Multilingual speakers’ L1, L2, and L3 fluency across languages: A study of Finnish, Swedish, and English

Abstract

The present study provides a multilingual perspective on speech fluency by examining fluency across first language (L1), second language (L2), and third language (L3) productions in Finnish, Swedish and English among L1 Finnish (Group 1, G1) and Finnish-Swedish bilingual (Group 2, G2) university students in Finland. The two research questions focused on differences in speech fluency across the three languages between the groups and correlations across speech fluency measures in the different languages. 90 speech samples in Finnish (L1), Swedish (L1/L3), and English (L2) from 30 participants were analyzed in the present study. The speech samples consist of short picture narrations based on comic strip prompts. The fluency analyses focused on temporal fluency and stalling mechanisms. The data were analyzed quantitatively with Mann-Whitney Utests and Spearman's rank-order correlation coefficients. The results demonstrated minor differences between the groups in their Finnish (L1) and English (L2) productions, but a higher level of fluency in Swedish for G2 (their L1) than G1 (their L3). For G1, the correlations were strongest between their L1 Finnish and L2 English, whereas for G2, mostly moderate to strong correlations were found between the different language pairs. The results suggest connections in fluency across the languages in a multilingual speaker's repertoire, but the strength of the correlations varies depending on the participants' L1(s) and proficiency level in the additional languages. Based on the findings, further research on fluency among multilingual speakers is needed. The results have implications for L2 teaching and assessment.

Listener perceptions of L1 speech in L2 and L3 fluency assessment contexts: a qualitative approach

“…While a consistent finding in several utterance fluency studies using correlation and regression analyses has been that there are connections between learners' L1 and L2 fluency features (e.g., De Jong et al 2015;Duran-Karaoz & Tavakoli 2020), perceived fluency studies have rarely addressed connections between L1 fluency and listeners' assessments of L2 fluency. Some utterance fluency studies also indicate that the connections between L1 and L2/L3 fluency measures are potentially influenced by cross-linguistic differences and learners' proficiency level in the target language (Huensch & Tracy-Ventura 2017;Peltonen 2018;Peltonen & Lintunen 2022). In one of the few studies examining L1 and L2 fluency links in an assessment context, Pinget et al (2014) examined connections between L2 Dutch fluency ratings and both traditional L2 fluency measures and L1-adjusted fluency measures (residuals obtained from correlating L1 and L2 fluency measures) for L1 Turkish and English (N=30) speakers.…”

“…The speech data used in the present study were collected as part of a project Fluency across Multilingual Speakers (MultiFluency; funded by the Swedish Cultural Foundation in Finland) and selected based on our previous research on these speakers' utterance fluency in Finnish, Swedish, and English (see Peltonen & Lintunen 2022). The speakers were 20 Finnish-speaking and 10 Finnish-Swedish bilingual university students of language subjects from two Finnish universities, most of whom studied English as either their major (n=21) or minor (n=5) subject (see Table 1 for a summary of the speakers' background information).…”

“…Over the last decade, connections between language learners' speech fluency in their first (L1) and target languages (L2, L3…) have received increased attention (e.g., De Jong et al 2015;Duran-Karaoz & Tavakoli 2020;Peltonen & Lintunen 2022;Gao & Sun 2024), and the accumulating results from these studies indicate that several L2 fluency features are affected by L1 fluency features. However, to our knowledge, examining how a speaker's L1 fluency affects their target language fluency has not been studied from the listener's perspective, even though it has been suggested that individual differences in L1 speech fluency should be recognized in L2 fluency assessment contexts (De Jong 2018;Gao & Sun 2024).…”

Fluency profiles in L2 monologue production: the role of underlying cognitive factors

“…Instead of group-level analyses of fluency and proficiency, a more individual analysis can reveal the mechanisms of how speaking styles and differences in general cognitive processing have an effect on L2 fluency within speakers; especially with more proficient language users, these impacts are more highlighted. This can be seen in, for example, the frequency and duration of pauses, which are influenced by L1 speaking style, especially for proficient language users (e.g., Derwing et al 2009; for fluency features in Finnish, see, e.g., Peltonen & Lintunen 2022;Toivola et al 2009). In addition, repairs seem to depend on many factors, including resources for monitoring for mistakes and choosing to engage in a repair, as in whether the language user prefers speed or accuracy of the speech (see, e.g., the Complexity-Accuracy-Fluency framework, Kuiken & Vedder 2022).…”