L2 Pronunciation Research Symposium

Home , Swedish orthography

6.-7.11.2019 Jyväskylä, Finland

Abstracts

Picture: University of Jyväskylä photo archives

Design: Valtteri Nyyssönen

Table of Contents

1. Programme ...... 1

2. Invited speakers ...... 4

Exploring the roots of perceptual problems in L2 speech learning ...... 5 Ocke-Schwen Bohn

SPEAKING (!) to the use of mixed methods in the study of language attitudes 7 Barbara Soukup

Is being comprehensible the same as being communicatively acceptable? Adoptions and adaptions in research on L2 pronunciation in L2s beyond English ...... 8 Lisa Tulaja

Active listening, imitation and pronunciation ...... 9 Elisabeth Zetterholm

L2 Speech Rhythm—Greetings from Melbourne ...... 10 Michael O’Dell

L1 speech attrition: opportunities, new areas and methods for future research 11 Ineke Mennen

3. Oral presentations ...... 13

L2 English pronunciation research: focus on fluency, beliefs and feedback ...... 14 Pekka Lintunen, Aleksi Mäkilähde & Pauliina Peltonen

Phonetic convergence to non-native speech: acoustic and perceptual evidence 15 Mónica A. Wagner, Mirjam Broersma, James M. McQueen & Kristin Lemhöfer

The effect of Swedish vowel-consonant quantity contrast on intelligibility of Swedish as a Second Language ...... 16 Mara Haslam & Elisabeth Zetterholm

When duration matters: the case of Estonian ...... 17 Einar Meister & Lya Meister

Researching L2 pronunciation learning in the project “Focus on learning pronunciation: Swedish as a L1/L2” ...... 19 Mikko Kuronen, Maria Kautonen, Elina Tergujeff, Riikka Ullakonoja, Hannele Dufva, Henna Heinonen & Johanna Moilanen

Project ”Accent perceptions in societal gatekeeping; ’Broken Finnish’” ...... 20 Mia Halonen & Sari Ohranen

Comprehensibility and accentedness of English spoken by Finnish-speaking and Swedish-speaking Finns ...... 21 Elina Tergujeff

4. Poster presentations ...... 22

Session 1

Swedish youths as listeners for diverse English accents: discussing the method ...... 23 Anna Elgemark & Hyeseung Jeong

The comparison of phonetic imitation and read-aloud of Finnish words produced by a Somalian: a pilot study ...... 25 Anne Leinonen

Effect of noise and accent on attitudes and intelligibility in EMI lectures ...... 26 Christian Jensen & Jacob Thøgersen

Analyzing L2 fluency with global wavelet spectrum ...... 28 Heini Kallio, Antti Suni, Štefan Benuš & Juraj Šimko

Children learning L2 pronunciation: Does frequent exposure to music facilitate learning through auditory training? ...... 29 Katja Immonen, Jemina Kilpeläinen, Maija S. Peltola & Paavo Alku

Prosodic marking of information structure by Russian L1 learners of English and native speaker perception of learners' intonation ...... 30 Maria Kunevich, Brechtje Post & Dora Alexopoulou

Session 2

Study proposal - Familiarity effects in L1 and L2 accent processing ...... 31 Anna Persson

Comprehensibility of L1 Finnish speakers’ L2 Swedish ...... 32 Henna Heinonen

Comprehensibility and intonation in L2 pronunciation - is there a link? ...... 33 Leena Maria Heikkola & Antti Saloranta

L2 pronunciation of Swedish on different proficiency levels ...... 34 Maria Kautonen

A single-case study of L2 Swedish word-final stop voicing pronunciations ...... 35 Maryann Su Lin Tan, Xin Xie & Florian Jaeger

When similar meets similar and new: Perception and production of French L3 obstruents in Scandinavian learners of French as a foreign language ...... 36 Pernille Berg Johnsson

Varieties of World Englishes: Nigerian Attitudes and Perceptions ...... 38 Temitayo Olatoye

Wednesday 6.11., Seminarium, S212 8:30-9:00 Registration 9:00-9:15 Foreword: Riikka Ullakonoja 9:15-10:00 Invited speaker: Ocke-Schwen Bohn Exploring the roots of perceptual problems in L2 speech learning 10:00-10:30 Pekka Lintunen, Aleksi Mäkilähde & Pauliina Peltonen L2 English pronunciation research: focus on fluency, beliefs and feedback 10:30-10:45 Elevator pitches for posters 10:45-11:45 Poster session 1, coffee break

Swedish youths as listeners for diverse English accents: discussing the method Anna Elgemark & Hyeseung Jeong

The comparison of phonetic imitation and read-aloud of Finnish words produced by a Somalian: a pilot study Anne Leinonen

Effect of noise and accent on attitudes and intelligibility in EMI lectures Christian Jensen & Jacob Thøgersen

Analyzing L2 fluency with global wavelet spectrum Heini Kallio, Antti Suni, Štefan Benuš & Juraj Šimko

Children learning L2 pronunciation: Does frequent exposure to music facilitate learning through auditory training? Katja Immonen, Jemina Kilpeläinen, Maija S. Peltola & Paavo Alku

Prosodic marking of information structure by Russian L1 learners of English and native speaker perception of learners' intonation Maria Kunevich, Brechtje Post & Dora Alexopoulou

11:45-12:30 Invited speaker: Barbara Soukup SPEAKING (!) to the use of mixed methods in the study of language attitudes

12:30-13:00 Mónica A. Wagner, Mirjam Broersma, James M. McQueen & Kristin Lemhöfer Phonetic convergence to non-native speech: acoustic and perceptual evidence

13:00-14:00 Lunch (not included in the registration fee) 14:00-14:45 Invited speaker: Lisa Tulaja Is being comprehensible the same as being communicatively acceptable? Adoptions and adaptions in research on L2 pronunciation in L2s beyond English 14:45-15:15 Mara Haslam & Elisabeth Zetterholm The effect of Swedish vowel-consonant quantity contrast on intelligibility of Swedish as a Second Language 15:15-15:45 Einar Meister & Lya Meister When duration matters: the case of Estonian 15:45-16:15 Coffee break 16:15-16:45 Mikko Kuronen, Maria Kautonen, Elina Tergujeff, Riikka Ullakonoja, Hannele Dufva, Henna Heinonen & Johanna Moilanen Researching L2 pronunciation learning in the project “Focus on learning pronunciation: Swedish as a L1/L2” 16:45-17:30 Mia Halonen & Sari Ohranen Project ”Accent perceptions in societal gatekeeping; ’Broken Finnish’” 19:00- Dinner (Restaurant Harmooni, Väinonkatu 1)

Thursday 7.11., Seminarium, S212 9:00-9:45 Invited speaker: Elisabeth Zetterholm Active listening, imitation and pronunciation 9:45-10:15 Invited speaker: Michael O’Dell L2 Speech Rhythm—Greetings from Melbourne 10:15-10:30 Elevator pitches for posters 10:30-11:30 Poster session 2, coffee break

Study proposal - Familiarity effects in L1 and L2 accent processing Anna Persson

Comprehensibility of L1 Finnish speakers’ L2 Swedish Henna Heinonen

Comprehensibility and intonation in L2 pronunciation - is there a link? Leena Maria Heikkola & Antti Saloranta

L2 pronunciation of Swedish on different proficiency levels Maria Kautonen

A single-case study of L2 Swedish word-final stop voicing pronunciations Maryann Su Lin Tan, Xin Xie & Florian Jaeger

When similar meets similar and new: Perception and production of French L3 obstruents in Scandinavian learners of French as a foreign language Pernille Berg Johnsson

Varieties of World Englishes: Nigerian Attitudes and Perceptions Temitayo Olatoye

11:30-12:00 Elina Tergujeff Comprehensibility and accentedness of English spoken by Finnish- speaking and Swedish-speaking Finns 12:00-12:45 Invited speaker: Ineke Mennen L1 speech attrition: opportunities, new areas and methods for future research 12:45-13:00 Afterword: Riikka Ullakonoja 13:00-14:00 Lunch (not included in the registration fee) 14:00-15:30 Research meeting & networking 15:30-15:50 Coffee break 15:50-17:30 Planning a Nordic research network on L2 pronunciation

Invited speakers

Exploring the roots of perceptual problems in L2 speech learning

Ocke-Schwen Bohn Aarhus University, Denmark

Nonnative speakers typically produce their L2 with a foreign accent. Most, if not all researchers agree that foreign accented speech has a perceptual basis: As posited by current models of L2 speech learning, nonnative speakers are biased perceivers whose perception of L2 sounds differs systematically from the ways in which native speakers of the L2 perceive these sounds. The basic assumption of both the Speech Learning Model (Flege 1995) and the Perceptual Assimilation Model (Best & Tyler 2007) is that L2 speech learning problems can be predicted from how L2 learners map the sounds of the L2 to their native categories. That is, these models depend on reliable and valid methods of establishing the perceived similarity of speech sounds to account for the perception and production problems of nonnative speakers.

This presentation reports on several experiments which examined how well four different types of similarity predict nonnative speech perception. The different types are ecphoric and perceptuali cross-language similarity, perceived within-language similarity, and acoustic similarity. One set of experiments examined how well cross-language perceptual assimilation (English to Danish) and within-language similarity ratings (English-English) of English consonants by Danish listeners predict Danes’ identification of English consonants. Another set of experiments explored the roots of English listeners’ discrimination problems for the closely spaced Danish unrounded front vowels by relating these problems to two types of perceived cross-language similarity (ecphoric and perceptual) and to acoustic similarity. Results suggest that each of the four types of similarity accounts for some of the perceptual problems, but none does so exhaustively, probably because nonnative speech perception is affected by additional factors including universal perceptual biases which can’t be attributed to the nonnative speaker’s L1 (see Bohn 1995 and Polka & Bohn 2011).

[Research supported by grants from the Carlsberg Foundation and Inge Lehmanns Legat af 1983]

References

Best, C. T., & Tyler, M. D. (2007). Nonnative and second-language speech perception: Commonalities and complementarities. In: Bohn, O.-S. & Munro, M. J. (eds.), Language experience in second language speech learning: In honor of James Emil Flege, 31–52.

Bohn, O.-S. (1995). Cross-language speech perception in adults: First language transfer doesn’t tell it all. In Strange, W. (ed.), Speech perception and linguistic experience: Issues in cross-language research, 279–304.

Flege, J. E. (1995). Second language speech learning: Theory, findings, and problems. In Strange, W. (ed.), Speech perception and linguistic experience: Issues in cross-language research, 233– 277.

Polka, L., & Bohn, O.-S. (2011). Natural Referent Vowel (NRV) framework: An emerging view of early phonetic development. Journal of Phonetics, 39(4), 467–478.

Tulving, E. (1981). Similarity relations in recognition. Journal of Verbal Learning and Verbal Behavior, 20(5), 479–496.

______i According to Tulving (1981, 488), Ecphoric similarity refers to “similarity between a test item and the stored relevant episodic information”, whereas perceptual similarity refers to “the similarity between test items in a given set, such as a test pair”.

SPEAKING (!) to the use of mixed methods in the study of language attitudes

Barbara Soukup Austrian Academy of Sciences, Austria

In this contribution, I theorize and demonstrate a mixed-methods approach to the study of language attitudes that integrates quantitative methods of attitude elicitation (notably, speaker evaluations using the matched-guise technique) with qualitative methods of interactional discourse analysis of conversational phenomena.

A mixed-methods approach to language attitude research has two main implications/benefits: (1) It necessitates and thus mandates an integrative conceptualization of ‘language attitudes’ that accounts for their properties as both emergent, interactive processes as well as cognitively solidified entities of discourse; and (2) it enhances the validity of claims based on experimental findings, as well as their practical applicability to the explication of actual human behavior contingent on language attitudes (like strategic language choice or sociolinguistic gatekeeping).

On the first point, I argue that (and explicate how) Ron Scollon’s (2003) notion of ‘human epistemological constructs’ is a particularly useful tool for the conceptual integration of mixed methodologies in the investigation of language attitudes. Regarding concrete study design in view of the real-life applicability of findings, I suggest that Dell Hymes’s (1972) heuristic known as the ‘SPEAKING grid’, which helps to capture and describe configurations of particular speech events, is a practical and ethnographically sound foundation on which to build mixed-methods language attitude research.

I illustrate and discuss my arguments and their implications in a case study of strategic style- shifting between standard and dialect in conversational data from an Austrian TV discussion show (see Soukup 2009, 2015).

References

Hymes, Dell. (1972). Models of the interaction of language and social life. In: John J. Gumperz & Dell Hymes (eds.), Directions in Sociolinguistics: The Ethnography of Communication, 35–71. New York: Holt, Rinehart and Winston.

Scollon, Ron. (2003). The dialogist in a positivist world: Theory in the social sciences and the humanities at the end of the twentieth century. Social Semiotics, 13(1), 71–88.

Soukup, Barbara. (2009). Dialect Use as Interaction Strategy: A Sociolinguistic Study of Contextualization, Speech Perception, and Language Attitudes in Austria. Vienna: Braumüller.

Soukup, Barbara. (2015). Mixing methods in the study of language attitudes: Theory and application. In: Alexei Prikhodkine & Dennis R. Preston (eds.), Responses to Language Varieties: Variability, Processes and Outcomes, 55–84. Amsterdam: John Benjamins.

Is being comprehensible the same as being communicatively acceptable? Adoptions and adaptions in research on L2 pronunciation in L2s beyond English

Lisa Tulaja Kiel University, Germany

A central aim in L2 pronunciation research is finding answers to questions related to practical issues that are useful to L2 teachers and learners (Levis 2016). While there has been growing interest for the field, current studies mostly address English as a second language. However, the value of adapting research designs for research on languages beyond ESL should be taken into consideration.

Regarding social ratings of L2 accented speech, Munro & Derwing’s (1995) concepts of comprehensibility and accentedness have had a huge impact. However, when it comes to Danish as an L2, the notion of comprehensibility does not go far enough. As Danish L1 speakers exhibit a relatively low degree of tolerance for variation as compared to the standard variety (Kristiansen 2003) and are generally quite competent in English, Danes tend to switch to the lingua franca as soon as they perceive an L2 accent – even if the L2 speech is highly comprehensible. In order to make recommendations for explicit teaching contexts (Saito 2011), it would be advantageous to know which features of L2 accent trigger this behaviour. In my research, I extended Munro & Derwing’s (1995) approach by adding an extra dimension labelled acceptability – a construct that is considered to capture another aspect of social behaviour not covered by comprehensibility. However, in contexts of ESL this dimension has received little attention so far (Thomson 2018).

In my presentation I will discuss the implementation and adaption of the dimension acceptability as an example of how a practical problem, relevant to a specific language pair, can lead to asking new questions and creating a research design that benefits from the bigger field of ESL if matched with the specific conditions. I will argue that, in turn, the field of ESL can also profit from adaptations developed from a context with a specifically Nordic focus.

References

Kristiansen, Tore. (2003). Language attitudes and language politics in Denmark. International Journal of the Sociology of Language 159, 57–71.

Levis, John. (2016). The interaction of research and pedagogy. Journal of Second Language Pronunciation, 2(1), 1–7.

Munro, Murray J. & Tracey M. Derwing. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 45(1), 73–97.

Saito, Kazuya. (2011). Examining the role of explicit phonetic instruction in native-like and comprehensible pronunciation development: an instructed SLA approach to L2 phonology. Language Awareness, 20(1), 45–59.

Thomson, Ron. (2018). Measurement of accentedness, intelligibility, and comprehensibility. In: Kang, Okim & April Ginther (eds.), Assessment in second language pronunciation, 11–29. London/New York: Routledge.

Active listening, imitation and pronunciation

Elisabeth Zetterholm Linköping University, Sweden

Perception training and corrective feedback with support of form-focused instructions seem to have a positive effect on second language learners’ metalinguistic awareness and pronunciation (e.g. Lee & Lyster 2016). Two different studies with a focus on learning Swedish as a second language using active listening and imitation will be presented. Teaching based on active listening when reading a text is shown as a successful method for improved pronunciation with increased intelligibility even after five weeks of training. Furthermore, participants mention a better understanding of spoken Swedish and improved self-esteem in speaking Swedish (Zetterholm, Emgård, Vahlén 2019). Learners were guided to focus on specific significant features in Swedish prosody such as stressed words and syllables as well as prolonged speech sounds. Another study shows the importance of segment. Learners of Swedish should focus on the pronunciation of final consonants. They were instructed to listen to recordings of a native Swedish speaker and imitate the pronunciation. Training sentences were constructed using words ending with a consonant some of which were linked together with the next word starting with a vowel. Pronunciation improvement was established after six weeks of training (Zetterholm 2014). Another study focusing on reading a text, listening, auditory training and imitation in a language unknown for the learner indicate that mimicking an L1 speaker could be effective especially when learning the prosodic features. However, a transparent orthography is not easier to pronounce for a novice L2 learner than an L2 with an opaque orthography (Kuronen, Zetterholm & Nyyssönen, under review). Prosodic features can be seen as the larger, more general characteristics of a language, which is why they might be easier to mimic than segmental features. To summarize, prosodic features may be more salient for listeners and therefore more important for an intelligible pronunciation, although segments cannot be neglected.

References

Kuronen, M., Zetterholm, E. & Nyyssönen, V. (under review). The Impact of Orthography on Pronunciation in the First Stages of L2 Learning.

Lee, A.H. & Lysteer, R. (2016). The effects of corrective feedback on instructed L2 speech perception. Studies in Second Language Acquisition, 38(1), 35–64.

Zetterholm, E. (2014). Final stops or not? The importance of final consonants for an intelligible accent. In: Proceedings of the 4th Pronunciation in Second Language Learning and Teaching Conference, Iowa University, USA. Sept 20–21, 2013. 22–29. Zetterholm, E., Emgård, H. & Vahlén, B. (2019). Pronunciation teaching based on active listening. Abstract and presentation at PSLLT 2019.

L2 Speech Rhythm—Greetings from Melbourne

Michael L. O’Dell University of Tampere, Finland

More than 100 papers at the International Congress of Phonetic Sciences (ICPhS) in Melbourne in August 2019 dealt with various aspects of the phonetics of second and foreign language. A general overview of these papers will be presented, looking especially in more detail at one aspect which has not yet received adequate attention: speech rhythm, tempo and timing from the L2 perspective. The importance of rhythm, tempo and timing for speech in general and for second language and L2 pronunciation research in particular will be discussed on the basis of the ICPhS papers and other research.

L1 speech attrition: opportunities, new areas and methods for future research

Ineke Mennen University of Graz, Austria

While it is widely acknowledged that an L2 accent will almost inevitably retain traces of a speaker’s L1 (see Edwards & Zampini, 2008), there is growing evidence that the reverse is also possible. That is, the L2 learning experience can also affect pronunciation in the L1 (Flege, 1987; Mennen, 2004; Dmitrieva et al., 2010; Mayr et al., 2012; de Leeuw et al., 2012, 2013), even to the extent that individuals may sound foreign in their native language (de Leeuw et al., 2010; Hopp & Schmid, 2013; Schmid & Hopp, 2014; Bergmann et al., 2016). This decline from an individual’s previous L1 linguistic abilities in healthy individuals immersed in an environment where the L2 is dominant is commonly referred to as L1 attrition (Köpke & Schmid, 2004; Schmid, 2010).

The field of L2 speech learning has gained a fairly good understanding of the factors that contribute to the degree of foreign accent, which aspects of pronunciation are most likely to deviate from the L2 norm, and the relative contribution of these deviations to foreign accent perception. However, much less is known about L1 speech attrition. For instance, it is unclear how common L1 phonetic attrition is; whether some areas of pronunciation – and if so, which – are more susceptible to attrition than others; which deviances from L1 norms lead to the perception of foreign accent; at which stage L1 attrition may set in; and which predictor variables govern attrition. This talk will give an overview of the current state of research on L1 speech attrition, highlighting those areas that need further research attention. I will illustrate this with some preliminary research findings from a longitudinal study of Arnold Schwarzenegger’s speech.

References

Bergmann, C., Nota, A., Sprenger, S., & Schmid, M. S. (2016). L2 immersion causes non- native-like L1 pronunciation in German attriters. JPhon, 58, 71–86. de Leeuw, E., Mennen, I., & Scobbie, J. M. (2013). Dynamic systems, maturational constraints, and L1 phonetic attrition. Int J of Biling, 17, 683–700. de Leeuw, E., Mennen, I., & Scobbie, J. M. (2012). Singing a different tune in your native language: first language attrition of prosody. Int J of Biling, 16(1), 101–16. de Leeuw, E., Schmid, M. S., & Mennen, I. (2010). The effects of contact on native language pronunciation in an L2 migrant setting. Biling-Lang Cogn, 13(1), 33–40.

Dmitrieva, O., Jongman, A., & Sereno, J. (2010). Phonological neutralization by native and non-native speakers: The case of Russian final devoicing. JPhon, 38, 483–9.

Edwards, J. G.; & Zampini, M. L. (Eds.) (2008). Phonology and second language acquisition. Philadelphia, PA: John Benjamins.

Flege, J. E. (1987). The production of 'new' and 'similar' phones in a foreign language: evidence for the effect of equivalence classification. JPhon, 15(1), 47–65.

Hopp, H., & Schmid, M. S. (2013). Perceived foreign accent in first language attrition and second language acquisition: The impact of age of acquisition and bilingualism. Appl Psycholing, 34(2), 361–94.

Köpke, B., & Schmid, M. S. (2004). First language attrition: The next phase. In: Monika S. Schmid, Barbara Köpke, B., Merel Keijzer, M. & Lina Weilemar (Eds.), First language attrition: Interdisciplinary perspectives on methodological issues. Amsterdam: John Benjamins, pp. 1–43.

Mayr, R., Price, S., & Mennen, I. (2012). First language attrition in the speech of Dutch– English bilinguals: The case of monozygotic twin sisters. Biling-Lang Cogn, 15(4), 687–700.

Mennen, I. (2004). Bi-directional interference in the intonation of Dutch speakers of Greek. JPhon, 32(4), 543–63.

Schmid, M. S. (2010). Languages at play: The relevance of L1 attrition to the study of bilingualism. Biling-Lang Cogn, 13(1), 1–7.

Schmid, M. S., & Hopp, H. (2014). Comparing foreign accent in L1 attrition and L2 acquisition: Range and rater effects. Lang Testing, 31(3), 367–88.

Presentations

L2 English pronunciation research: focus on fluency, beliefs and feedback

Pekka Lintunen, Aleksi Mäkilähde & Pauliina Peltonen University of Turku, Finland

The English Department at the University of Turku has a long tradition in L2 pronunciation research. The focus of research has been on the teaching and learning of pronunciation. In this presentation, our aim is to review the English Department’s most recent projects on L2 pronunciation and illustrate what their main contributions are to L2 pronunciation research nationally and internationally.

This presentation will focus on three areas: 1) fluency in L2 speech and how fluency and pronunciation are linked, 2) language learners’ beliefs about pronunciation learning, 3) learning to give constructive feedback on L2 pronunciation. Firstly, fluency is a multifaceted concept that is often mentioned in assessment grids and learning objectives (e.g. Lintunen et al. in press). Many fluency measures are temporal in nature, and fluency analysis can be useful for researchers interested in pronunciation or second language acquisition more broadly. Secondly, language learners’ beliefs affect learning outcomes. We have investigated, for instance, learners’ awareness of their pronunciation problems, pronunciation learning strategies and learners’ beliefs concerning their own pronunciation priorities and development (e.g. Lintunen & Mäkilähde 2018). Lastly, we have recently focused on pronunciation feedback and investigated how peer feedback is provided and structured (e.g. Lintunen et al. 2017). Our studies suggest that giving feedback on learners’ pronunciation skills should be part of L2 teacher training.

References

Lintunen, P., Mutta, M. & Peltonen, P. (in press). Fluency in L2 Learning and Use. Bristol: Multilingual Matters.

Lintunen, P. & Mäkilähde A. (2018). Short- and long-term effects of pronunciation teaching. In Jan Volín & Radek Skarnitzl (eds). Pronunciation of English by Speakers of Other Languages. Newcastle Upon Tyne: Cambridge Scholars Publishing, 46–72.

Lintunen, P., Mäkilähde, A. & Peltonen, P. (2017). Learner perspectives on pronunciation feedback. In Proceedings of the Phonetics Teaching and Learning Conference. University College London: London, 67–71.

Phonetic convergence to non-native speech: acoustic and perceptual evidence

Mónica A. Wagner, Mirjam Broersma, James M. McQueen, Kristin Lemhöfer Radboud University, the Netherlands

This study aims to assess phonetic convergence to a non-native speaker using a novel paradigm and an approach combining both acoustic and perceptual measures. Relatively few studies have evaluated whether people converge to non-native speech and the findings so far have been inconsistent. This variability may be due to differences in measurement techniques and experimental paradigms, which may induce more or less social engagement. We hope to contribute to this literature by evaluating phonetic convergence to a non-native accent with vowels and consonants both acoustically, along multiple dimensions, and perceptually with judgments from raters. To this end, we developed a novel experimental paradigm in which a shadowing task is disguised as a memory task in order to minimize awareness of the study’s goal.

We tested 93 native speakers of Dutch who listened to and repeated pre-recorded audios of a native speaker of Croatian proficient in Dutch. Critical items were chosen that contained target sounds that deviated from the native Dutch targets spectrally and/or in terms of duration. Before the task began, baseline measurements were acquired by having participants read the words out loud.

So far, the vowels have been analyzed. Euclidean distances between participants’ and the model speaker’s normalized critical vowels were calculated. Convergence in duration was measured by calculating the absolute difference between the model’s and participants’ normalized durations. Comparison of the baseline and shadow distances revealed convergence in duration but not in vowel spectra. AXB perceptual data from 16 native Dutch raters is also being collected in order to compare patterns of perceived convergence with the acoustic data.

Our preliminary findings are already consistent with the varied results that characterize the literature on speech accommodation. In a next step, individual differences in convergence will be related to L2 pronunciation data which has also been collected for these participants.

The effect of Swedish vowel-consonant quantity contrast on intelligibility of Swedish as a Second Language

Mara Haslam1, Elisabeth Zetterholm2 1Stockholm University, Sweden; 2Linköping University, Sweden

Sweden has received a great number of immigrants in recent years, and these learners need to be able to speak Swedish intelligibly in order to participate in the Swedish society and workforce. Teachers of Swedish as a Second Language may wonder which of all the features of Swedish phonology they should focus on. While some research is available, more information on prosodic features is needed. This project attempts to partially address that issue by testing the vowel-consonant length contrast in stressed syllables of Swedish. When a stressed syllable contains a vowel followed by a consonant, the lengths of the vowel and consonant must contrast; that is, a long vowel must be followed by a short consonant or a short vowel must be followed by a long consonant. This is often encoded in Swedish orthography, such that V:C syllables are represented with one consonant letter, such as the word fin [fi:n] “nice”, while VC: syllables are represented with two consonant letters, such as the word finn [fin:] “find”. We hypothesized that this length contrast is necessary for intelligibility for native Swedish listeners. 25 non-native speakers of Swedish participated in perception and production activities targeted to this contrast. Their productions were identified by native- speaker judges and compared to acoustic measurements. Perception data show that learners performed significantly slightly better than chance at identification of the contrast in focus (59.8% correct) but significantly worse than control native speakers (99% correct). Logistic regression of preliminary production data show that the hypothesis that consonant-vowel length contrast is necessary for intelligibility holds true for one-syllable words but not for two- syllable words with the target contrast in the first syllable. Detailed results will be presented and discussed.

When duration matters: the case of Estonian

Einar Meister1, 2 & Lya Meister1 1Tallinn University of Technology, Estonia; 2University of Eastern Finland

Languages vary in terms of how different acoustic cues are exploited in the production of phonological contrasts. In quantity languages, e.g. Finnish and Estonian, a difference in the duration of a vowel or a consonant can result in a change of grammatical function or lexical meaning of the word, while in non-quantity languages, e.g. Russian and Spanish, duration- based phonological oppositions are lacking.

Our studies on the acquisition of the Estonian ternary quantity contrasts (referred to as Q1, Q2 and Q3) by subjects with Russian, Finnish, Japanese and Latvian (Meister & Meister, 2011, 2012, 2013a, 2013b, 2014a, 2014b, 2017; Meister, Nemoto & Meister, 2015) language backgrounds have shown that all L2 groups were successful in producing the Q1-Q2 opposition but failed to distinguish the Q2-Q3 vocalic contrasts (not revealed in the orthography). It was found that Q2 and Q3 temporal patterns by Finnish and Latvian L2 subjects were close to the native Estonian Q3 pattern while L2 subjects with Russian and Japanese language background produce durational patterns close to Estonian Q2. However, all L2 groups were able to distinguish Q2-Q3 consonantal contrasts, especially in the case of intervocalic plosives where the contrast is revealed in the orthography.

Over the years, L2 research has given much time and attention to the acquisition of the correct pronunciation of L2 consonants and vowels mostly concentrating on well-researched languages (e.g. English, French, Spanish), and little attention has been devoted to the acquisition of languages that employ durational cues in making phonological contrasts. The results on the L2 production of Estonian quantity oppositions provide clear evidence on the limitations of the current L2 theoretical models and point to the need to devote more attention to the understudied acquisition of L2 duration-based contrasts. In the talk we will envisage some preliminary insights to address the role of duration in the L2 theoretical models.

References

Meister, Lya; Meister, Einar (2011). Perception of the short vs. long phonological category in Estonian by native and non-native listeners. Journal of Phonetics, 39(2), 212−224.

Meister, Lya; Meister, Einar (2012). The production and perception of Estonian quantity degrees by native and non-native speakers. Interspeech 2012: 13th Annual Conference of the International Speech Communication Association, 886−889.

Meister, Einar; Meister, Lya (2013a). Production of Estonian quantity contrasts by native speakers of Finnish. Interspeech 2013: 14th Annual Conference of the International Speech Communication Association, 330−334.

Meister, Einar; Meister, Lya (2013b). Native and non-native production of Estonian quantity degrees: comparison of Estonian, Finnish and Russian subjects. Nordic Prosody: Proceedings of the XIth Conference. Frankfurt am Main: Peter Lang Verlag, 235−243.

Meister, Einar; Meister, Lya (2014a). L2 production of Estonian quantity degrees. Speech Prosody 7: Proceedings of the 7th international conference. 929−933.

Meister, Einar; Meister, Lya (2014b). Estonian quantity degrees produced by Latvian subjects. Linguistica Lettica, 22, 85−106.

Meister, Einar; Meister, Lya (2017). Production of Estonian consonantal quantity contrasts by Latvian subjects. Linguistica Lettica, 25, 88−108.

Meister, Einar; Nemoto, Rena; Meister, Lya (2015). Production of Estonian quantity contrasts by Japanese speakers. Journal of Estonian and Finno-Ugric Linguistics // Eesti ja soome-ugri keeleteaduse ajakiri, 6(3), 79−96.

Researching L2 pronunciation learning in the project “Focus on learning pronunciation: Swedish as L1/L2”

Mikko Kuronen, Maria Kautonen, Elina Tergujeff, Riikka Ullakonoja, Hannele Dufva, Henna Heinonen & Johanna Moilanen University of Jyväskylä, Finland

In this presentation, the research project Focus on learning pronunciation: Swedish as L1/L2 and some of its main results are introduced. The main aim of the project has been to (i) study the phonetic aspects of L2 Swedish with speakers of Finnish, Russian and English as L1 and (ii) studying aspects of Swedish L1 speakers’ production of these languages as their L2. The project focuses on the Finland-Swedish variety, but also Swedish as spoken in Sweden is studied.

L2 learners in the chosen language pairs have been studied on different stages of the learning process with both acoustic and auditory methods. In the presentation, we will give an overview of the results concerning specifically learning of Swedish as L2. In addition to the theoretical aim of gaining new knowledge on L2 pronunciation learning, the research findings of the project contribute to language teaching practices and the design of teaching materials.

The project has been funded by The Society of Swedish Literature in Finland (SLS) 2015– 2019. https://www.jyu.fi/fokus/presentations-and-publications

Project “Accent perceptions in societal gatekeeping; ‘Broken Finnish’”

Mia Halonen & Sari Ohranen University of Jyväskylä, Finland

‘Broken Finnish’ focuses on pronunciation and ‘accent’ perceptions in the National Certificates of Language Proficiency test in Finland. It explores if and how these perceptions affect the ratings of speech proficiency and studies where the perceptions might arise from. As many countries - Finland included - have language requirements for citizenship, language testing is a possible source of social inequality. The main aim of the project is to ensure the fairness of the test system.

To explore whether there was any bias towards test takers, we decided to focus on test takers from L1 groups that we already knew, through previous research, face negative stereotyping in Finland: Arabic, Estonian, Russian, Thai, and Finland Swedish, the hypothesis being that these speakers, if any, might face biased assessments. The data were collected through an online platform in 2015 and 2016. Altogether 49 examinees’ samples of their speaking performance were rated by 44 certified raters. The raters rated the performances both holistically and separately for the six analytical criteria for speaking: fluency, flexibility, coherence/cohesion, range and accuracy of vocabulary, pronunciation/phonological control and grammatical accuracy. The raters were also asked to identify the speaker’s L1 and to indicate how certain that identification was. Multifaceted Rasch analyses were conducted to discover any rater biases toward particular L1 groups or interactions between criteria and learner groups and whether the raters’ identification of the speaker’s L1 affected assessments.

In the symposium we will present and discuss some of our results. Our results confirm the hypothesis: recognition of L1 affects the assessment, and, in most cases, negatively. Interestingly, the criterion most affected by L1 identification was pronunciation. In addition to discussing some preliminary explanations for this phenomenon, we will discuss the relationship between the criteria: how they behave in relation to each other and to the holistic assessment. https://www.jyu.fi/hytk/fi/laitokset/solki/broken-finnish/in-english

Comprehensibility and accentedness of English spoken by Finnish-speaking and Swedish-speaking Finns

Elina Tergujeff University of Jyväskylä, Finland

Research on spoken English by Finns has so far gained information about typically difficult features (through contractive and practical studies, e.g. Lintunen 2004), speakers’ fluency resources (Peltonen 2017) and how pronunciation is taught in schools (Tergujeff 2013). Little is known about how Finns’ English is perceived by listeners, and overall the studies have paid little attention to L1 Swedish-speaking Finns’ English. This study aims to explore how comprehensible (i.e. how easy or difficult to understand) and foreign accented the English of Finnish-speaking and Swedish-speaking Finns is, as perceived by L1 English listeners and the speakers’ peers.

Speech materials consisted of 20-second samples extracted from language assessment data with Finnish-speaking (n=20) and Swedish-speaking (n=20) teenaged Finns as speakers. Half of them were A2-level and the other half B1-level speakers of English on the CEFR scale. L1 English-speaking (n=34) teenagers rated the speech samples for comprehensibility and foreign accentedness. Similarly, L1 Finnish-speaking (n=31) and L1 Swedish-speaking (n=30) teenagers rated their peers’ speech samples. The ratings were done on a 9-point scale.

Results reveal that L1 English-speaking listeners rate A2-level speakers from the two language groups very similarly. However, B1-level Swedish-speaking Finns are rated significantly more comprehensible and less accented than Finnish-speaking Finns. Hence, there seems to be a difference in the English of these two L1 groups, affecting comprehensibility and accentedness ratings but not the overall proficiency assessment. In addition, the two L1 groups act very differently when rating their peers’ English: L1 Swedish speakers rate their peers more comprehensible and less accented than what the L1 English-speaking listeners do, while L1 Finnish speakers rate their peers equally comprehensible and more accented when compared to ratings given by L1 English listeners.

This study was funded by the Swedish Cultural Foundation in Finland (2016–2017) and the Academy of Finland (2018–2021), together with the University of Jyväskylä.

References

Lintunen, P. (2004). Pronunciation and phonemic transcription: A study of advanced Finnish learners of English. Turku: University of Turku.

Peltonen, P. (2017). Temporal fluency and problem-solving in interaction: An exploratory study of fluency resources in L2 dialogue. System, 70, 1–13.

Tergujeff, E. (2013). English pronunciation teaching in Finland. Jyväskylä: University of Jyväskylä.

Posters

Swedish youths as listeners for diverse English accents: discussing the method

Anna Elgemark, Hyeseung Jeong University West, Sweden

The Swedish youth have received recognition for having high proficiency in English and for being highly motivated in learning English (Norrby, 2015). However, their accent preference and listening input are skewed towards certain Inner Circle Englishes, especially American English (AE) (Cabau, 2009), while it is not known if they are willing and/or capable of understanding a wide range of accents in globalised communication. Our study, to be commenced in early 2020, seeks to address this gap by investigating how Swedish youths are as listeners for diverse English accents. In this poster presentation, we will discuss the method of our study.

The method is experimental. We operationalise listener intelligibility (LI), listener comprehensibility (LC), accentedness perception (AP) and accentedness acceptance (AA), adapted from previous studies (e.g. Kennedy & Trofimovich, 2008; Munro & Derwing, 1995; Tulaja, 2019), as well as listener variables found to have influence on the listener perception (e.g. Kang & Rubin, 2014). The stimulus will be the recordings of five university lecturers (one AE L1 and four L2 speakers) reading the same texts – adapted from previous Swedish national English 6 tests and the 40 true/false sentences from Munro and Derwing (1995). Listeners will be five classes of high school students, and each class will listen to one of the five speakers’ readings. We will administer tests to determine LI, LC, AP, and AA, and a questionnaire to identify individual variables. The analysis will target the LI, LC, AP and AA for different speakers, their relationships, interaction between the effects of speakers and listener variables, and speakers’ phonetic features compromising LI. At this stage of the research process, we welcome feedback, particularly on the justification of speaker selection, instruments for measuring the four constructs and for identifying individual variables and ways to remove training effect in test administration.

References

Cabau, B. (2009). The irresistible rise and hegemony of a linguistic fortress: English teaching in Sweden. International Multilingual Research Journal, 3(2), 134–152. doi:10.1080/19313150903073786

Kang, O., & Rubin, D. (2014). Listener expectations, reverse linguistic stereotyping, and individual background factors in social judgments and oral performance assessment. In A. Moyer & J. M. Levis (Eds.), Social dynamics in second language accent account (pp. 239–253). Berlin: De Gruyter Mouton. Kennedy, S., & Trofimovich, P. (2008). Intelligibility, comprehensibility, and accentedness of L2 speech: The role of listener experience and semantic context. Canadian Modern Language Review, 64(3), 459–489.

Munro, M., & Derwing, T. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 45(1), 73–97. doi:10.1111/j.1467- 1770.1995.tb00963.x

Norrby, C. (2015). English in Scandinavia: Monster or mate? Sweden as a case study. In J. Hajek & Y. Slaughter (Eds.), Challenging the monolingual mindset (pp. 17–32). Bristol: Multilingual Matters.

Tulaja, L. (2019). Pronunciation errors by German L2 Danish learners: Ratings in accentedness, comprehensibility and acceptability. Paper presented at the The New Sounds, Waseda, Japan.

The comparison of phonetic imitation and read-aloud of Finnish words produced by a Somalian: a pilot study

Anne Leinonen University of Jyväskylä, Finland

Phonetic imitation often takes place in language acquisition either as a part of communication or formal teaching. The impact of imitation in learning L2 pronunciation is still unclear. Previous findings (Zajac 2013; Ballier & Burin 2017) indicate that L2-learners tend to accommodate for vowel duration and speech rate of a L2-model.

This presentation reports the results of a pilot study about pronunciation of Finnish words joustava, tunnollinen, reipas, päättäväinen and nokkela produced by a Somalian female (Finnish as a L2). Each of the words contain segments that Somali speakers often find difficult in Finnish (see more Aho, Toivola, Karlsson & Lennes 2016). The focus of this study is on the differences between sounds, word stress and speech rate in read-aloud and imitated speech.

This study tested whether the informant’s production of the words was influenced by imitation in the level of phonetic outcome. The study consisted of three separate tasks: a spelling, a read-aloud and an imitation tests of a native model. In a spelling test, the listener heard twice 12 Finnish words and was asked to write down what she heard. In a read-aloud test she was instructed to read the same words aloud, and in an imitation task she repeated the same words after a native model.

This study compares the production of read-aloud and imitated words in the levels of orthography, acoustics and listening. The results of the speaker’s imitation and read-aloud speech were compared with the L2- model. The read-aloud and imitated words were analyzed acoustically by Praat. The words were also transcripted by a phonetician. This pilot study shows positive impact of imitation, especially in speech rate. It also demonstrates that imitation affects the production of single sounds only slightly.

References

Aho, E., Toivola, M., Karlsson, F. & Lennes, M. (2016). Aikuisten maahanmuuttajien suomen ääntämisestä. Puhe ja kieli, 36(2), 77–96.

Burin, L. & Ballier, N. (2017). Accommodation of L2 speech in a repetition task. EPIP5, May 2017, Caen, France.

Zajac M. (2013). Phonetic imitation of vowel duration in L2 speech. Research in Language, 11(1). DOI 10.2478/v10015-012-0009-5.

Effect of noise and accent on attitudes and intelligibility in EMI lectures

Christian Jensen & Jacob Thøgersen University of Copenhagen, Denmark

In several studies on English-Medium Instruction (EMI) in the Nordic countries, students comment that they often find it difficult to understand the lecturers’ accents (Bolton & Kuteeva 2012; Hellekjær 2010). It has also been demonstrated that (perceptions of) the lecturers’ English skills, including their pronunciation, affect students’ perceptions of the lecturer’s general lecturing and disciplinary competence (Jensen et al. 2013). Derwing & Munro (1995), however, showed that foreign-accented speech is often fully intelligible, even if it is subjectively felt to be difficult to understand (low comprehensibility, in their terms).

In a previous study (Jensen & Thøgersen 2017), we found that accented speech which yielded low comprehensibility ratings, was in fact less intelligible when test subjects were presented with cognitively demanding tasks, simulating actual EMI lectures. We proposed that this effect of accent was due to the increased cognitive load required by the task combined with the increased cognitive load of parsing accented speech. In a recent study involving accented speech in noise and quiet, Dragojevic & Giles (2016) showed that native-listener attitudes towards a native or non-native accent are more negative when noise is added. Dragojevic & Giles suggest that the negative evaluation of the accented speaker is a direct result of the extra cognitive work required of the listener.

This study is inspired by Dragojevic & Giles but set in an EMI context with non-native listeners. It examines how noise affects attitudes to and intelligibility (and recall) of a simulated lecture given in either a more familiar native (American) English accent or a less familiar (Japanese) English accent. We hypothesize main effects of both accent and noise, and furthermore an interaction effect, viz. that noise will have a stronger effect on intelligibility and attitudes for the less familiar accent because of the increased processing cost, i.e. that adverse listening conditions – such as are often present in EMI teaching contexts – will affect non-native speakers more than native speakers. Results will be discussed in relation to the challenges of EMI at Nordic universities.

References

Bolton, Kingsley and Maria Kuteeva. 2012. “English as an academic language at a Swedish university: parallel language use and the ‘threat’ of English.” Journal of Multilingual and Multicultural Development 33 (5): 429–447. doi:10.1080/01434632.2012.670241.

Munro, Murray J., and Tracey M. Derwing. 1995. “Foreign Accent, Comprehensibility, and Intelligibility in the Speech of Second Language Learners.” Language Learning 45 (1): 73–97.

Dragojevic, Marko and Howard Giles. 2016. “I Don't like You Because You're Hard to Understand: The Role of Processing Fluency in the Language Attitudes Process.” Human Communication Research 42 (3): 396–420

Hellekjær, Glenn Ole. 2010. “Lecture comprehension in English-medium higher education.” Hermes 45: 11–34.

Jensen, Christian and Jacob Thøgersen. 2017. ˈForeign accent, cognitive load and intelligibility of EMI lectures.” Nordic Journal of English Studies 16 (3): 107-137.

Jensen, Christian, Louise Denver, Inger M. Mees and Charlotte Werther. 2013. “Students’ attitudes to lecturers' English in English-medium higher education in Denmark.” Nordic Journal of English Studies 12 (1): 87-112.

Analyzing L2 fluency with global wavelet spectrum

Heini Kallio,1,2, Antti Suni1, Štefan Benuš2,3 & Juraj Šimko1 1University of Helsinki, Finland; 2Slovak Academy of Sciences, Slovakia; 3Constantine the Philosopher University in Nitra, Slovakia

Fluency is one of the most commonly used descriptors of second or foreign language (L2) speaking skills. From a narrow perspective, fluency consists of temporal aspects of speech that include breakdown fluency, speed fluency and repair fluency (Lennon, 2000; Skehan, 2009). These aspects include disfluencies such as hesitations, unplanned breaks, and repetitions, often present in L2 speech. These disfluencies together with challenges in the production of L2 stress patterns (Hahn, 2004; Loukina et al., 2011) affect negatively the temporal regularity of speech, causing irrelevant prosodic phrasing and unpredictable changes in the speech tempo. In this study we examine L2 fluency as temporal regularity at different levels of prosodic hierarchy, proposing a new method for analyzing and characterizing L2 fluency.

We used wavelet spectrum of speech envelope for capturing temporal regularities in L2 speech at various time scales (for more detailed description, see Suni et al. 2019). The data consisted of read aloud stories produced in L1 Slovak and L2 English (n=61), and L1 English (n=7). Three professional teachers of English assessed the fluency, pausing, and speech rate separately from the speech samples. Linear regressions were used to find out which spectral components best correspond to L2 assessments.

Our results indicate that both syllable and phrase level regularities are present in the spectra. The results show clear differences between L1 and L2 speech, possibly arising from the irregularities in prosodic chunking in L2 speech, where disfluencies blur the rhythmic hierarchy. In our poster we will demonstrate, how the method captures both the holistic perceptions of fluency as well as individual aspects of pausing and speech rate.

References

Hahn, L. D. (2004). Primary stress and intelligibility: Research to motivate the teaching of suprasegmentals. TESOL Quarterly, 38(2), 201–223.

Lennon, P. (2000). The lexical element in spoken second language fluency. In Perspectives on fluency (pp. 25–42). University of Michigan.

Loukina, A., Kochanski, G., Rosner, B., Keane, E., Shih, C. (2011). Rhythm measures and dimensions of durational variation in speech. The Journal of the Acoustical Society of America, 129(5), 3258–3270. Skehan, P. (2009). Modelling second language performance: Integrating complexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510–532.

Suni, A., Kallio, H., Benuš, Š., & Šimko, J. Characterizing Second Language Fluency With Global Wavelet Spectrum. ICPhS 2019.

Children learning L2 pronunciation: Does frequent exposure to music facilitate learning through auditory training?

Katja Immonen1, Jemina Kilpeläinen1, Maija S. Peltola1 & Paavo Alku2 1University of Turku, Finland; 2Aalto University, Finland

Our earlier training studies have focused on different age groups (e.g. Taimi et al. 2014, Jähi et al. 2015, Peltola et al. 2017). The results have shown that children learn to produce difficult L2 sound contrasts quickly through listen and repeat training (Taimi et al. 2014). The present study investigated whether daily exposure to music in school would affect children’s L2 production learning.

Two groups of 9–11-year-old Finnish speaking children were tested (N=23). Group 1 consisted of 11 pupils from a music-oriented fourth grade. Group 2 had 12 pupils from a regular fourth grade. The aim was to see whether the difference in the amount of exposure to music would affect training results. The experiment consisted of four recording and four training sessions on two consecutive days. The first day started with a baseline recording and the second day ended with a final recording. The stimuli were two semisynthetic pseudo words: /tʉ:ti/ (target) and /ty:ti/ (non-target). Each training session included 30 repetitions of the target and non-target words (60 words in total per session). The children were instructed to listen to the words without repeating them during training. The recordings included 10 repetitions of both stimuli.

The results showed that both groups changed their pronunciation of the target vowel after three short trainings, which supports our earlier findings that children are fast learners. However, closer examination of the formant values revealed that Group 1 produced /ʉ/ closer to the target values already in the baseline recording and throughout the experiment, leaving less room for change as an effect of training. The results prove that children can learn to produce a difficult L2 sound through simple auditory training. Furthermore, our findings give an indication that daily exposure to music in school might benefit L2 production learning.

References

Jähi, K., Alku, P. & Peltola, M. S. (2015). Does interest in language learning affect the non- native phoneme production in elderly learners? In The Scottish Consortium for ICPhS 2015 (Ed.), Proceedings of the 18th International Congress of Phonetic Sciences. Glasgow, UK: the University of Glasgow. ISBN 978-0-85261-941-4. Paper number 0234.

Peltola, K. U., Rautaoja, T., Alku, P. & Peltola, M. S. (2017). Adult learners and a one-day production training - Small changes but the native language sound system prevails. Journal of Language Teaching and Research, 8, 1–7.

Taimi, L., Jähi, K., Alku, P. & Peltola, M. S. (2014). Children Learning a Non-native Vowel – The Effect of a Two-day Production Training. Journal of Language Teaching and Research, 5, 1229– 1235.

Prosodic marking of information structure by Russian L1 learners of English and native speaker perception of learners' intonation

Maria Kunevich, Brechtje Post & Dora Alexopoulou University of Cambridge, UK

This project reports the results of a production and a native speaker perception experiments that explore the acquisition of prosodic information structure (IS) marking by Russian L1 learners of English at different levels of proficiency.

Previous research has demonstrated an asymmetry between the acquisition of phrase-final focus accents and non-final focus accents. The asymmetry between these stress patterns emerges in production studies when L2 learners with varied L1 backgrounds are asked to respond to phrases with different IS contexts (broad vs. narrow focus). In general, L2 learners have difficulties in the correct placement and realisation of nuclear pitch accents and tend to overuse pitch accents as well as accent the last word of an utterance irrespective of their L1 and the word’s pragmatic status (cf. Raiser, Hiligsmann, 2007, 2009; Gut, 2009; Ramirez Verdugo, 2006).

Here we explore the link between phrase-final and phrase-initial accent patterns further in two studies with Russian L1 adult learners of English. Our aim is to investigate whether Russian learners of English mark focused words prosodically and whether native speakers can perceive such focused words as prominent.

In Experiment 1 (production experiment) we tested 36 Russian learners in three proficiency groups (A2, B1, B2 levels) on their ability to correctly produce nuclear pitch accent in six IS contexts. In Experiment 2 (online perception experiment) we tested whether native speakers of English can perceive certain words in learners’ productions as more prominent.

The findings show that there is a clear developmental path in acquisition of IS-prosody marking and learners’ experience in the language plays a role. Moreover, different IS contexts have a different impact on the shape of learner utterances. Native speakers are significantly better at correctly identifying prominent words for Subject narrow focus, followed by Object narrow focus, and finally broad focus patterns.

References

Gut, U., “Non-native Speech. A Corpus-based Analysis of Phonological and Phonetic Properties of L2 English and German”, Peter Lang, 2009.

Rasier, L., Hiligsmann, P. “Prosodic transfer from L1 to L2. Theoretical and methodological issues”, Nouveaux cahiers de linguistique francaise 28: 41–66, 2007.

Verdugo, D.R., “Prosodic realization of focus in the discourse of Spanish learners and English native speakers”, World Englishes 14, 9–32, 2006.

Study proposal - Familiarity effects in L1 and L2 accent processing

Anna Persson Stockholm University, Sweden

Studies on perception of accented speech indicate that accent familiarity affects both accuracy and speed of processing (e.g. Bradlow & Bent, 2008; Clarke & Garrett, 2004). As an increased processing cost for L2 accents relative to unfamiliar regional L1 accents has been found (Adank et al., 2009; Floccia et al., 2006), L2 accents have been suggested to impose additional processing costs over L1 varieties. That is, an accent that indexes non-nativeness may trigger different perceptual processes. However, this difference due to non-nativeness may be modulated by listener experience (Poretta et al. 2016).

This study aims to 1) establish whether the difference in processing cost between unfamiliar regional L1 accents and unfamiliar L2 accents as reported in previous studies are also true of L1 speakers of Central Swedish (CS) from Stockholm, and 2) investigate how training with a L2 variety affects processing times and possibly transfers to another L2 variety.

L1-CS speakers will be tested on their processing of four varieties using a cross-modal matching task: L1-CS speech, L1-Malmö speech, L2-CS speech, and L2-Malmö speech; the L2 speakers will have the same L1 and the same perceived accentedness level. In a second experiment, a different group of L1- CS speakers will be trained on either the L2-Malmö, the L2-CS, the L1-Malmö, or the L1-CS in noise, and subsequently tested on the L2-Malmö.

If the different processes hypothesis holds true, processing times for the L2 varieties after training should be reduced but still not comparable to the processing times for the L1-Malmö; furthermore, processing times for the L2 varieties should not differ. However, if experience modulates perceptual processes, training on the L2 varieties should result in processing times comparable to the L1 regional variety. In addition, the L2-Malmö should generate increased processing costs relative to the L2-CS. This study may advance our current understanding of perception of variable speech and bring further insight into L1 and L2 speech processing.

References

Adank, P., Evans, B., Stuart-Smith, J. & Scott, S. (2009). Comprehension of Familiar and Unfamiliar Native Accents Under Adverse Listening Conditions. Journal of Experimental Psychology: Human Perception and Performance, 35(2), 520–529.

Bradlow, A. & Bent, T. (2008). Perceptual adaptation to non-native speech. Cognition, 106(2), 707–729.

Clarke, C. & Garrett, M. (2004). Rapid adaptation to foreign-accented English. The Journal of the Acoustical Society of America, 116(6), 3647–3658.

Floccia, C., Goslin, J., Girard, F. & Konopczynski, G. (2006). Does a regional accent perturb speech processing? Journal of Experimental Psychology: Human Perception and Performance, 32(5), 1276–1293.

Porretta, V., Tucker, B. & Järvikivi, J. (2016). The influence of gradient foreign accentedness and listener experience on word recognition. Journal of Phonetics, 58, 1–21.

Comprehensibility of L1 Finnish speakers’ L2 Swedish

Henna Heinonen University of Jyväskylä, Finland

Comprehensible and functional pronunciation is considered to be an appropriate learning goal for second language (L2) learners (Levis, 2005). It is well-known that prosodic features have a great impact on comprehensibility (e.g. Thorén, 2008). Still, prosody has had only a minor role in formal education, especially in textbooks (Tergujeff, 2013), even if prosodic features seem to be challenging to learn without explicit instruction (Derwing et al., 1998). Therefore, it is relevant to investigate, to what extend the learning goal is being achieved.

In this poster presentation I discuss comprehensibility, i.e. ease of understanding, in L1 Finnish speakers L2 Swedish and specific features’ effect on comprehensibility. I also present how these aspects are acknowledged in textbooks that are the prime input source in language teaching in Finland (Luukka et al. 2008).

Research material consists of read aloud excerpts (n=21) by L1 Finnish high school students. Firstly, comprehensibility of the excerpts was rated by 64 L1-listeners on a 7-point verbalized scale. Secondly, pronunciation of specific features in excerpts was analyzed scalarly by 4 expert listeners and comprehensibility and pronunciation ratings were analyzed statistically. 10 Finnish textbooks were analyzed considering the results of the earlier to studies.

The results show that comprehensibility ratings fall in three different categories. In relation to pronunciation, the categories differ statistically from each other in primary stress and segment quality. The textbooks analysis showed that textbooks lack explicit instructions on prosody, especially primary stress, and the crucial segments are not paid much attention to. Thus, there is potential to improve pronunciation exercises to be more functional in respect of gaining comprehensible pronunciation.

References

Derwing, T. M., Munro, M. J., & Wiebe, G. (1998). Evidence in favor of a broad framework for pronunciation instruction. Language Learning, 48(3), 393–410.

Levis, J. (2005). Changing context and shifting paradigms in pronunciation teaching. Tesol Quarterly, 39(3), 369–377.

Luukka, M. R., Pöyhönen, S., Huhta, A., Taalas, P., Tarnanen, M., & Keränen, A. (2008). Maailma muuttuu - mitä tekee koulu?: äidinkielen ja vieraiden kielten tekstikäytänteet koulussa ja vapaa-ajalla. Jyväskylä: University of Jyväskylä.

Tergujeff, E. (2013). English pronunciation teaching in Finland. Jyväskylä studies in humanities, 207. Jyväskylä: University of Jyväskylä

Thorén B. (2008). The priority of temporal aspects in L2-Swedish prosody. Studies in perception and production. Stockholm: Stockholm university.

Comprehensibility and intonation in L2 pronunciation - is there a link?

Leena Maria Heikkola1 & Antti Saloranta2 1Åbo Akademi University, Finland; 2University of Turku, Finland

Background and aims: Comprehensibility, that is, how easily listeners understand an L2 speaker, can be used as a measure of L2 pronunciation (Derwing & Munro, 1995). In L2 Finnish, intonation has been shown to have an effect on the comprehensibility (Kuronen, 2016) but research on the topic is scarce, and has mainly focused on a few language groups, namely native speakers of Russian, Swedish or Thai (Kuronen, 2016; Nikonen, 2012; Toivola, 2011). The objective of the study is to examine how comprehensibility and intonation develop during two intensive courses of Finnish, and whether intonation patterns are linked to perceived comprehensibility.

Methodology: 70 Finnish language learners (FLLs) were tested at the start and end of two intensive courses of Finnish. The students read out loud 60 sentences in Finnish, which were recorded. The same sentences were read in the pre- and posttest, although the order was randomized for the two tests. Native Finnish speakers (NFSs) will be tested in September 2019. Intonation will also be rated by the authors on a 9-step Likert scale (compare, Munro & Derwing, 1995), and both FLLs’ and NFSs’ intonation patterns will be analyzed, and compared in Praat. Different FNSs will rate the sentences for comprehensibility on a 9-step Likert scale (compare, Munro & Derwing, 1995). Possible links between intonation patterns, and perceived comprehensibility will be analyzed statistically.

Results: We expect that the intonation patterns of FLLs’ become more native-like, and their pronunciation becomes more comprehensible during the courses. We also expect that more native-like intonation is linked to higher perceived comprehensibility.

Conclusions and implications: On the basis of the results, we will see, whether and how intonation is linked to comprehensibility in L2 Finnish. Also, we will be able to recommend possible implications for teaching L2 Finnish pronunciation to improve comprehensibility of L2 Finnish speakers.

References

Kuronen, M. (2016). Uttal av S2-finska med fokus på svenskspråkiga talare. (Pronunciation of L2 Finnish with a focus on Swedish speaking speakers.) Puhe ja kieli, 36(3), 147–174.

Munro, M. J. & Derwing, T. M. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 45(1), 73–97.

Nikonen, M. (2012). Kysymysten prosodiaa ja syntaksia. Thainkielisten suomenoppijoiden kysyvät konstruktiot keskustelussa. (Prosody and syntax of questions. Thai language Finnish language learners’ asking constructions in conversations.) MA Thesis. University of Helsinki. Available at: http://urn.fi/URN:NBN:fi-fe201301281669

Toivola, M. (2011). Vieraan aksentin arviointi ja mittaaminen Suomessa. (Assessment and measurement of foreign accent in Finland.) PhD Thesis. University of Helsinki. Available at: http://urn.fi/URN:ISBN:978-952-10-7217-8

L2 pronunciation of Swedish on different proficiency levels

Maria Kautonen University of Jyväskylä, Finland

In this presentation, I explore L2 pronunciation on different levels of oral proficiency. The proficiency levels are from A1 to C2 in the Common European Framework of Reference. Research that combines pronunciation and oral proficiency level is scarce even internationally, not to mention pronunciation analysis of free speech (e.g. Piske et al. 2001). The present study provides thus useful knowledge on L2 pronunciation on different proficiency levels and the learning paths in L2 pronunciation learning that will benefit both language teaching and assessment.

The learner group in this study consists of Finnish-speaking learners of Finland-Swedish. The speakers’ pronunciation in a monologue task is studied in two language tests; assessment of learning outcomes in Swedish and National Certificates of Language Proficiency. Pronunciation on proficiency levels A1 to B1 (n=68) is analyzed acoustically by four listeners with phonetic expertise, and pronunciation on levels B2 to C2 (n=12) by Finnish-speaking and Swedish-speaking university students of Swedish (n=18). Moreover, intonation of speakers on levels B1 and B2 (n=8) is analyzed acoustically by comparing the fundamental frequency of speech and intonation contours with L1 speakers of Finland-Swedish (n=5).

Even though Finland-Swedish and Finnish share several phonetic similarities, Finnish- speaking L2 learners face challenges in L2 pronunciation of Swedish on both segmental and prosodic level. These challenges are relatively similar on the different proficiency levels, but there is also clear improvement in intonation and sentence stress towards a more native-like pronunciation as the proficiency level gets higher. These findings support previous research findings stating that prosody has an important role for oral skills assessment (cf. Kang 2013).

References

Kang, O. (2013). Linguistic analysis of speaking features distinguishing general English exams at CEFR levels. Research Notes, 52, 40–48.

Piske, T., MacKay, I. & Flege, J. (2001). Factors affecting degree of foreign accent in an L2: A review. Journal of Phonetics, 29, 191–215.

A single-case study of L2 Swedish word-final stop voicing pronunciations

Maryann Su Lin Tan, Xin Xie, Florian Jaeger Stockholm University, Sweden

Acquiring native-like pronunciation is one of the most difficult feats for L2 learners. Non- native accents are pervasive even after years of L2 learning, with potential social consequences (Fuertes et al., 2009). Formal instruction is not necessarily effective, partly because pronunciation is not prioritised in foreign language teaching (Piske et al., 2001).

We assess the effectiveness of a common instructional approach to L2 accent reduction— repeating (‘mimicking’) words immediately after hearing the native pronunciations. We examine pronunciations of word-final dental-stops of a beginning L2 learner of Swedish with Flemish L1. Stop voicing was expected be challenging for this speaker given that Flemish stops are neutralised word-finally.

Speech stimuli comprised 32 minimal pairs of monosyllabic Swedish words (e.g. rötröd). We analyze pronunciation along three phonetic dimensions known to cue voicing: vowel, closure, and burst durations. To assess the effects of mimicking, we compare native L1 to L2 productions elicited under two conditions: unassisted or mimicked.

Native vs. unassisted L2 speech. All three measures exhibited a significant interaction between voicing and elicitation condition (ps < .006) but the pattern of interaction differed across the three measures. The L2 speaker did not distinguish the /d/-/t/ contrast on the primary native cue (closure), but instead used a cue that is secondary to native speakers (burst).

Unassisted L2 and mimicked L2 speech. All measures exhibited a significant interaction (ps < .0002). When mimicking, L2 production of cues in /d/ vs /t/-final words came into closer alignment with native production although they still differed significantly from native speech.

This presents the first small-scale analysis of the phonetics of L2 word-final stops in Swedish (previous research has employed phonologists’ judgments, e.g., Hammarberg, 1996). We find that immediate repetition seems to substantially reduce the non-nativeness of pronunciations, though our results leave open how long lasting this effect is.

References

Fuertes, J. N., Gottdiener, W. H., Martin, H., Gilbert, T. C., and Giles, H. (2012). A meta-analysis of the effects of speakers' accent on interpersonal evaluations. Eur. J. Soc. Psychol., 42, 12–133.

Hammarberg, B. (1996). Conditions on transfer in phonology. In A. James & J. Leather (Eds.), Second-language speech: Structure and process (161–180). Berlin: Mouton de Gruyter.

Piske, T., MacKay, I. R. A., & Flege, J. E. (2001). Factors affecting degree of foreign accent in an L2: A review. Journal of Phonetics, 29(2), 191–215.

When similar meets similar and new: Perception and production of French L3 obstruents in Scandinavian learners of French as a foreign language

Pernille Berg Johnsson Uppsala University, Sweden

Textbooks in French phonetics at Scandinavian universities (cf. Rankka et al. 1995, Jensen & Thorsen 2005, Girard & Lyche 2005) often tend to focus on sounds that are absent in the learners’ native language, such as /z, ʒ/ (and /ʃ/ in Danish) or /ɛ̃, ɔ̃, ɑ̃/ (new sounds), and on sounds that are similar to sounds in the L1 inventory, but phonetically different from these, such as /p, t, k, b, d, g/. French obstruents thus represent many of the well-known segmental difficulties for ‘Scandinavian’-speaking learners. Two previous studies in Scandinavia have addressed learners’ difficulties by considering their perception of spoken French, both focusing on word recognition (Stridfeldt 2005, Hedevang 2007) rather than on the relationship between perception and production and without following learners’ development.

This newly started PhD project aims at investigating 90 Swedish/Norwegian/Danish- speaking French language students’ perception and production of new (e.g. /z/) versus similar (e.g. /p/) French L3 obstruents during the first semester of their bachelor degree. Considering the similarities of the Scandinavian students’ native languages as well as their differences; how do their perception and production of French L3 obstruents develop when they are receiving phonetic instruction? Does phonetic (dis)similarity influence them and in what way? How do French native listeners perceive the students’ productions?

We are planning to conduct four pre- and post-tests (discrimination, identification, picture naming and free production), with the identification and the picture-naming task being our main focus in this presentation, as well as important considerations when selecting stimuli for the two tasks in question. By gaining more insight into French L3 obstruents’ degree of difficulty, the extent to which Scandinavian learners’ pronunciations of these sounds affect native French listeners’ perceived acceptability and to which learners improve when receiving phonetic instruction, we will provide new and nuanced perspectives with possible implications for the teaching of French L3 pronunciation in Scandinavia.

References

Girard, Francine & Lyche, Chantal S. (2005). Phonologie et phonétique du français (4th edition). Oslo: Universitetsforlaget.

Hedevang, Lise. (2007). Talt fransk i danske ører. Ordgenkendelse og ordgenkendelsesstrategier hos danske fransklearnere. Aarhus: Department of languages, literature and culture, Aarhus University.

Jensen, Ole Kongsdal & Thorsen, Oluf M. (2005). Fransk fonetik & fonologi (2nd edition). Copenhagen: Department of English, Germanic and Romance Studies, University of Copenhagen.

Rankka, Erik, Barjac, Jean, & Axelsson, Christiane. (1995). Kurs i franskt uttal: För användning i språkinlärningslaboratorium (SL) (6th edition). Uppsala: Department of Romance languages, Uppsala University.

Stridfeldt, Monika. (2005). La perception du français oral par des apprenants suédois. Skrifter från moderna språk: Umeå University

Varieties of World Englishes: Nigerian Attitudes and Perceptions

Temitayo Olatoye University of Eastern Finland

The conflict between native and non-native norms in multilingual societies, in which English is chiefly acquired as a second language remains unresolved (Bamgbose 1998). In Nigeria, British English remains the prescriptive variety of English used in the educational system. However, a growing influence of American English cannot be ignored (Igboanusi 2003), while an indigenized “Nigerian” variety is in fact prevalent. In an increasingly globalized world, in which encounters between native and non-native speakers of different varieties of English are on the rise, is there an ongoing reorientation of language attitudes among educated Nigerians?

This research examines the attitudes of educated Nigerians towards six varieties of English: Standard British English (SBE), General American English (GAE), Scottish English, Ivorian English, Acrolectal Nigerian English (ANE) and Mesolectal Nigerian English (MNE). Using the Verbal Guise Technique (VGT), the attitudinal evaluations of educated Nigerians were obtained via a six-point semantic differential scale with nine personality traits, and subsequently analysed along status and solidarity dimensions. To gain a better understanding of their dialectal awareness, a speaker identification task was also included.

An analysis of 102 responses indicated a preference for the ANE and GAE varieties in terms of status and solidarity. The MNE was stigmatized in terms of status, but favourably rated in terms of solidarity. On the other hand, the SBE was considered prestigious in terms of status, but both SBE and SCOT received the least favourable ratings in terms of solidarity. These findings differ from previous studies, in which heavily accented speakers of a local variety were considered more socially attractive (McKenzie 2008) or stigmatized in terms of both status and solidarity Chan 2016). These results suggest that educated Nigerians have an overwhelming preference for a local variety, devoid of heavily marked features.

References

Bamgbose, A. (1998). Torn between the norms: innovations in world Englishes. World Englishes, 17(1), 1–14.

Chan, J. Y. H. (2016). A multi-perspective investigation of attitudes towards English accents in Hong Kong: Implications for pronunciation teaching. Tesol Quarterly, 50(2), 285–313.

Igboanusi, H. (2003). Knowledge, use and attitudes towards Americanisms in Nigerian English. World Englishes, 22(4), 599–604.

McKenzie, R. M. (2008). Social factors and non-native attitudes towards varieties of spoken English: A Japanese case study. International Journal of Applied Linguistics, 18(1), 63–88.