The phonetics and pragmatics of spontaneous speech intonation in Mexican Spanish

examine phonetic trends such as fundamental frequency (F0) peak placement and downstepping contours. These studies provide important insight into the intonation of laboratory (i.e. read or scripted) speech, however, much remains undiscovered about spontaneous speech intonation in Spanish. One study addressing this issue, Face (2003), compares the intonation of Peninsular Spanish declaratives in spontaneous speech to those of laboratory speech. The most significant differences found concern five phonetic features: downstepping, final lowering, F0 rises through stressed syllables (or lack there of, i.e. deaccenting), and F0 peak alignment. The present study uses Rao's (2004) investigation of the spontaneous speech of Madrid Spanish declaratives, which explores the manifestations of the phonetic features detailed in Face (2003) based on pragmatic meaning conveyed, as a point of departure and attempts to fill in remaining gaps in Spanish intonation research by focusing on the connection between phonetics and pragmatics in spontaneous speech declaratives of Mexican Spanish. The features presented in Face (2003) are analyzed based on F0 contours of utterances belonging to the five pragmatic categories of speech acts developed by Searle (1977), which are representatives (subjective assertions about the world), directives (attempts to get the hearer to perform an action), commissives (promises), expressives (inner states of a speaker that represent known information, i.e. apologies or congratulatory remarks), and declarations (signals of a change in a state of affairs). Inspired by the methodology of Hualde (2002), a native-speaker linguist of the Mexico City dialect was provided with three situations for each of the five pragmatic categories mentioned, and produced an intonation pattern consisting of a few sentences that she found appropriate for each situation. Face (2003) notes that although this method of data collection may seem contrived, it is an efficient means of accessing isolated examples of spontaneous speech without having to utilize a corpus of data. The results show that the frequency of downstepping in areas of F0 contours between pauses or hesitations is 66% for representatives, 40% for directives, 30% for commissives, 17% for expressives and 33% for declarations. Furthermore, final lowering is only present in representatives, commissives and declarations. Deaccenting is found in 17% of accentable words in representatives, 15% of accentable words in directives, 18% of accentable words in commissives, 17% of accentable words in expressives, and 7% of accentable words in declarations. Finally, F0 peaks align with stressed syllables in following quantities: 22% for representatives, 36% for directives, 30% for …