Simultaneous bilingualism and speech style as predictors of variation in allophone production: Evidence from Finland-Swedish

Re-examining the /eː-ɛː/ merger in Finland-Swedish: Regional and stylistic variation

“…According to the Vowel Space Paradox (Van der Harst, 2011), while the overall vowel space of a speaker shrinks in spontaneous speech, individual vowels are expected to increase their vowel space and, consequently, their sociogeographical variation. In contrast to expectations, the findings by Strandberg et al (2021) suggested decreased sociogeographical variation in spontaneous speech. While examining the /eː-εː/ merger, the current study explores whether a similar pattern of regional variation is found in the production of other vowels in Finland-Swedish.…”

“…In contrast, participants' attention to speech is expected to increase with more formal tasks, that is, the reading of a passage of text or a list of words. To our knowledge, besides Strandberg et al (2021), this is the first Table 1. Mean F1 and F2 Hz values for /eː/ in Finland-Swedish (FS) from Reuter (1971:243) and Kuronen (2000:140), and for /εː/ in Sweden-Swedish (SS) from Fant et al (1969:28) and Kuronen (2000: study to use sociolinguistic interview data for acoustic analysis of Finland-Swedish vowel production.…”

“…The current study seeks to investigate the /eː-εː/ merger in Finland-Swedish, as well as to provide a quantitative acoustic analysis of sociogeographical variation of the vowels. In an analysis on phonetic variation in allophones [øː] and [oeː] by monolingual Finland-Swedish and bilingual Finland-Swedish and Finnish participants, Strandberg, Gooskens, and Schüppert (2021) found regional variation to be present in more formal speech styles but not in informal speech. When comparing height and fronting of the two vowels, the analysis indicated that Ostrobothnian participants produced the allophones in significantly different ways than speakers from the south (i.e., Helsinki and Southern Finland) when reading a list of word pairs and a passage of text, but not in samples from spontaneous speech.…”

“…Whereas Strandberg et al (2021) examined allophones [øː] and [oeː], which are both orthographically represented by <ö>, the current study explores the merger of /eː/ and /εː/, represented by the graphemes <e> and <ä>, respectively. Considering that the distinction between the phonemes is preserved in Swedish orthography, the differences of reading tasks versus spontaneous speech could potentially be larger when participants are faced with the written form.…”

The role of L1-L2 dissimilarity in L2 segment learning – Implications from the acquisition of English post-alveolar fricatives by Mandarin and Mandarin/Wu speakers

“…SLM has been successful in characterizing the learning of L2 sounds in various studies ( Flege et al, 1996 ; McAllister et al, 2002 ; Georgiou, 2021 ; Strandberg et al, 2021 ; Yang et al, 2022 ); nevertheless, exceptions are still occasionally noticed. Some studies demonstrate that degrees of L1-L2 dissimilarity do not always indicate how well the L2 segments are learned ( Wester et al, 2007 ; Iverson et al, 2008 ; Schmidt, 2018 ; Yang, 2019 ), and that assessing the perceived L1-L2 dissimilarity can be challenging and thus produce contradictory results ( Cebrian, 2006 ; Iverson and Evans, 2007 ).…”