An experiment was conducted to assess phonetic evidence for categorically distinct prosodic organization associated with restrictive vs. non-restrictive relative clauses in English (RRC vs. NRRC).

While most theories presuppose that syntactic differences are mapped to categorical prosodic variations, we do not take this for granted but rather examine whether there is evidence for categorically distinct prosodic structures in phonetic data.

Acoustic and articulatory data were collected from twelve participants. They read two types of RCs in a wide range of variation in speech rate which was elicited by a moving visual analogue rate cue. We first assessed whether the functional relations between speech rate and various phonetic measures at phrase boundaries differed by syntactic context. In addition, linear and regression mixture models (i.e. nonlinear models) were fit to each of the articulatory and acoustic measures within each syntactic context, and the Akaike Information Criterion (AIC) was used to determine whether the nonlinear model provides a substantially better fit than the linear model.

Both of our analyses showed weak evidence for the existence of distinct prosodic categories.