This paper reports on an acoustic analysis of the phonemic tap-trill contrast (/ɾ/-/r/) for first and second generation speakers of Mexican Spanish who live in the Chicagoland area. First, it is shown that speakers most commonly produce phonemic trills with a single apical occlusion, although there is much individual variation. Second, nearly all speakers realize the tap-trill contrast by means of segmental duration, and this is especially true for speakers who favor zero or one closures in the phonemic trill. These data suggest that heritage speakers make use of the limits of phonetic variation to the extent that phonological contrasts are intact in their grammar. The findings have implications for disentangling the sources of individual variation in heritage pronunciation. Specifically, we propose that between-speaker instabilities in rhotic production result more directly from an inherently variable speech signal than from the language contact situation.

The current study is the first technically-principled examination of second language Spanish intonation as it develops in a study abroad context. As such, it aims to identify patterns of development related to overall F0 contours and final boundary movements by creating individual profiles for each L2 learner at the beginning and the end of their stay abroad. Our data come from 5 English-speaking learners of Spanish enrolled in a study abroad immersion program for 7 weeks in León, Spain. We collected data through a computerized production task that elicited three context-based sentence types: declaratives, absolute interrogatives, and pronominal interrogatives. This task was administered once upon arrival to León and again at the end of the 7-week program (approximately 6 weeks elapsed between tasks). In our analysis, we calculated the initial and final tonal levels in addition to the Low and High tones of pitch accents (when applicable). For each learner the most common strategy employed at each recording time was identified. Our findings show that despite individual variation, most L2 learners modify their intonational patterns (i.e. pitch accent and boundary movements) through an increase in frequency of use of their more dominant intonation patterns and/or changes in their final boundary tone inventory.

(cf. Cruttenden, 2007; but see also Lickley, Schepman, & Ladd, 2005). The experimental findings serve to clarify a number of assumptions about the syntax-prosody interface underlying wh-question utterance signaling; they also have implications for research methods in intonation and task-based variation in laboratory phonology.

This paper is an experimental investigation on the tonal structure and phonetic signaling of declarative questions by speakers of Manchego Peninsular Spanish, a dialect of Spanish for which little experimental research on intonation is currently available. Analysis I examines the scaling and timing properties of final rises produced by 16 speakers under various pressures of tonal crowding. The quantitative results provide evidence for two contrasting nuclear pitch accent specifications: L* vs. H*. These data are consistent with the findings and analyses of final rises in Dutch, although certain time pressure effects had not been reported in this previous body of research. Analysis 2 provides a phonetic comparison of the two question contours uncovered in Analysis I with those of lexically and syntactically identical declarative statements. The findings indicate that speakers differentiate the two question contours from corresponding statement contours in dissimilar fashion and that for the L*H% contour, the terminal rise may be the only F0 signal of its question intent. Some speculation on the possible causes of this variation is offered in conclusion.

In this paper we provide a preliminary characterization of the phonemic trill (i.e., /r/) as produced by twenty-four speakers of northern and central Peninsular Spanish. The acoustic analysis revealed a considerable number of non-canonical variants containing one or zero apical occlusions. The quantitative results showed robust effects of the following three factors on trill articulation: Speaker dialect, gender, and preceding vowel. Regarding social factors, central Peninsular speakers and male speakers showed the greatest propensity to produce fewer occlusions per phonemic trill. Regarding linguistic factors, non-canonical variants were especially common in contexts of preceding /u/; we interpret this result on articulatory grounds given the antagonistic gestures required for the trill and the high back vowel. All in all, these findings offer empirical support that geographically-oriented studies within a sociophonetic framework offer critical information on the diachrony of trill consonants.