A misleading syllable-based generalization about Japanese SJ+/zu/ compounds

Abstract

Abstract In 1910, Ogura Shinpei published a meticulous critique of Benjamin Smith Lyman’s famous 1894 article on rendaku . In the course of a thorough examination of compounds consisting of a single Sino-Japanese morpheme followed by /su/∼/zu/ ‘to do’, Ogura noted that all the examples with rendaku (i.e., with /zu/) have a monosyllabic first element. This observation invites the inference that there is a causal connection between first-element monosyllabicity and rendaku in X+/zu/ compounds, but a careful review of the history of these vocabulary items indicates that the correlation between monosyllabicity and rendaku is just an accident. There is no reason to believe that first-element monosyllabicity has ever been an active phonological constraint, and the pattern that Ogura identified cannot be used to bolster the view that syllables distinct from moras are real units in modern Tokyo Japanese.