The origins of Japanese h from an element-based perspective

Abstract

This paper examines the historical and phonological properties of Japanese h in an Element Theory approach (Nasukawa 2005, Backley 2012). It argues that the element |U| is naturally weak in Japanese, which accounts for two synchronic idiosyncrasies — the restricted distribution of labials and rounded vowels, and the patterning of h with labials. The analysis also offers insights into how diachronic change may be implemented. In modern Japanese, labiality is phonetically weak: the ‘rounded’ segments u/w are produced as unrounded [ɯ]/[ɰ], while labial p is banned from certain contexts. These facts suggest that |U| is also phonologically weak in Japanese, which is expressed in terms of structural headedness: headed (strong) |U| represents labials while non-headed (weak) |U| represents velars (Backley & Nasukawa 2009). Moreover, Japanese |U| has become weak, giving it unrestricted distribution in (non-headed) velars but a contextually conditioned distribution in (headed) labials. The restriction on labials is captured by claiming that for headed |U| to be realised, it must co-occur with another ‘dark’ element. The division between dark {|A|, |U|, |L|} and light elements {|I|, |H|, |Ɂ|} is grounded in acoustics but also has an impact on phonological patterns cross-linguistically, such as the behaviour of h and labials in Japanese.