An Optimality-Theoretic account of Sa-insertion in Japanese

1
An Optimality-Theoretic account of Sa-insertion
in Japanese

• Shin-ichiro Sano
• (Graduate Student, Sophia Univ. )
• shinichirosano_at_gmail.com

2
Sa-insertion

• Sa-insertion
• variation of causatives in Japanese
• Formation of causatives in Japanese
• attaching causative suffix (as, ase) to verb
  stems
• Standard causative V - ase
• Sa-insertion V - as-ase

3
Examples (causative)

• Standard causative
• Watasi-ga Taroo-ni sore-o yar-ase-ru.
• I-Nom Taro-Dat it-Acc do-Cause
• Sa-insertion
• Watasi-ga Taroo-ni sore-o yar-as-ase-ru.
• I-Nom Taro-Dat it-Acc do-Cause-Cause
• I let Taro do it.

4
Examples (honorific)

• Standard causative
• Watasi-ga sore-o yar-ase-teitadaku.
• I-Nom it-Acc do-Cause-benefactive
• Sa-insertion
• Watasi-ga sore-o yar-as-ase-teitadaku.
• I-Nom it-Acc do-Cause-Cause-benefactive
• I will do it. (honorific)

causative honorific
Standard causative
Sa-insertion -
5
Quantitative analysis

• Corpus
• On-line full-text database of the minutes of
  the Diet
• Sampling by birth-year ? 81 members from Tokyo
• Table 1. The rate of Sa-insertion in all
  causatives

Sa-insertion 317
Standard causatives 4,708
rate of Sa-insertion () 6.31
6
Chronological change of Sa-insertion
Pseudo R20.096
7
Summary of quantitative analysis

• Sa-insertion
• is first observed in 1949.
• does not contain the sequence sasa.
• Some Sa-insertion is observed (gradually
  increasing) until the 1980s. After the 1980s, it
  suddenly exploded.
• ?
• is in the beginning of a Language change.
• The change will proceed in an S-curve fashion.

8
Double causative

• Okada (2003)
• Sa-insertion is a double causative.
• The meaning is not double causative but
  honorific.
• Double causative
• Double causative is not allowed in Japanese.
• (Shibatani 1973, among others)
• ? suppressed to a single causative
• V - as - ase ? V - ase
• Variation suppressed ? Standard causatives
• unsuppressed ? Sa-insertion

9
2 types of Japanese Verbs (Bloch 1946)

• Consonant verbs (e.g. hasir- run, yar- do)
• Vowel verbs (e.g. mi- see, tabe- eat)
• Sa-insertion is restricted to Consonant verbs.

Standard causative Sa-insertion
Consonant verb yar-ase yar-as-ase
Vowel verb tabe-sase tabe-sa-sase
10
OT analysis Double causative

• Double causative is suppressed to a single
  causative.
• OCP (morph)
• No identical morphological categories are
  adjacent.
• Double causative is blocked by OCP (morph) and
  single causative surfaces.

Input /yar-as-ase/ OCP (morph)
a. ya.ra.sa.se !
?b. ya.ra.se
11
Paradigm Contrast

• Contrast between causative and honorific (reduces
  functional load)
• ParContrast (Paradigm Contrast, Ito and Mester
  2004)
• The cells of a paradigm are pair-wise
  phonologically distinct.
• OCP (morph) ParContrast Standard causative
• ? ParContrast OCP (morph) Sa-insertion
• Form and Content - 2 to 2

Candidate paradigms for /yar-as-ase/ ParContrast OCP (morph)
?a. causative ya.ra.se honorific ya.ra.sa.se
b. causative honorific !
ya.ra.se
12
sasa

• Sa-insertion does not contain the sequence sasa.
• OCP (µ)
• No identical moras are adjacent.

Candidate paradigms for /das-as-ase/ OCP (µ) ParContrast OCP (morph)
a. causative da.sa.se honorific da.sa.sa.se !
?b. causative da.sa.se honorific da.sa.se
13
Sa-insertion consonant verbs

• Sa-insertion is restricted to consonant verbs.
• Any Sa-insertion with vowel verbs contains sasa.
• ? phonological reason (not lexical or
  morphological)

Candidate paradigms for /tabe-sas-ase/ OCP (µ) ParContrast OCP (morph)
a. causative ta.be.sa.se honorific ta.be.sa.sa.se !
?b. causative ta.be.sa.se honorific ta.be.sa.se
14
Allomorph Correspondence

• Paradigm optimization
• AlloCorr (Allomorph Correspondence, Ito and
  Mester 2004)
• Morphs in a relation of allomorphy are
  identical.
• Form and Content - 1 to 1

Candidate paradigms for /yar-/ and /tabe-/ OCP (µ) ParContrast AlloCorr OCP (morph)
?a. consonant verb ya.ra.sa.se vowel verb ta.be.sa.se
b. consonant verb ya.ra.se vowel verb ta.be.sa.se !
15
Language Change

• Change of the ranking
• OCP (morph) ParContrast, AlloCorr ?
  ParContrast, AlloCorr OCP (morph)
• Partially ordered grammar
• OCP (morph), ParContrast OCP (morph)
  ParContrast, AlloCorr ? Standard caus.
• AlloCorr
  ParContrast, AlloCorr OCP (morph) ?
  Sa-insertion

Initial stage (a) Middle stage (a) (c) Final stage (c)
OCP (morph) ParContrast, AlloCorr ParContrast, AlloCorr, OCP (morph) ParContrast, AlloCorr OCP (morph)
Standard causative 100 Sa-insertion 0 Standard causative 50 Sa-insertion 50 Standard causative 0 Sa-insertion 100
16
Conclusion

• Sa-insertion - Language change (initial stage)
• Contrast and Paradigm optimization
• Standard causative Sa-insertion
• Contrast Form 1, Content 2 Form 2, Content 2
• ParaOptim Form 2, Content 1 Form 1, Content 1
• ? change of ranking
• OCP (morph) ParContrast, AlloCorr
• Paradigm optimization in Ra-deletion (variation
  in passives) (Ito and Mester 2004)
• ? Global change in Japanese voice system