Published online by Cambridge University Press: 27 June 2016
Does phonological conditioning have a privileged status in allomorphic alternations? Most synchronic descriptions appear to assume that if simple phonological conditions can account for some alternation, they automatically constitute the relevant factors chosen by speakers in their internalized grammars (but cf. Hooper 1976:87 or Klausenburger 1979:29–36 for divergent views). Intuitive notions of simplicity cannot provide a solution because simplicity is actually an empirical issue (cf. Chomsky 1965:37–47). The answer can only come from the analysis of external evidence, i.e., errors, language acquisition, historical changes, etc.
I would like to thank my mentor Rajendra Singh for encouraging me to write this paper, and for the help provided in countless discussions. He is of course not responsible for any errors, mistakes and distortions which may still appear here.