A produção da lateral pós-vocálica em uma comunidade bilíngue: aspectos do Português sob a influência do Polonês como língua de imigração

Abstract

This study aims to describe the acoustics and articulation of the production of lateral liquid consonant /l/ in postvocalic position in the speech of a community characterized by the use of Polish as an immigrant language. In the rural areas of southern Brazil that are influenced by immigrant languages, Altenhofen and Margotti (2011) point out that there is a preservation of the segment in syllable-final position and that it is, therefore, produced as velarized or alveolar. The segment is acoustically characterized by higher values for the second formant (F2), as compared to more velarized productions (NARAYANAN, 1997). As for articulation, less velarized productions are realized with anterior tongue-body movement, which indicates that the apical gesture is more prominent than the tongue-dorsum gesture (RECASENS, 2004, 2016). To describe the lateral segment, we observed data from 12 female subjects, aged 15 to 59 years old, six of whom are Polish-Portuguese bilinguals and six are monolingual speakers of Portuguese. All the subjects are residents of the community of Barra do Arroio Grande, located in the rural area of the district of Dom Feliciano-RS, Brazil. In the first stage of the data collection, aimed at the acoustic analysis, a list of pre-established questions was used to prompt spontaneous speech. After the collection of spontaneous speech, the subjects performed two picture-naming tasks, which allowed the production of words in Polish and Portuguese, inserted in carrier phrases. The two tasks made it possible for /l/ to be produced in word-medial and word-final stressed syllables. Concerning the vocalic contexts, the first task enabled the production of the lateral in the context of the seven Portuguese vowels, while the second task enabled its production in the context of the five Polish vowels. The subjects performed each task six times. The data were recorded using a Zoom H4n audio recorder. In the second stage, the data were collected from a new round of the picture-naming tasks, but this time with words produced in isolation. In this stage, aimed at the articulatory analysis, besides the audio data, the articulatory imaging of the subjects was recorded using a Mindray DP 6600 portable ultrasound machine equipped with a micro-convex probe. A probe-stabilization headset by Articulate Instruments was used in the ultrasonic data collection. In order to test the data collection methodology, a pilot collection was carried out in the Laboratório Emergência da Linguagem Oral, located in the Center for Languages and Communication of the Federal University of Pelotas. All the other subjects had their /l/ production data were collected in their community. For the acoustic analysis, we considered the values os F1 and F2, differences between the values of first and second formants and the segment duration. For the articulatory analysis, we considered point maximun of constriction of the articulator, in order to identify a more anterior or more posterior production. The results reveal gradual velarization for the post-vocalic lateral. The spontaneous speech data show that the bilingual subjects velarize /l/ less than the monolingual ones. The articulatory analysis confirmed a graduality of the velarization of /l/. It was also observed that the segment has consecutive dorsum retraction and lowering tongue-body gestures in more velarized productions, and simultaneous apical and tongue-body gestures in less velarized productions.