Phonetic behavior in proficient bilinguals

This chapter discusses a selection of the literature on the phonetic behavior of proficient bilinguals. It examines both perception and production, and it focuses on what is known about a particular bilingual group: Catalan–Spanish bilinguals. This population has received a lot of attention because it allows for the exploration of bilingual individuals with different experience profiles who reside in a speech community where both languages enjoy similar social status and are thus likely to be used by any member of the community in any given day. Phonetic research on this bilingual population has been concerned mostly with addressing the following question: What is the role of the age of first exposure to an additional language in the manner in which a bilingual will represent and process this language? Research on this population has sparked a wealth of investigations on other populations in order to address this question from multiple perspectives.

Production of French close rounded vowels by Spanish learners

This chapter introduces the corpus-based L;2 French phonology research program Interphonologie du Français Contemporain (IPFC, Interphonology of Contemporary French) and provides an illustration of its methodological approach with a population of Spanish university students learning French as an L;2. For these learners the phonemic contrast between the two close rounded French vowels /y/ and /u/ is known to be difficult to acquire, but most studies in the past relied only on acoustic analyses of laboratory speech data elicited from rather few subjects. Within the IPFC framework, on the basis of a single multitask survey protocol for all populations of learners of L;2 French to ensure data comparability, data processing is carried out with an ad hoc auditory coding procedure which integrates contextual information and target-likeness assessment. In this chapter the results of three different approaches to process the /y-u/ Spanish production data are compared.

Articulatory setting, articulatory symmetry, and production mechanisms for Catalan consonant sequences

This study reports articulatory and acoustic data for three Catalan dialects (Eastern, Western, Valencian), showing that the sequences /tsʃ/ and /sʃ/, and /tʃs/ and /ʃs/, are implemented through analogous production mechanisms and therefore that fricative+fricative and affricate+fricative sequences behave symmetrically at the articulatory level. Analysis results also reveal a clear trend for regressive assimilation in the case of /(t)sʃ/ and for blending or a two-target realization in the case of /(t)ʃs/; differences in degree of articulatory complexity among the segmental sequences under analysis account for these production strategies. Moreover, the final phonetic outcome is strongly dependent on the dialect-dependent articulatory differences in fricative articulation; thus, in Valencian, /(t)sʃ / may undergo regressive assimilation or blending and /(t)ʃs/ regressive assimilation, owing to a more anterior lingual constriction for /ʃ/ than in the other dialects.

Modeling assimilation

This chapter discusses recent developments in the study of voicing assimilation as manifested in Spanish preconsonantal sibilants. It uses experimental results to develop a model to capture the observed patterns and make predictions about the behavior of voicing assimilation in Spanish. The central argument is that voicing assimilation in Spanish is the result of gestural blending at the laryngeal level and can be couched within the framework of Articulatory Phonology. The main evidence for the gestural blending account comes from a review of the data available on the topic and, more precisely, from the reanalysis of the data from two experiments run by Rebeka Campos-Astorkiza that provide acoustic data to substantiate the claims put forward here. Furthermore, these experiments allow the evaluation of some of the challenges of working with acoustic data when analyzing voicing assimilation.

An articulatory account of rhotic variation in Tuscan Italian

Rhotic variation in a spoken variety of Tuscan Italian is investigated. The chapter takes a multilevel articulatory approach, based on real-time synchronization and analysis of acoustic, electropalatographic (EPG), and ultrasound tongue imaging (UTI) data. Contrary to the expectations based on the received dialectological literature, it emerges that speakers produce various alveolar variants: taps, trills, fricatives, and approximant realizations. To examine the factors that may constrain the variation of /r/, a multiple correspondence analysis is carried out. The result is that there are significant associations between the phonetic properties of /r/ variants and their preferred contexts of occurrence. A particular focus is then placed on the articulatory properties of the singleton–geminate distinction. It is shown that the length contrast is maintained but contrary to expectation, trills are not primarily used for geminates. Instead, each speaker differentiates the singleton from the geminate according to a variety of production strategies.

Romance sounds

The study of Romance sounds, and their structure, has for centuries occupied a fundamental position in core phonetic and phonological research. And for good reason. By examining the typological symmetries and asymmetries among the different Romance languages we have learned much about the universal properties of language and the production/perception mechanisms which underscore acquisition and sound change. This provides a rich terrain in which to formulate and test new hypotheses related to sound systems and their development. The commissioned authors in the current volume present recent research in the acoustic, articulatory, phonological, perceptual, and acquisition domains from an array of theoretical foci. The work presented here is sure to have a far-reaching impact in the speech sciences for many years to come.

Production and perception in the acquisition of Spanish and Portuguese

Learning to listen to and produce the sounds of a new language is a difficult task for many second-language learners. While there is a large corpus of literature that investigates Spanish and Portuguese learners’ perception and production of an L;2, particularly English, there is relatively little research available for the opposite scenario, namely, how speakers of other languages learn to perceive and produce the sounds of Spanish and Portuguese. The purpose of this chapter is to provide a critical review of the available literature in this less studied area. First some general facts relating to non-native and L;2 speech perception and production are presented, including the theoretical models that aim at explaining these phenomena. A review follows of the empirical findings currently available for L;2 speech production and perception in Spanish and Portuguese, and how these two abilities relate in the process of acquiring the sounds of these languages.

Language proximity and speech perception in young bilinguals

Linguistic experience shapes speech perception from the earliest stages of development. Infants growing up in bilingual contexts are exposed to a more complex linguistic input from which they will gradually build language-specific phonetic and phonological categories, eventually characterizing words in their early lexicons. Input languages can show different levels of proximity relative to their rhythmic, phonetic, phonological, or lexical properties. Does language proximity affect early speech perception processes, from language differentiation to perceptual narrowing and phonological representation of words in the bilinguals’ vocabulary? Data from infants growing up in Catalan-Spanish contexts, acquiring a close pair of Romance languages, are reviewed and contrasted with data from infants exposed to more distant language pairs. It is argued that language proximity can determine specific adjustments in bilinguals’ early phonetic perception and phonological encoding of words. Language proximity factors can account for differences among bilingual infants’ trajectories previously reported in the literature.

Perceived phrasing in French

Studies of prosodic structure in French have tended to concentrate on production in order to identify what levels of phrasing speakers differentiate. But what are listeners hearing? The chapter details the examination of some specific structures (dislocations and focused phrases) to see whether listeners’ perceptions align with the prosodic structures that have been proposed. The approach used is Rapid Prosody Transcription, in which untrained listeners are asked to identify words they perceive as prominent or places where they perceive a break in the flow of speech (phrasal boundary). Descriptions of French agree that prominences occur either initially or finally in the smallest-sized phrase. While the majority of locations where listeners marked boundaries were perceived as preceded by a prominent word, because more prominences than boundaries were marked these did not necessarily co-occur with boundaries. Dislocated words were perceived as both prominent and separated from the main clause by a boundary.

Brazilian Portuguese rhotics in poem reciting

Rhotics are diverse in nature and conditioning factors. In this work the use of Brazilian Portuguese rhotic varieties in the context of recitation of a poem by a professional actor is analyzed and articulatory and acoustic descriptions of the Brazilian Portuguese rhotic varieties are provided. The choice of varieties in the analyzed speech data is interpreted as deriving from the shaping of acoustic characteristics so as to produce impressive meaningful effects. These effects are achieved through the symbolic use of sound, and the fact that sounds are sense-impressive and meaning-expressive is taken to be the foundation of speech expressivity.