Volume 82, Issue 1

The emergence of multiethnolects, i.e. specific speaking styles or varieties associated with second and third generation speakers from immigrant backgrounds, has been observed and studied in several major cities in Europe and elsewhere in the world. The multiethnolect that is the focus of this study is one such variety of colloquial German. Most previous research on multiethnolectal German has focused on grammatical features. This paper reports on the first comprehensive study of the vowel system (vowel quality and global vowel space size) of multiethnolectal German, based on data from Stuttgart. The results show that the vowel space of multiethnolectal speakers is in generally more centralized than that of a comparison group. A more detailed analysis reveals that the linguistic background plays an important role, as speakers with a Turkish or South Slavonic language background are responsible for this effect.

Mnong Râlâm is a South Bahnaric language (Austroasiatic) that is traditionally described as preserving a voicing contrast in onset obstruents, contrary to other languages of the Mnong/Phnong continuum. Acoustic results yield evidence that this voicing distinction is less robust than previously suggested and is redundant with a register contrast realized on following vowels through modulations of F1 at vowel onset (and more limited variations of F2 and voice quality). A perception experiment also shows that F1 weighs heavier than closure voicing in identification. The existence of a language in which voicing and register are redundant allows us to revisit previous models of registrogenesis. It suggests that the acoustic properties of register may develop and become distinctive while voicing is still present rather than as a direct consequence of devoicing. The unexpected discovery of register in Mnong Râlâm also challenges previous reconstructions of Proto-South-Bahnaric, in which only voicing is postulated, based on hitherto limited available descriptive materials.

This paper explores the perception of two diachronically related and mutually intelligible phonological oppositions, the onset voicing contrast of Northern Raglai and the register contrast of Southern Raglai. It is the continuation of a previous acoustic study that revealed that Northern Raglai onset stops maintain a voicing distinction accompanied by weak formant and voice quality modulations on following vowels, while Southern Raglai has transphonologized this voicing contrast into a register contrast marked by vowel and voice quality distinctions. Our findings indicate that the two dialects partially differ in their use of identification cues, Northern Raglai listeners using both voicing and F1 as major cues while Southern Raglai listeners largely focus on F1. Production and perception are thus not perfectly aligned in Northern Raglai, because F1 plays a stronger role in perception than production in this dialect. We conclude that mutual intelligibility between dialects is possible because they both use F1 for identification.