摘要：Some prominent models of speech production assume that words are stored as linear sequences of phonemeswhich are syllabified at the moment of production (Levelt et al., 1999). The rationale for this is that syllables arecompletely predictable from the sequence of phonemes. It is also assumed that when a language often resyllabifieswords in connected speech (as in English), storing syllable structure is redundant and wasteful (Levelt el al., 1999 ).
