). This set includes the first half of a geminate  (Q), the 
moraic nasal (N), the second half of a diphthong (J) and the second half 
of a long vowel (R). Possible syllables are shown in 1.

(1) Possible Japanese syllables
a. [hi] "day"
b. [ko:] "this way"
c. [hoN] "book"
d. [bat:a] "grasshopper"
e. [kai] "clam"

These moras are the object of analysis in recent work by Vance (2008), 
Labrune (2012) and others. One hallmark of their deficiency is the 
absence of accent on a special mora in a lexically accented word, e.g. 
*/kaí/ "clam".  However, it is simply assumed that the set of deficient 
moras cannot support a tone as it is not the head position within a 
syllable. Alternatively, Labrune (2012) has proposed that deficient 
moras are deficient as they are not a preferred CV syllable. She further 
proposes a hierarchy of moras and a constraint banning accentuation on 
deficient moras (NADM) to account for cases of non-accentuation. 
Labrune's proposal ties deficiency to representation in an important 
manner. However, I argue that Labrune's analysis can be improved upon 
with revised representations and that the analysis of tone and tone 
bearing units can be derived more easily and more accurately with an 
altered understand of the mora.

I revisit the deficiency of these moras and claim that 1) R, J and N are 
all vocalic segments, 2) that these moras may support a high tone in 
limited contexts and 3) Q must be treated differently. Special moras are 
to be divided into vocalic and non-vocalic dependent moras. This can 
account for marginal tone assignment on special moras as well as the 
facts of tone spreading in Tokyo Japanese, which I present in the full 
talk.

I frame my analysis within the framework of Strict CV (Lowenstamm 1996), 
building on previous work by Yoshida S. (1996) and Yoshida Y. (1999). 
Moras (here, CV units) are divided based on the status of the nucleus, 
namely its filled or licensed/governed status.

I derive degeneracy and TBU patterning from the dependent or empty 
status of the V position. In doing so, segmental hierarchies and 
constraints can be excluded.

In my conclusion, I briefly discuss cases where "special moras" are 
regularly tone bearing units, focusing on the Kansai dialect.

-- 
Mag. Dr. Markus A. Pöchtrager
Institut für Sprachwissenschaft / Department of Linguistics
Universität Wien / University of Vienna