Part ofSonic Signatures: Studies dedicated to John Harris
Edited by Geoff Lindsey and Andrew Nevins
[Language Faculty and Beyond 14] 2017
► pp. 17–32
On the surface Acadian French appears to have word-initial and word-internal branching onsets. A closer look at the data however, reveals that no branching onsets can be followed by schwa. This fact leads me to propose that the TR clusters of Acadian French are bogus and that onsets do not branch in this dialect of French. [TRV] sequences have the structure of two separate onsets with a p-licensed and unrealised empty nucleus occurring between the stop and the liquid. When such a bogus cluster (i.e. /TøR/) is followed by a nucleus which is lexically empty, the form surfaces as [TəR] as a result of ECP.