On the undecidable syllabification of /sC/ clusters in Italian: Converging experimental evidence

A widespread claim defines /sC/ clusters as universally heterosyllabic, although the actual distribution of these clusters, as well as their reactivity to phonological processes, may vary from language to language. This claim has also been put forth for Italian. Upon close examination, however, none of the purported arguments turn out to be compelling. Bertinetto (1999) thus concluded that the syllabification of /sC/ clusters in present-day Italian is likely undecidable: the phonology of the language is in most cases compatible with both types of syllabification, tautosyllabic and heterosyllabic. The present paper addresses the issue from the experimental perspective, summarizing the results of a number of psycholinguistic investigations conducted in recent years. In doing so, the paper does not engage in issues relating to theoretical phonology. The problem at hand is how the native speakers concretely, i.e. observably, deal with /sC/ clusters. This does not mean, however, that the paper has no theoretical ambitions. The discussion will specifically focus on the underlying assumption, present (or implicit) in most approaches to syllabification, that all segments should be unambiguously syllabified. The various psycholinguistic tasks reviewed yielded contrasting results, but altogether converged in suggesting that Italian /sC/ clusters should not be confused with the typically tautosyllabic clusters, nor with the typically heterosyllabic ones, strongly suggesting once again the undecidable nature of the /sC/ clusters’ syllabification in Italian. This is only compatible with views that do not demand that every consonantal sequence be uniquely, i.e. unambiguously syllabified.