Minority language learning in Kven through conversation

ABSTRACT

Language learners of Kven participated in informal learning settings using the Master Apprentice method, a method often used in language revitalization. The use of this method is studied in the light of sociocultural theory of language learning, which focuses on the relationship between collaborative learning and learner autonomy. The students of Kven improved their oral proficiency when using the language with the mentors in informal conversations. The informal learning setting increased learner autonomy among the participants because they exercised learner agency and used language learning strategies. Language learning takes place not only cognitively, but also emotionally. Anxiety can discourage learners from using language; self confidence influences the courage to speak. The MA method helped learners to master their negative feelings, and to lower anxiety. Positive feedback helped them take initiative for the use of the Kven language independently after first practising it with their mentor. However, beliefs about language learning seemed to guide the choice of learning activities. Practical activities, which are usually included in the use of the MA method, were not included, even though they would probably have influenced language learning in a positive way.

KEYWORDS:

• Kven
• language revitalization
• learner autonomy
• Master Apprentice method
• new speaker
• sociocultural theory

Introduction

The article presents the use of the Master Apprentice method (hereafter, the MA method) in the Kven language studies unit at UiT The Arctic University of Norway (UiT) during the academic year 2015/2016. The learning of oral ability in an endangered language like Kven is a challenging task, because only a few resources are available, due in part to a low number of mother-tongue speakers, limited exposure to the language in the media, and limited teaching materials. The MA method creates an environment for oral language use to counter these hindrances. A student is assigned a mentor, and the language is learned through informal conversations (Hinton et al. 2018).

Language learning beyond the classroom environment has received increased research interest in recent years, not only in revitalization studies, but in a variety of other contexts (Benson 2011). In so-called “beyond classroom” studies, language learning is considered basically to be a social process, as seen in this study. In addition, “the social turn” (Swain and Deters 2007; Benson 2011, 2014) in language learning studies tries to capture learners’ perspectives using qualitative analyses of students’ narratives of their learning.

Achieving oral competence in an endangered language is important in language revitalization. This article focuses on how interaction facilitates oral language learning, asking the questions:

• Did the MA method help students improve their oral ability in Kven?

• Did their autonomy in language learning develop while using the MA method?

• Did the use of Kven extend from mentor meetings to other contexts?

These questions are connected with the overarching question of whether the MA method was helpful in revitalizing Kven.

The first section of this article presents the efforts being made to revitalize Kven. The following section frames the MA method within sociocultural theory, and presents the data and method of the study. The next section details the results and discusses the significance of the findings to the revitalization of Kven. Lastly, a short conclusion recaps the study.

The Kven language today

Kven is a language closely related to Finnish that has been spoken among the Kven minority in Northern Norway for at least the past three hundred years. The existence of the Kven language in Norway is a result of migration from Finnish-speaking areas in Northern Finland and Sweden to Northern Norway mostly during the eighteenth and nineteenth centuries. The Kven language lost many speakers during the period of modernization of Norwegian society between the 1850s and the Second World War (Lindgren 2009, 111–113). Today, the number of those who can speak Kven is estimated at between 1500–10,000, although no exact numbers exist (Målrettet plan 2017, 14). An “ethnic renaissance” (Lindgren 2009, 116) started to take root among the Kven minority in the 1980s and resulted in the creation of an ethnic organization, Norske Kveners Forbund (NKF, The Norwegian Kven Association) in 1987. The revitalization began seriously after Kven got a status as a language in 2005. Kven is recognized by the Norwegian Government as a national minority language in Norway, following the European charter for Regional and Minority languages (Lindgren 2009, 116–118). In this section, the focus is on the revitalization measures of the Kven language, and especially on its new domains of use, with a description of institutions, activities and actors in the revitalization process.

Writing in Kven represents a new domain of use. In 2004, Alf Nilsen-Børsskog published the first novel written in Kven, and he continued as an active writer until his death in 2014. A written standard of Kven was created after the recognition of it as a language in 2005, and was initially based on the standard set in Nilsen-Børsskog’s first novel. It was particularly recognized that a standard form of the language was needed to create teaching materials for schools and for third-level courses (Söderholm 2010). Söderholm’s Kven grammar book was published in 2014, followed, in 2017, by a translation into Norwegian (Söderholm 2014, 2017).

Another new domain has been the use of Kven in education. Teaching materials and a grammar book were produced for primary- and secondary-level students (Keränen 2018), though not many to date have chosen to study Kven. One reason for this has been the challenge to find qualified teachers. Courses for those who wish to qualify in teaching Kven language and culture only started in the 2019/2020 academic year at the UiT campus in Alta. Credits in Kven language and culture are included in the curriculum of teacher education. Students can choose optional courses, totalling 60 ECTS (one year full-time study according to the European credit system), in Kven language and culture during the five-year Master’s in Education at UiT.

Institutions can support language revitalization in many ways (Dwyer et al. 2018). The Kven Institute, launched in 2007 and located in Børselv in Porsanger municipality, has been responsible for many revitalization activities, the most important being corpus planning and the creation of written Kven (Keränen 2018). In 2014, they launched a project to establish Kven at the kindergarten level, a domain where Kven was not used for many decades, and a key communication point between adults and children.

In terms of media, Kven is used in the Kven newspaper, Ruijan Kaiku, and within the NRK, the Norwegian Broadcasting Corporation. New Kven language centres have been established in the municipalities of Vadsø and Kvænangen in 2018, and in Porsanger in 2019. In addition, the Halti centre for Kven culture in Nordreisa, and the language centre of Storfjord in Skibotn work with the revitalization of Kven.

Kven was introduced in a totally new domain, higher education, in 2006, when studies in Kven were launched at the UiT. Previously, only studies in the Finnish language were given at UiT, and many with a Kven background chose to study it. However, as a consequence of the new language status of Kven, university courses in Kven were established. Among students who participated in Kven courses were Kvens who had oral competence in their language and were interested in learning how to write it. However, Kvens without that oral competence who had previously studied Finnish, and other students without any ethnic Kven background, participated in these studies. Since 2006 Kven language courses have been included in a one-year study, whereby students could complete 60 ECTS in Kven language and culture; today, both a bachelor’s and a master’s programme in the Kven language have been established. Expanding this even further, the MA method in Kven studies has opened a new practical domain, namely that between students and their Kven mentors.

Since 2005 revitalization has developed in many other areas. The primary activists are the Kvens themselves, working through their own ethnic organizations. Some Kven language activists are still mother-tongue speakers, others turned to using Kven after learning Finnish in an educational context. Finnish speakers, either mother tongue or bilingual speakers, can also be found among staff members in the institutions involved in Kven revitalization, as teachers and producers of teaching materials.

The MA method project in Kven was initiated as a collaborative project between the UiT, the Kven organization NKF and the Kven Institute. The project received financing from the education fund at the UiT, and UiT was the project lead. NKF contributed to financing and helped with finding mentors among its members. The Kven institute participated by giving an introductory course about the method. The MA method project is seen as an important part of revitalizing Kven as oral use of Kven among minority members is declining. One other motivation to initiate the MA method project was to increase the use of the language among those Kvens who are able to speak it.

The main principles of the Master Apprentice method in the light of sociocultural theory of language learning

This section will present the basic principles of the MA method (Hinton 2002; Hinton et al. 2018), followed by a discussion of them in the light of sociocultural theory of language learning (Lantolf and Thorne 2006). The focus is on interaction and autonomy in language learning. The concept of a “new speaker” will also be presented.

The MA method follows the principles of language immersion: A mentor (master) who is a fluent speaker – usually a mother-tongue speaker – of an endangered language and a person (apprentice) who wants to revitalize his or her language, are connected, and they use the language in informal conversations, typically by participating in everyday practical activities. The MA method is used to reach the generation of people who never learned their language in natural environments, at home or in a language community, due in part to negative attitudes towards indigenous and minority languages which led to parents no longer speaking their language to their children. The MA method was first developed in the USA (Hinton 2002; Hinton et al. 2018); afterwards, it was used in, for example, the revitalization of the Enare Saami language in Finland (Olthuis, Kivelä, and Skutnabb-Kangas 2013; Pasanen 2015). The MA method has been adapted around the world, including Australia and Canada (Hinton et al. 2018).

One of the most important principles of the MA method is that participants use only the endangered language – or target language (TL) – during interaction (Hinton 2002). Equally, interaction in the TL has been a central issue in sociocultural theory of language learning. Interaction itself is considered to facilitate language learning, not only to offer input. The importance of interaction is based on Vygotsky’s theory of collaborative learning. Language learning starts as a social activity, and afterwards it is internalized by the learner on a psychological level (Mitchell and Myles 1998, 145, 162).

Sociocultural theory describes learning, including language learning, as a process of internalization. The relationship between an individual and the external, social-material world is mediated by artefacts. Artefacts are concrete tools, or conceptual and semiotic artefacts at a higher level, the most important of these being language. Regulation is a form of mediation, which refers to the control of human activity. Regulation can be external, called other-regulation, or internal, called self-regulation. Other-regulation happens e.g. in collaborative language learning when a learner gets feedback from an interlocutor. Self-regulation means that a learner is already proficient enough to make the regulation by herself. Internalization is a developmental process, where the reliance on external regulation decreases, whereas internal (self) regulation increases (Lantolf, Thorne, and Poehner 2015, 208–212).

The most famous concept of Vygotsky is “the zone of proximal development”(ZPD). Murray (2014, 234) calls ZDP “the metaphorical space between what learners can do on their own and with help from others.” The goal of learning is autonomy, and ZPD helps to understand the relationship between autonomy and collaboration in language learning (ibid. 234). ZPD is most often studied in a formal learning context; however, it does not refer to an instructional strategy, but to a developmental process (Ohta 2017, 65). ZPD refers not only to developing performance in the TL, but indicates that learners take more responsibility for appropriate performances in the TL (Lantolf, Thorne, and Poehner 2015, 213–214).

In the MA method, the responsibility to plan and carry on meetings with mentors is placed on the apprentices. Learners are encouraged to be active, taking charge of their own learning (Hinton et al. 2018, 129). Mentors are often elderly people, and they have no education as language teachers. Students are responsible for planning the meetings, so they need to develop autonomy in language learning.

The ability to have access to learning strategies is sometimes considered to lead to learner autonomy (Benson 2007, 28). Different strategies facilitating language learning are presumed. First of all, metacognitive abilities regulate learning in general terms. Metacognitive abilities are used in the planning, goal setting and evaluation of learning. Other strategies are more specific, like: affective strategies which can, for example, help lower one’s anxiety; social strategies which increase interaction with the TL; cognitive strategies which are used in a specific learning task; or communication strategies which promote communication with others and contact with the TL (Williams and Burden 2004, 149–156).

Even though learning strategies also help learners (Reinders 2011, 181), it has been asked how central these concepts are for autonomy in language learning (Benson 2007, 28). Autonomy is seen to include more than learning strategies. Sociocultural approaches, in particular, consider learners as intentional human beings who have “human agency”, and are actively engaged in learning (Lantolf and Thorne 2006, 142). Agency is related to Vygotsky’s concept of self-regulation (van Lier 2008, 174). According to Benson (2007, 30), agency leads to autonomy. Still, autonomy is a complex concept which in addition to agency, is also related to the concepts of motivation and identity (Benson 2007, 34).

Additionally, the European Language Portfolio focuses on the importance of learner autonomy (Little 2012). It includes a self-assessment tool called the Common European Framework of Languages (CEF). Learners can evaluate their competence in the different areas of language use: reading, writing, oral interaction, oral production and listening. Oral language competence is divided into three different competences: interaction, production and listening. The students in this study were asked to evaluate their oral language competence in Kven both before and after the MA method project following the European Language Portfolio descriptions.

A further question is whether learning beyond the classroom fosters the capacity to autonomy (Benson 2007, 27). The use of the MA method means that language learning is situated outside the classroom, in an informal context. In addition, participating in common everyday activities using the TL is one of the central principles of the MA method (Hinton 2002). Informal learning is described as non-intentional; where knowledge, skill and ability are acquired from daily experiences (Manolescu, Florea, and Arustei 2018). Language learning is often informal by its nature. Tomasello (2003, 19–31) describes how language learning happens when a child participates in “joint attentional frames.” In this case, the focus is not on language learning but on social activities, and language learning happens when a child understands a caretaker’s communicative intentions in social situations. Lantolf and Thorne (2006, 173) suggest that Tomasello’s theory is also relevant for adult second-language learning. There are many examples of adults who learned a language in an informal context, like Kvens who learned Norwegian informally and became bilingual during the period of modernization and even earlier (Lindgren 2009, 109–110).

Sometimes, learning outside the classroom is called non-formal: it is still intentional, and its purpose is to help learners maximize their language learning. Non-formal learning is different from informal learning as it is guided by procedures which aim to reach the learning targets (Manolescu, Florea, and Arustei 2018). Indeed, a manual written by Hinton (2002) presents a kind of procedure for the use of the MA method. Thus, using the MA method in language learning could also belong to non-formal education, as the purpose is to help learners acquire an endangered language in the most effective way possible.

Participants that have learned language in revitalization programmes are traditionally called “second language learners” (e.g. Hinton et al. 2018, 135); however, the concept of new speaker has recently been established. One motive for using this concept is the specific environment for language learning, namely, different educational contexts that lie outside a language community. However, some learners can have contact with a few speakers (Hornsby 2015; Lane and Makahara 2016); this makes endangered language learning different to language learning in many other contexts. Another reason to use this concept is in referring to learners whose aim is language revitalization; who are, in many cases the people who form the future language community (Jaffe 2015, 23–25). This means that new speakers have to become autonomous in the TL. Most obviously, new speakers have both high motivation and a strong language identity – the concepts related to learner autonomy (Benson 2007, 34).

New speakers can be differentiated from old speakers, referring to those who learned the language in their communities. New speakers sometimes experience a lack of acceptance as “authentic” or “legitimate” speakers by old speakers. This is because learning a language in a formal context often creates speakers without identifiable local features, which are important for minority language speakers, as they are not used to standardized language (Hornsby 2015; Lane and Makahara 2016, 314).

Using the frame of sociocultural theory makes it possible to analyze the particular significance of collaboration between mentors and students when using the MA method. Collaboration is an essential process in developing autonomy in language learning, and further, autonomy is an important goal in the process of becoming a new speaker in an endangered language.

Data and the method of the study

The use of the MA method was initiated by different actors among those who work with the revitalization of Kven language and culture (see the second part of this article). As the MA method was not previously used in the revitalization of Kven, this study presents a new context where this method has been used. The author was responsible for managing the project, together with a project assistant. Consequently, the project can be termed as action research because the author participated in the project being analyzed.

Eight students in the Kven studies unit and six mentors participated in the MA method project. Before commencing, all participants were offered a two-day introductory course where they learned about the principles of how to use the MA method in language learning. This course was arranged by the Kven Institute, one of the project partners.

The mentors were found through the help of the Kven organization, NKF. Only six mentors started in the programme. The goal was to find mentors close to where the students lived, but this was not always possible. All mentors were fluent speakers of Kven.

Students had different levels of knowledge of the Kven language before they started. Many of them came from Kven families or had Kven ancestry. Some were beginners, others had used Kven before; they may have had a receptive knowledge of the language or even been able to communicate in Kven. In addition, some students already knew Finnish. All students recruited were doing Kven studies at the UiT. Participation was voluntary and all those who started Kven studies in the autumn term of 2015 also signed up as participants in the project.

The project was evaluated in two ways. One was by log writing, often used in professional studies to increase students’ ability to evaluate phenomena they meet in practice. Students write down thoughts, experiences and impressions. In an ideal case, writing is done immediately after training to capture the immediate senses of situations. The goal is to increase reflection on training, and in this way to increase its quality (Dysthe, Hoel, and Hertzberg 2010, 70).

Students were asked to write about three different topics in their log writing: a plan before a meeting with their mentor; how the plan was implemented, and an evaluation afterwards. Six students delivered 38 logs of varying lengths which give a picture of conversations and activities during the project.

After the project period (the academic year 2015/2016) was over the students were asked to fill in a questionnaire, the second method used in the project. The questionnaire included 17 questions and students were asked to evaluate the meetings with their mentors, and to evaluate their oral ability in Kven both before and after the mentor project. Both general and specific questions about the MA method were included, together with an open-ended question. Six out of eight students completed the questionnaire. As some of the questions overlapped with the information delivered in the logs, both sources were evaluated together in the analysis. The students who wrote logs used them when they answered the questionnaire; those who did not write logs gave their evaluation only in the questionnaire.

The method used conveys an emic perspective on language learning (Swain and Deters 2007, 820), as it gives a voice to students who learned Kven using the MA method. Qualitative analyses are based on narratives delivered in logs and questionnaires where students tell and reflect about the learning of Kven. The narrative inquiry has been receiving growing interest as a research method for language learning under terms like “language biography”, “language learning history”, or “autobiography”. Narrative inquiry focuses on reflective accounts of learning experiences. Narratives are analysed using qualitative methods, like thematic analysis (Benson 2014, 154–157).

It was a challenge to keep the anonymity of the informants because they were so few. When excerpts appear from logs and questionnaires, there is no separation between informants. However, all excerpts of logs and questionnaires presented are from one log entry, and not a composite of several logs. The students mostly wrote their answers in Norwegian and one student wrote in Kven. The excerpts below are given in English translation, in an effort to keep their sources as anonymous as possible. The project received research permission from the Norwegian Centre for Data Research (NSD) (Project 47165 Tilegnelse av kvensk). All the participants who delivered logs and questionnaires gave permission to use these documents in research.

The analysis seeks to answer the research questions by presenting analysis from the logs and questionnaires. The first question about improving the oral ability is analyzed both from the students’ self-evaluation of their oral ability in Kven, and from the way the questionnaire answers were worded. The question of developing autonomy in language learning is analyzed in narratives about activities done together with mentors, and in narratives that describe learning strategies or demonstrate learning agency in logs and questionnaires. The analysis also concentrates on the question of whether non-formal or informal learning characterizes students’ descriptions of mentor meetings. The use of Kven outside mentor meetings is analyzed from students’ descriptions of where and with whom they practised oral Kven.

Results

The analysis of logs and questionnaires is structured as follows: first, by presenting how the students described and evaluated meetings with the mentor; second, a description of the development of oral ability; third, other topics concerning the use of the MA method which emerged from the analysis are considered.

How students described and evaluated meetings with mentors

The first focus is on the students’ plans for meetings and their implementations, followed by their reflections. Plans included what they would talk about, time and place for meetings. Meetings were arranged at the participants’ homes, in libraries, or in language cafés, where other Kvens joined in. Sometimes Skype or telephone was used.

Topics

In their conversations, students often used pictures (see Hinton 2002, 114–119) that were distributed to them during the introductory course:

1. Conversation with the mentor around two of the “comic strips/ cartoons” that were given out during the introductory course. The word choice was initially simple, such as the name of the room, appearance of a person, clothes and shoes, household items. Through repetition, new words came up and the sentences became more complete. (Logs)

The use of pictures is a well-known method in language classes as well as in mentor programmes. Students practised the vocabulary in the pictures, and made stories based on them. Objects like kitchen implements were also used to learn vocabulary. Other resources used in conversations included texts in Kven (like the textbook in the Kven course), children’s books, newspapers (like the Kven paper Ruijan kaiku), as well as documentary films about Kvens like those made by Anstein Mikkelsen.

Topics of conversations covered work or hobbies, the weather, holidays, families, relatives and the home background of mentors. Maps were sometimes used when talking about place names in Kven. Those students who were already at a communicative level in Kven were able to talk about topics like the Kven language, language revitalization, or minority politics. Students who knew Finnish were interested in learning more about the differences between Finnish and Kven.

Not all topics were easy to talk about, as the Kven language has not been used in the domains concerning modern societal issues (Lindgren 2009, 112–113). Consequently, it was easier to talk about the past, like a mentor’s home town or issues and stories from their childhood. When talking about modern topics, conversations changed into Norwegian, although this was not the only reason for Norwegian being used. Beginners found it particularly difficult to use only Kven and pointed out that it was sometimes necessary to use Norwegian words just to be able to carry on a conversation. Thus, the use of Norwegian was a communicative strategy used when the communication could otherwise have stopped.

Interaction and learning strategies

Students described how they sometimes needed help from their mentors to continue a conversation.

• 2) The mentor spoke Kven, and I tried to answer. I managed to give some answers, with a little help. (Logs).

Students described how the mentor helped them especially in the first meetings: the need for other-regulation was still necessary for production (Lantolf, Thorne, and Poehner 2015, 209). Often it was a student who needed help, but if a student already knew some Kven, the student and the mentor sometimes tried to find the correct words and phrases together. Therefore, both participants were engaged in helping each other to produce Kven; the roles were reversed (cf. Lantolf, Thorne, and Poehner 2015, 213).

Students were often concerned about how to improve their Kven vocabulary:

• 3) Most everyday words and phrases are familiar to me, but I am unable to reactivate them as quickly as I would like to. Therefore, I avoid words that I experience as difficult to retrieve. I felt that it was easier to find the words when I was able to calm down and I could repeat them several times, as we did. Having a concrete topic with pictures made the conversation easier. (Logs)

Excerpt 3 demonstrates how the learner used different learning strategies: Avoidance is a communication strategy which can be used when learners meet difficulties with a specific area of language, in this case the vocabulary. “To calm down” is an affective strategy, with the aim of making the conversation situation less anxious, and repetition is a cognitive strategy used to help with memorizing (Brown 2000, 125, 128–129; Williams and Burden 2004, 152–154). As in excerpt 3, students often mentioned that it was easiest to talk about concrete and familiar topics, in which cases the pictures were considered to be helpful. However, not all students were interested in talking about pictures, because such conversations soon became boring. A student commented that the best thing about the MA method was when “I was able to talk about something real” (Questionnaires).

A problem that was noted in both the logs and the questionnaires is that the students were advised not to write down unfamiliar words during the meetings. This advice was based on the assumption that writing activities can take over; accordingly, the oral use of language will suffer (Hinton 2002, 2). However, students felt that by writing down new words, they could have learned them more effectively. It became especially problematic to memorize words when students and mentors did not meet very often. Still, some students emphasized that it was effective to try to concentrate only on listening:

• 4) It's good to repeat words many times. Some words are easy to get stacked in your mind; others are harder to remember. The setting with only visual, oral and auditory focus is new, but relaxing when putting away the habit of noting everything. We're expanding the conversation areas a little bit every time and this means that I can learn something new out of what I have already learned earlier. (Logs)

Excerpt 4 describes that it is demanding to get used to a new way of learning. Taking notes is an activity belonging to a formal learning context, which students are used to doing. However, the excerpt demonstrates that it was possible to make changes in established learning habits; moreover, the new way of language learning was even relaxing. In addition, this log also indicates that repetition is a cognitive strategy that the students often used.

The importance of practicing listening is in focus in the following reflection:

• 5) I'm still trying to talk more and more, feeling it going forward, but slowly. Realize that I still have to listen to more, I hear [recordings of] Aikamatka from the start again now, and understand most of it because I have heard it many times now. But as soon as something new comes, I have trouble understanding. I have plans to listen several times to the Finnish program on NRK in order to get the language to “get into the ear” I have to work more. (Logs)

This log demonstrates that the MA method raised an awareness that not only production, but understanding are important parts of oral competence. It is apparent that understanding oral language is seen as a result of practicing, and along with audio files of the Kven course book, Aikamatka, the student describes plans to find other resources for listening. So, the excerpt reveals the use of communicative strategy when finding more resources for listening (Williams and Burden 2004, 150); in addition, it shows learner agency.

Informal vs. non-formal learning and learner agency

Focus on grammar is unusual in the MA method. Still, one of the students described how some demanding parts of Kven grammar, like pronouns, became understandable when they were practised with the mentor:

• 6) Concerning the pronouns, I have now understood that the case “is interfering” here. [Inflection in case is also a matter concerning pronouns]. I got an Aha! moment here recently. That “mistä” means whence and is in elative. Has it not appeared in the teaching earlier?? Or have I just slept during the course?? Or was it said before I started to understand anything at all? There is suddenly a logic here that I have not understood before, as Norwegian does not have the case in the same way as Kven. (Logs)

The students in this project had also received formal instruction in Kven, including grammar concepts. But learning language in an informal context is essentially connected to an implicit knowledge of language, which means that it is unconscious (Lantolf, Thorne, and Poehner 2015, 219–221). It is not unusual that learners do not understand everything that is taught to them in a classroom, and sometimes explicit knowledge like grammar concepts are understood only when practising, as this excerpt demonstrates.

Conversations were activities practised with the mentors, but participants felt there was a lack of practical activities. One student commented:

• 7) I should have been more consistent with using Kven, be tolerant when I didn’t understand and repeat or explain in other words. Possibly we ought to have had more variation and done more together; gone hiking, cook meals, etc. We shouldn’t just have talked about things in the pictures. (Questionnaires)

Communication while engaged in practical activities might have increased real communication, as this student also seemed to expect. Excerpt 7 shows that the student noticed there were other communication strategies to be used alongside the use of Norwegian in situations where communication did not work. To explain things using “other words” is a communication strategy to compensate for words that the learner does not know (Brown 2000, 128). However, using such a strategy demands more time and patience in communication compared to using an instantaneous translation.

According to the principles of the MA method, the responsibility to plan meetings lies with students. However, in this project mentors sometimes took on the responsibility. They could send pictures to students, and suggest what to talk about. In their evaluation, some students asked for more tools that could help mentors in their work:

• 8) I would suggest that the mentor get some aids in the beginning, for better achievement and so the mentor can control the conversations to accomplish those goals. These could be: Phrases for everyday use, different themes: colours, numbers, nature, animals, etc. If such a menu were available, mentor and student would be safer. (Questionnaire)

This reflection demonstrates a request for a learning procedure which could be followed. The excerpt also reveals that the mentor is thought to have a similar role as a teacher: leading and controlling learning activities. Thus, learning with a mentor resembles language learning in a traditional classroom.

In contrast, informal learning in real-life situations cannot follow a structured form, where everything is planned beforehand. The MA method includes features both from non-formal and informal learning, and some students focused on the non-formal side of the method, while others felt that following fixed procedures became too rigid. Students also expressed that they were able to make changes if necessary:

• 9) When something didn't work, however; it was up to us, so we changed course along the way. E.g. that it was not 100% effective for learning a language to go on a trip. (Questionnaire)

This excerpt shows that the student understood they were in charge of the meetings, and when something did not work, changes needed to be made. It also demonstrates a metacognitive ability to evaluate which activities promoted learning and which did not, and an agency to make a change in situations which did not work.

How students evaluated their oral ability

In the questionnaire, the students evaluated their oral ability in Kven using the scale of the CEF, where oral ability is evaluated using three groups A, B and C, each divided into two levels. Levels A1 and A2 refer to basic, levels B 1 and B2 to independent, and levels C1 and C2 indicate proficient language user. Students received a description of all these levels, and were asked to place themselves on the appropriate level both before and after the mentor project, in terms of their proficiency in listening, oral interaction and oral production.

In Table 1, only the changes are presented. A student’s change from e.g. A1 to A2 gets a numeric value 1, and a change e.g. from A1 to B1 gets a numeric value 2. It was decided to not refer to the different levels in the CEF to be able to keep the anonymity of the informants. Because of the low number and the heterogeneity of the informants, it could be possible to identify them just on the basis of their self-evaluation with regard to the CEF. Instead, information on how many hours they met with their mentors is given in the table.

Table 1. Evaluation of oral ability before and after the mentor project.

Informant	Time spent with mentor	Oral interaction	Oral production	Listening
1.	80 hours	2	2	2
2.	ca. 60 hours	1	1	2+
3.	56 hours	1	1	2+
4.	50 hours	1	1	-
5.	40 hours	1	1.5	1.5
6.	27 hours	-	-	-

The table demonstrates that all students save one claimed that the mentor project had a positive impact on their oral ability. It is also possible to make some connection between the time spent in meetings with mentors and the development of oral ability, even though this is not so clear among all the students.

The students were asked to verbalize their oral development. The positive effect of the MA method project is shown in their evaluations:

• 10) Earlier I was just able to answer using only one word, now I can tell something. (Questionnaire).

Moreover, the students make special mention of vocabulary learning and better understanding. They describe increased oral competence: it became easier to retrieve words, to memorize them, and they even became used to fast speech. All these comments demonstrate processes of internalization in language learning (Lantolf, Thorne, and Poehner 2015, 208–212), because the students became more independent about expressing themselves in Kven during the use of the MA method.

How students evaluated the MA method project

This section will introduce the answers to the questions connected to specific topics, such as the number of hours used for the mentor meetings. It will also include a more general evaluation of the project by the students and a discussion on how well the project promoted the use of the Kven language in Kven communities.

All the students could use up to 100 hours in meetings with their mentor during the academic year; the actual hours used were between 80 and 27 (Table 1.) There are several explanations for the low number of hours. Most students were in full time work or studies beside the mentor project; in addition, they were also students in Kven studies. It was not easy to arrange regular meetings when the mentor and the student did not live near each other or not even in the same region. To arrange meetings by telephone was not successful either, though contact by Skype worked better. It was most effective to have a mentor who lived in the local area. Otherwise, there were too many and too long gaps between meetings, and what had already learned became forgotten. Still, when evaluating the optimal number of hours to be used during an academic year, some students felt that 100 hours was still a possibility. Meetings with mentors combined with more intensive informal learning; e.g. in the form of a language camp was also suggested.

The overall evaluation of the project by students was positive. All those who completed the questionnaire felt that the UiT ought to use this method in future. The practice of oral language with the mentor increased motivation in Kven studies. One student wrote that the mentor project was their main reason for continuing their studies in Kven.

The use of the Kven language expanded from meetings with mentors to be used in other contexts:

• 11) There was a positive atmosphere [in the conversation] at the table, and I got some help with words when I was stuck. This is how I was more motivated to chat and make contact with other Kven people in the café the next day and exchange some everyday phrases with them. Likewise, when I was visiting the nursing home in the next few days, I got in touch with old Kvens. I understood my interlocutors well and felt that the vocabulary and pronunciation did not reveal that I was still not so competent in Kven. (Logs)

The log shows the Vygotskyan concept ZPD at work: a positive experience when collaborating with others using Kven improved self-confidence, and it fostered independence to use Kven. Positive feedback from others helped the learner to overcome fear of not being competent enough when producing Kven. The log demonstrates learner-agency because the student started to use Kven independently; in addition, the ability to control one’s emotions when using the TL seemed to be involved in the development towards autonomy. Even though the performance in Kven did not change, the autonomy to use Kven increased.

The oral use of Kven extended outside the project into other contexts in Kven communities. One example is the new language café in Kven which started up as a direct result of the MA method project. The project also helped increase use of Kven at the homes of some students, for example when other Kvens paid a visit during a meeting.

Discussion

This section will focus on the significance of the results for language learning. Did the students advance their oral language proficiency? Were the principles of the MA method followed? The results are compared with research on the use of the MA method elsewhere; moreover, the results are studied in the light of sociocultural theory. Two questions are then posed: whether learner autonomy increased when using the MA method, and if it is possible to create new speakers using the MA method.

The first research question is whether the MA method helped students to improve their oral ability in Kven. Most students felt they had advanced in all three competence areas of oral language use: interaction, production and listening. In their evaluation, they focused on vocabulary learning and listening. Repetition is mentioned as a strategy for learning new words. The MA method also helped students to realize the importance of listening as a crucial part of oral competence. Actually, listening was the part of oral competence that developed more compared to other parts of oral competence during the project (Table 1).

The second research question is about how autonomy in language learning developed while using the MA method. The students described how practising with their mentor helped them to develop their understanding Kven more easily, reactivating receptive knowledge, and learning more vocabulary and other language elements. Students described how mentors gave them words and phrases when they got stuck. Gradually, they became more independent language users. So, the students’ narratives described the process of internalization, as other-regulation decreased, but self-regulation increased (Lantolf and Thorne 2006).

A basic principle in the MA method is not to use the majority language at all; still, it turned out that Norwegian was used when discussions stalled. Yet, the use of a majority language instead of the TL also happens in other MA method projects (Hinton et al. 2018, 131). Communicating in an endangered language often demands special effort; for instance because of the lack of vocabulary applicable to modern domains. Accordingly, it becomes demanding to keep the conversation going; so, the easiest and most available communication strategy is translation.

Both concrete and semiotic resources like texts and documentary films were in use to facilitate conversations in Kven. Pictures were often used and were especially helpful for beginners, as visualizing helped them understand more easily what the conversation was about. But one of the most important motivations for language learners is to be able to use the TL in real-life situations. Language learning connected to practical activities is one essential way of learning according to the MA method (Hinton et al. 2018, 130). Such learning resembles more informal learning than non-formal learning: the focus is on activities, and a language is learned in “a joint attention frame” (Tomasello 2003, 26). Therefore, communication in Kven while participating in practical activities could have increased proper communication. The fact that practical activities were not included in meetings illustrates that beliefs about language learning (Benson 2007, 28) possibly dominated the choice of activities. Accordingly, an activity like cooking a meal while learning a language does not feel like study, whereas talking about pictures feels more like a language learning occupation, as such activities are frequently used in classrooms, while participating in concrete activities is seldom found in formal learning settings. Still, even the choice of meeting places could dictate the occurrence of any activities. When meetings were arranged in libraries or language cafés, practical activities were not possible. Indeed, some students commented that concrete, practical situations should have taken place and that meeting places should have been more varied. Therefore, students demonstrate that they knew about this important principle of the MA method, though it was not put into practice.

The students’ narratives give the impression that well-known activities connected to formal language learning contexts were often those that took over. In formal contexts, language is often seen as an object of study and not always as a medium of communication. One impact from the formal language learning context was also evident when a mentor took over the responsibility of organizing the meetings. However, students’ narratives also show learning-agency when they took the initiative to expand the learning resources, or they demonstrated metacognitive ability to organize their learning better. Moreover, many learning strategies for increasing autonomy in language learning can be found in the students’ narratives.

The project analysis provides examples of how the use of Kven expanded from mentor meetings into other contexts; the third research question. We can see that mentor meetings activated the use of Kven in local Kven communities. The overall research question is whether the MA method was helpful while revitalizing Kven. It is connected to the question of whether it is possible to create new speakers of Kven using the MA method. This question is a crucial one for the revitalization of Kven and depends, of course, on what the criteria are for becoming a new speaker of the language.

If the criteria for a new speaker are that one must reach a communicative level in the TL, the number of hours spent with the mentor sets limits to this goal. In other programmes, to reach a medium-to-high level of communication in the TL takes at least three years, with at least 300 hours per year spent with the mentor (Hinton et al. 2018, 129). In this case, the number of hours spent with the Kven mentors was low. Even the student who used most hours in the project (80 hours, Table 1) spent fewer hours compared to those in the programmes in California. So, it must be concluded that the beginners did not have the chance to reach the best communicative levels in Kven in the course of one year. On the other hand, those with an earlier knowledge of Kven, or of Finnish, and who were able to communicate in Kven from the very beginning would have been able to become new speakers. Some of them even wrote logs in Kven. Therefore, more mentor meetings are needed to lift beginners onto a higher communicative level in Kven. It is worth noting that after the project was completed, those students who continued their Kven studies continued to have meetings with their mentors.

Communication, rather than a perfect mastering of the TL, seems to be important in other programmes where the MA method is used (Hornsby 2015, 138; Hinton et al. 2018, 135). To become a legitimate speaker of Kven means that the language community accepts one’s language use even though it is still different to that of mother-tongue speakers (Jaffe 2015). It seems that the use of the MA method helped some of the students to be accepted as a new speaker among other Kvens. This also illustrates how collaborative learning works (Lantolf and Thorne 2006): after practising the language together with the mentor, students were able to use Kven independently in other contexts. The use of the MA method helped the new speakers to communicate in Kven without a feeling of not being good enough, as interlocutors did not pay any attention to shortcomings; instead, they praised the speaker for using the language.

However, in many cases reaching a communicative level in the TL can still be too demanding a task. Hornsby (2015, 139–149) discusses endangered languages as “postvernacular” languages, which refer to languages that have lost their communicative functions. Postvernacular use of language includes symbolic expressions for identity, i.e. the use of words and phrases of an endangered language when speaking the majority language. Music, theatre or other types of performances expressing identity are also examples of postvernacular use. Vernacular and postvernacular practices don’t have to be mutually exclusive; still, the practice of postvernacular expands the possibilities of becoming a new speaker.

Conclusion

This article focused on the development of autonomy in language learning while participating in the use of the MA method, a method developed to contribute to language revitalization and to create new speakers of endangered languages. So far, new speakers have been mostly studied from the sociolinguistic perspective. More research on developing learner autonomy in language revitalization contexts is needed. This is because in such a context, learners are often engaged, drawing on strong motivation and identification with the language they are learning. These are dispositions closely associated with learner autonomy.

Not many mother-tongue speakers of Kven are left today; instead, a category of new speakers of Kven is becoming established. New speakers of Kven have a diversity of backgrounds. Some have Kven as a heritage language, others have other language backgrounds, most often Finnish. The heterogeneous backgrounds of new speakers in Kven will have an impact on the development of the Kven language in future.

Disclosure statement

No potential conflict of interest was reported by the author(s).