Spanish Stop Lenition and the Temporal Dimension: An Acoustic Approach

Identificador:  TDX:4452

Handle: https://hdl.handle.net/20.500.11797/TDX4452

Autors:  Soler Cervera, Antonia

Resum:
This dissertation presents a study that investigates Spanish stop lenition as a gradual weakening phenomenon which is conditioned by multiple factors. Voiced stop weakening in Spanish has been depicted as a context-dependent process by which /b,d,g/ surface as [β,ð,ɣ] in all contexts except in initial position, after nasals and, exceptionally after /l/ for homorganic /d/. From a phonetic perspective, previous studies have placed emphasis on describing lenited realizations and their features, mainly based on information provided by acoustic data. These analyses have revealed some variability in the resulting realizations of lenition processes, where the effect of stress and contextual factors has been tested in a number of studies. The present study hypothesizes that the temporal dimension has a fundamental role in Spanish stop lenition, where the target realization is not completely achieved due to timing constraints. Contextual factors are also hypothesized to exhibit different patterning across speaking rates. An experiment is presented that includes two speaking rates and different contexts, and an acoustic analysis procedure, based on Praat, is used to measure the corresponding acoustic cues. The results show that speaking rate has a major effect on the realization of lenited consonants, as the acoustic cues associated with constriction degree are significantly different across rates. Contextual effects are also conditioned by time, which triggers a range of constriction degrees for context-rate interactions. These results can be best interpreted from an Articulatory Phonology perspective, which provides a framework to explain how time and constriction degree go along in Spanish stop lenition.