This talk exploits our improved understanding of tongue root harmony systems, particularly in NE Asia (see Joseph et al 2020 and the work cited there), to reconsider the long-mooted issue of the “laws of syllable combination” in Old Japanese identified by Arisaka Hideyo (1934). Building on an insight of Hayata (2006), I show that the Old Japanese vowel co-occurrence patterns have the properties of a vestigial NE Asian RTR (retracted root tongue)-dominant system. I explore and extend previous work on the reflexes of this system in Eastern Old Japanese suffix selection.

The talk then focuses on the observation of Joseph et al (2020) that the well-known Eastern Old Japanese alternation between /u/ and /o/ in the attributive suffix may reflect a TR harmony- conditioned pattern. Attributive -o occurs almost exclusively after stem-final [+RTR] vowels; of the 52 examples of -o collected by Ikier (2006), only one occurs after an unambiguous [-RTR] vowel; 4 occur after stems with final vowel /i/ or /u/, which may be the result of mid vowel raising from *e or *o.

The talk concludes with a discussion of the relation between RTR- versus ATR (advanced tongue root)-dominant harmony and vowel inventory in light of Van der Hulst’s (2017) formalization of a proposal by Casali (2003, 2008) that ATR-dominant systems typically have only one pair of contrasting high vowels.

Selected references: Hattori, Shirō 1976. J̄-dai Nihongo no boin taikei to boin ch̄-wa, Hayata, Teruhiro. 2006. Jōdai Nihongo boin chōwa obegaki, Joseph, Andrew, Seongyeon Ko, and John Whitman. 2020. A comparative approach to the vowel systems and harmonies in the Transeurasian languages and beyond, Van der Hulst, Harry. 2017. Deconstructing tongue root harmony systems, Whitman, John. 2016. Nichiryū sogo no on’in taikei to rentaikei, izenkei no kigen.

Full abstract and bio: https://tinyurl.com/udbku6bt