Reconstructions

Note

PAn *taRah ‘to wait’ is intriguing for several reasons. First, it is one of the few cognate sets which appears to provide evidence of homorganic nasal substitution in Formosan languages. Second, reflexes of *ma-naRah vs. *taRah are found even within the Atayalic group, distinguishing dialects of Atayal from those of Seediq (Li 2004:688), thus implying that *maŋ- was active in the relatively recent history of these languages, despite its general absence in Taiwan. Third, *taRah carries a variety of different affixes in all languages that reflect it, and it is very difficult to find examples of the simple base. Despite initial appearances, however, Atayal, Seediq and all known MP reflexes provide evidence for a doublet *naRa, and it is the prefixation of this base that has given the false impression of homorganic nasal subsitution. Part of this comparison was first recognized by Tsuchida (1976:232).