Previous research on /s/-lenition in Spanish has relied almost exclusively on impressionistic coding for the independent variable "/s/-realization" (cf. Brown 2008, File-Muriel 2009, Poplack 1979, Terrell 1979. In these studies, researchers are limited to the available devices accorded to them by the International Phonetic Alphabet (e.g. /s/ as [s], [z], [h], Ø, etc.). Although auditory acoustic analysis (i.e. transcription) is an extremely useful descriptive device, it can be influenced by the transcriber's background and expectations. Additionally, acoustic details are lost when tokens are forced into categorical groupings, such as short vs. long duration. The present study examines the production of /s/ by eight university students from Cali, Colombia in informal sociolinguistic interviews. This research departs from the audio transcription tradition, proposing a metric for quantifying the realization of /s/ by employing three scalar dependent variables: center of gravity, /s/ duration, and percentage of voicing. Such a method is less vulnerable to bias, given that it is based on acoustic measurements that are not necessarily perceptible to the human ear. The results of the linear regressions indicate that the dependent variables are significantly conditioned by a variety of linguistic factors: lexical frequency, local speaking rate, stress, speaker, position within the syllable, the preceding and following phonological context, word length, and lexical class. The magnitude of effect trends are elucidated by predicted probability plots. For example, as lexical frequency increases, /s/ duration decreases, center of gravity decreases, and percentage of voicing increases, all indicative of lenition. We argue that /s/ lenition is better explained in gradient terms, rather than categorical ones. This paper contributes to the growing body of literature documenting the influence of frequency effects on this otherwise well-studied phenomenon of phonological variation and change.

Several linguistic factors have been shown to condition the reduction of syllableand word-final /s/ in Spanish (e.g. ¿Cómo estás tú? > ¿Cómo e [h]tá [ø] tú? 'How are you?'). These include word length (mono vs. polysyllabic words), the prosodic stress of the syllable in which syllable-or word-final /s/ occurs (stressed vs. unstressed), the position of /s/ within the word (internal vs. final), as well as the factor group shown to the have the greatest conditioning effect, the phonological context that follows syllable-or word-final /s/ (consonant vs. vowel vs. pause).The influence of factors dealing with frequency has been shown only recently for this phenomenon (Brown 2004, Brown & Torres Cacoullos 2002, 2003, FileMuriel 2007, Minnick Fox 2006. These studies show that the token frequency of the word in which final /s/ occurs as well as the frequency with which a word occurs in a phonological context favorable to reduction (e.g. before a following consonant instead of a vowel or pause) significantly condition final /s/ reduction. Based in the usage-based model of phonology (Bybee 2001), this dissertation contributes to our understanding of the most-studied variable in Spanish, and consonant weakening more generally, by providing evidence that frequency factors significantly condition the reduction. An analysis was performed of 23,307 tokens of syllable-and word-final /s/ taken from four dialects of Spanish: Cali, Colombia (Travis 2005), northern New Mexico and southern Colorado (Bills & Vigil 1999, Brown 2004), Mérida, Venezuela (Domínguez & Mora 1995) and San Juan, Puerto Rico (Cortés -Torres 2005). The selection of these dialects is based on the overall rate of final /s/ reduction as high (Venezuela, Puerto Rico) and lower (Colombia, New Mexico). Tokens are coded for phonological context, the traditional, wellstudied factors mentioned above, the token frequency of the word in which final /s/ occurs and the frequency with which the word occurs in a phonological context favorable to reduction. The results of multivariate analyses (Sankoff 1988, Sankoff, Tagliamonte & Smith 2005) support the usage-based model of phonology and Exemplar Theory. The following phonological context loses conditioning effect as the overall rate of reduction approaches categorical reduction. Further, the results provide evidence that the token frequency of the word in which syllable-or word-final /s/ occurs significantly conditions this phenomenon, such that, in general, syllable-and wordfinal /s/ is reduced more often in high-frequency words than in low-frequency ones. Likewise, the string frequency of two-word combinations is also significant. When Brought to you by | University of California Authenticated Download Date | 6/7/15 2:32 AM

No abstract

Variability abounds in speech. According to usage-based accounts, lexical representations reflect phonetic variants of words resulting from contextual conditioning. Because faster speech contexts promote durational shortening of words and segments, words that occur more often in fast speech may be more reduced than words commonly used in slow speech, independent of the target’s contextual speech rate. To test this, linear mixed-effects models including a word form’s ratio of conditioning by fast speech contexts (FRCRATE) are used to predict the duration of Spanish /s/ and words containing /s/ in a corpus of spoken Spanish. Results show that words’ cumulative exposure to relatively fast speech affects phonetic realizations independent of factors operative in the production contexts. Thus, word and segment rates reflect cumulative (lexicalized) effects of words’ experience in fast speech contexts. The results suggest that lexically specific cumulative measures should be incorporated into models of linguistic variation and change.

There is mounting evidence that words that occur proportionately more often in contexts that condition a phonetically-motivated sound change end up changing more rapidly than other words. Support has been found in at least modern-day Spanish, Medieval Spanish, bilingual English-Spanish, and modern-day English. This study tests whether there is support for this idea with regards to the variable voicing of word-final /s/ in Spanish. An analysis of 1431 tokens of word-final /s/ spoken by 15 female speakers of Mexican Spanish living in Salinas, California, USA is performed. The response variable is the percentage of the /s/ segment that is voiced, and the effect of a handful of predictor variables shown in the literature to condition /s/ voicing is investigated. The variable of interest is forms’ ratio of conditioning (FRC), or the proportion of times with which word types occur in the context that conditions voicing of word-final /s/. The results of a series of 40 beta regression models indicate that FRC significantly conditions the percentage of voicing of word-final /s/ in these data. Also, the effect of manipulating two aspects of FRC operationalization is analyzed. This study adds to the growing body of literature documenting the importance of cumulative contextual information in the mental representation of words.