Motivation and the Support of Significant Others across Language Learning Contexts

,

The motivation to learn a new language is generally recognized to be as critical to the successful acquisition of a language as other psychological factors, such as language aptitude (Gardner, 1985).Because of its relative malleability, motivation is of particular interest to language teachers, administrators, and other people who wish to help the learner succeed in the learning process.Given that motivation can be influenced by people in the learners' social context, it is useful to understand how those who play important roles in language learners' lives can affect learners' motivation.
Although there was considerable interest in how the social milieu influences learners' motivation in the early years of research on motivation (see Gardner, 1985, for review), over the years, such studies have declined 1 , with perhaps the exception of how teachers' can motivate learners in formal classroom settings (Guilloteaux & Dörnyei, 2008;Mercer & Kostoulas, 2018).As a result, there is little recent research about the people who encourage or discourage learners' motivation (other than the teacher).Moreover, even less research has adopted a comparative perspective to understand how motivational support might vary depending upon the language learning context.
In light of this gap, this study explores how people, including teachers, peers, family members, and members of the target language (TL) community, support and/or undermine the motivation of language learners living in different types of social milieux, including heritage language (HL), modern language (ML), and English as a second language (ESL) contexts.To this end, we outline the theoretical lens that framed our research, Self-Determination Theory (SDT), and we then review existing research on the motivational support from teachers, peers, family members, and members of the TL community across these different language learning contexts.

Motivation and Self-Determination Theory
At its root, motivation concerns the "why" and "how" of action, including its initiation, direction, intensity, persistence, and the quality of engagement in that behaviour.
There are many metatheoretical and theoretical lenses through which these questions can be answered (see Ryan, 2012, for theoretical overviews); for the present study, we examine these questions with reference to Self-Determination Theory (SDT; Deci & Ryan, 1985;Ryan & Deci, 2017).SDT offers a comprehensive, coherent, and useful framework for understanding 1 The reasons for this decline in interest in the social milieu outside the classroom are manifold, but this trend corresponds with a shift to research focused on the self-dynamics of learners of English, primarily in classroom settings (see McEown, Noels, & Chaffee, 2014;Boo, Dörnyei, & Ryan, 2015).motivation, which has been empirically validated in many life domains, including language learning (see Noels, 2009;Noels, Chaffee, Lou, & Dincer, 2016;and Noels et al., in press, for review).Germane to the present study, SDT makes specific claims about the sociocommunicative and socio-psychological processes through which significant others affect learners' motivation.
SDT is grounded in organismic and humanistic psychology, which maintain that humans and other living creatures have an inherent tendency to curiously explore and develop their capacities to engage effectively in their social and physical worlds, and integrate the information acquired through this interaction into their meaning systems.Human development, then, occurs in a dialectical relationship with the environment, and involves an ongoing process of assimilation of novel information and accommodation of extant knowledge, such that over time and in optimal conditions, the self becomes increasingly elaborated, coordinated, and cohesive.Drawing from humanism, SDT maintains that the self is the central organizing principle for people's thoughts, feelings, and behaviours, and that the "well-lived" life comes about when people are true to their authentic self and live personally meaningful lives.
The pursuit of an activity because it is personally interesting (i.e., novel, challenging, meaningful) and satisfying (i.e., "fun") is referred to as intrinsic motivation.Individuals who are intrinsically motivated freely choose to engage in an activity, such as learning another language, because it stimulates their curiosity, offers the potential to develop mastery, and is associated with a sense of well-being and flourishing (termed "eudaimonia").People who are intrinsically motivated do not require any external incentive or pressure to perform this activity; in fact, externally imposed rewards may undermine intrinsic motivation because they constrain personal autonomy (Deci, Koestner, & Ryan, 1999).
In contrast to intrinsic motivation, extrinsic motivation refers to motivation that is not regulated by inherent interest and enjoyment in the activity, but instead for reasons external to that activity; however, these factors may be more or less internal to the self.SDT maintains that these reasons reflect forms of behavioural regulation that are more or less governed, or determined, by the self (Deci & Ryan, 1985, 1995;Grolnick, Deci & Ryan, 1997;Noels 2001a;Ryan & Deci, 2000).The notion of "forms of regulation" corresponds with the term "orientations" that is familiar to language learning scholars and teachers (Gardner, 1985).Thus, a given set of reasons reflects the locus of causality for the regulation of behaviour (i.e., whether the self or others "cause" the learners' engagement; Ryan & Deci, 2017), and describes the perspective that frames a learners' efforts for learning a language.
The least self-determined type of extrinsic motivation is external regulation, in which circumstances or people apart from the learner determine his/her behavior.For example, a student may be motivated to get a high grade in their language course because of the promise of a financial reward from their parents.Rewards and punishments are not solely administered by other people; people can regulate themselves in a similar manner through the unconscious internalization of others' expectations and attitudes, which is termed introjected regulation.
People who learn a language because of these introjects may work towards learning a language because they feel they ought to speak the language to be a good person, and would feel ashamed if they did not.Learners who regulate their learning through introjection are not acting voluntarily in spite of the fact that these pressures are internalized (Ryan & Deci, 2017).
In contrast with these more controlled forms of regulation, two forms of autonomous regulation are posited.Identified regulation is more internalized and self-determined than introjected regulation, and characterizes the motivational orientation of an individual who learns a language because they personally have identified it as important in order to accomplish things that they value.For instance, a person who learns Latin because it will facilitate their career as an archeologist, is autonomously regulating their actions.Eventually, learning and using another language may not just be considered important for achieving a valued aspiration, but it may become integrated into one's self-concept.The most autonomous form of extrinsic motivation, integrated regulation, describes the motivational orientation of a person who feels the language is an important part of their identity, and learning and using the language is a part of their self-expression.
Intrinsic motivation and self-determined extrinsic motivation are argued to be more sustainable than extrinsic motivation because they are not reliant on external forces to regulate one's behavior; when those external forces are no longer present, motivation is likely to dissipate.As well, self-determination is linked to well-being, thriving, and selfactualization, which a condition which sustains active engagement (Reeve, 2012).
Considerable research demonstrates that people who are self-determined experience the greatest degree of active engagement and sustained success in language learning (see Noels et al., in press for review).

Supporting Learners' Basic Psychological Needs
SDT assumes that humans have fundamental psychological needs that must be satisfied in order for them to behave in a self-motivated manner (Ryan & Deci, 2017).These needs include autonomy, competence, and relatedness.In order for individuals to feel autonomous, they must perceive that their behaviour is not being controlled by an outside source.To feel competent, they must be able to effectively deal with their environment and overcome new challenges.A sense of relatedness is fostered when the individual has the benefit of secure and satisfying relationships with other individuals and the larger social whole.
Other people in the learner's social milieu can help or undermine the fulfillment of these needs.A "significant other" is a person who is important to the learner's self-esteem and well-being, and has an impact on his or her behaviour (e.g., spouse, friend, teacher) (Sullivan, 1953;Woelfel & Haller, 1971).Significant others can support a learner's autonomy by acknowledging his/her perspective, and providing opportunities for him/her to act in an autonomous and personally meaningful manner.They can support competence by providing a useful and effective structure and clear expectations and feedback that help develop mastery.
They can support a sense of relatedness by conveying care and concern for the learner, which communication researchers refer to as a sense of interpersonal immediacy (Hosek, Houser, & Richmond, 2017;Wiener & Mehrabian, 1968 ).Significant others do not always support learners; they can also thwart learners' self-determination by being controlling, failing to provide relevant information, and/or being disinterested or unkind towards the learner.Close relationships are not necessarily solely positive or negative, and it is not uncommon for relationships to be ambivalent (Zayas & Shoda, 2015).As a result a learner's motivation can fluctuate even with reference to the same person.
A variety of individuals have been posited to influence language learners' motivation.
In this study, we consider teachers, peers, family members, and members of the L2 community (Palfreyman, 2011).
Teachers.It seems intuitive that teachers would play a large role in either nurturing or hindering the motivation of learners registered in formal language courses, and a growing body of literature supports this claim (Lamb, 2017;Mercer & Kostoulas, 2018).Following the tenets of the Socio-Educational Model, Gardner and his colleagues (for summaries, see Gardner, 1985;2010) demonstrated that if students hold positive attitudes towards their language teacher, they will be more motivated to learn the language and experience higher levels of achievement.Based on discussions with teachers coupled with personal experience, Dörnyei and Csizér (1998;Dörnyei, 1994;2001) developed a list of motivational strategies for teachers, eventually narrowing them down through empirical research to the "10 commandments of language teaching" that can maintain students' motivation (Cheng & Dörnyei, 2007;Guilloteaux & Dörnyei, 2008).
Other research has adopted a theory-driven approach to articulating aspects of teachers' socio-communicative style that either enhance or inhibit self-determination.Guided by SDT, Noels and her colleagues found that perceptions of a teacher using controlling techniques, such as threats or deadlines, instead of autonomy-supportive strategies, such as allowing students to choose their own learning activities, were related to more selfdetermined orientations and greater motivational intensity (Chaffee, Noels, & McEown, 2014;Dincer, Yeşilyurt, Noels, & Vargas Lascano, 2019;Noels, Clément, & Pelletier, 1999;Noels, Stephan, & Saumure, 2007;Noels, 2001bNoels, , 2005)).Informational feedback that provides structure, expectations and information about how to improve competence, has also been shown to sustain greater self-determination (Noels, 2001b;Noels et al., 2007;McEown et al., 2014); in contrast, vague admonitions for improvement are associated with feelings of incompetence, and thereby lessen self-determination and motivational intensity.Teachers who are perceived as warm and responsive to the learners' needs tend to nurture students who have greater self-determined motivation (McEown et al., 2014).
Friends/Classmates.Compared to teachers, less research has examined how friends and classmates encourage learners' motivation.Some studies assessed general L2 classroom climate, which refers to an overall sense of connection between members of the class.For instance, Clément, Dörnyei, and Noels (1994) found that students' appraisal of the classroom environment, particularly cohesion among classmates, was an important predictor of foreign language behaviour and competence.In a complementary manner, Dörnyei (2001) notes that a general lack of classroom motivation can be traced to fears of isolation or rejection from peers.
Some studies have shown little interpersonal influence among classmates.For instance, Kyricao and Zhu (2008) reported that Chinese senior high school students, on average, did not feel that their level of motivation to learn English was influenced by their peers (see also Csizér & Kormos, 2008).However, other studies do indicate that classmates and friends can encourage or discourage motivation.American Masters-level students in a Spanish immersion course claimed that having peers who are also learning the TL was beneficial in developing their own competence (Liskin-Gasparro, 1998; see also Csizér & Dörnyei, 2005).Bartram (2006) found that agreements with friends to study a foreign language helped participants to maintain motivation.Murphey (1997) details how peer role models who are from the same culture as the student, around the student's age, and successful in English positively influence Japanese students' English motivation.One study from an SDT perspective addressed whether peers in the language classroom affect learners' motivation, and found that a positive classroom social climate, including help from peers, led to the fulfillment of the needs for autonomy, competence, and relatedness, which are necessary for self-determined motivation (Joe, Hiver, & Al-Hoorie, 2017).
Family Members.The majority of research pertaining to the influence of family members on language learning relates primarily to children and/or HL learners.Research by Gardner and his colleagues (see Gardner, 1985, for summary) argued that parents can play both active and passive roles in their offsprings' language development, including, respectively, direct encouragement and modeling attitudes towards both language learning and the TL community (see also Cortés, 2002).Gardner, Masgoret, and Tremblay (1999) found that adult students who could recall parental encouragement towards L2 learning were more likely (particularly in early years) to have higher levels of motivational intensity.
Parental influence might wane as students age; Sung and Padilla (1998) found that parents were less important for the motivation of high school students compared to younger students.
That is not to say that parental support becomes irrelevant.In their study of 1,000 Malaysian, university-level learners of English, Vijchulata and Lee (1985) found a significant positive correlation between parental encouragement and students' motivational intensity.
Although parents might have the strongest familial influence on L2 learning, other family members can also play a role.Studies of children from minority ethnolinguistic families show mixed effects for the impact of siblings on second language development: some studies indicate that first-born siblings have better second language skills, presumably due to a dilution of resources as family size increases (e.g., Keller, Troesch, & Grob, 2015), but others suggest that later-born siblings might benefit from older siblings' bilingualism, through modeling, practice, and direct instruction (Bridges & Hoff, 2014).Spouses and their extended family have also been implicated in learners' motivation among couples with different ethnolinguistic backgrounds (Noels, 2013).
Family members are obviously implicated in the HL learning context.HL learners may be encouraged to study their ancestral language so that they can communicate with parents, grandparents and extended family members who use that language (Park & Sarkar, 2007;Melo-Pfeifer, 2015;Mu & Dooley, 2015), travel to the country of ancestral origin (Noels & Clément, 1989), and/or develop a cosmopolitan worldview (Guardado, 2010).
Motivational conflicts may arise when family members do not situate the HL learner as having a legitimate claim as a member of the TL community (Dressler, 2010).

Members of the L2 Community.
Considerable research supports the idea that a student's proficiency and attitude toward language learning is influenced by the support that they receive from the TL group.Research grounded in Clément's socio-contextual model of language learning shows that more frequent and better quality of contact with members of the TL community predicts linguistic self-confidence, motivational intensity, and language proficiency (Clément, 1980(Clément, , 1986)).In his study of students learning Spanish in Mexico, Yager (1998) learned that more contact with Spanish-speaking peers in informal settings was related to increased proficiency in the Spanish.Genesee, Rogers, and Holobow (1983) found that both achievement and the use of the second language depended upon the level of motivational support that the learner expected from members of the language community.In their study of English speakers who were learning French, perceptions of support from the French community were more predictive of proficiency, use, and social affiliation than was a student's own motivational orientation.

The Language Learning Context
The language learning context refers to the societal context within which learning takes place.This level of analysis focuses largely on the socio-political dynamics between ethnolinguistic groups, rather than the interpersonal and group dynamics within the classroom (i.e., between students and/or students and their teacher(s)), although broader societal dynamics can influence those classroom dynamics as well (Noels et al., in press).At least two characteristics define the learning context.The first is the relative socio-structural status, or "ethnolinguistic vitality", of the learners' own ethnolinguistic group compared to that of the TL group (Clément & Kruidenier, 1983;Landry, Allard, & Deveaux, 2009;Noels et al, in press).The second is the opportunity for interpersonal contact with TL speakers (whether face-to-face or mediated) in public and private domains (e.g., family, friends, school/work, community).
Since the inception of the field of language learning motivation, scholars have highlighted that language learning motivation is influenced by the relative ethnolinguistic vitality of the groups under consideration (e.g., Clément, 1980;Gardner & Lambert, 1972;Giles & Byrne, 1982;Landry et al., 2009;Noels, 2001a).Ethnolinguistic vitality refers to the dimensions that underpin the relative socio-structural status of ethnolinguistic groups (Ehala, 2015;Giles et al., 1977).It includes the demographic representation, prestige, and institutional support to which an ethnolinguistic group has access.For members of a lower vitality (or "minority") group, learning the language of the higher vitality (or "majority") group might result in losing the original language and culture, and becoming assimilated into that group (i.e., subtractive bilingualism; see Lambert, 1978).For members of the higher vitality group, learning the language of the minority is unlikely to upset their relatively higher status and vitality; learning the language can add advantages, including integration to the minority language group (i.e., additive bilingualism).
The opportunity for contact depends on the availability of members of the TL group for face-to-face interaction.This availability depends on the number of people in the learner's social network who speak the TL, the frequency/regularity with which they interact, and the quality (or intimacy/amicability) of those interactions.The important difference in opportunities for intergroup contact is illustrated by the distinction between English-as-asecond-language (ESL) vs. English-as-a-foreign-language (EFL) learners, with the former living in Anglophone societies, and the latter generally living in non-Anglophone societies.
However, even in nations that have large representations of TL speakers, learners may not have the opportunity to interact with them, because they live in different regions2 or are segregated into different communities.
Contact with speakers of other languages is a situated phenomenon; we interact with others in specific social situations, some of which are relatively intimate and/or informal (e.g., with friends and family) and others that are nonintimate and/or formal (e.g., school, work, or in public).For many language learners, intercultural communication is likely to take place in nonintimate situations before it takes place (if ever) in intimate situations (Noels, Clément, & Gaudet, 2004).HL learners, however, are more likely to encounter native speakers in private domains, particularly with family members.Moreover, their ancestral bonds with the ethnolinguistic community, most often formed in earliest childhood, are more likely to be integrated into their identity (Noels, 2005;Comanaru & Noels, 2009).Thus, it is important to differentiate HL from other ML learners, as their self-motivation and opportunities for intercultural contact are likely to involve different interpersonal and intergroup dynamics.

Objectives of the Present Study
The present study has three main objectives.First, it examines language learners' reasons for learning a language to determine the extent to which they can be described following the motivational orientations (i.e., forms of regulation) outlined by SDT.Second, it considers what learners have to say about how significant others can encourage or discourage their motivation, and how well learners' descriptions of positive and negative support from significant others correspond with SDT's notions of autonomy-support/control, informational/negative feedback, and interpersonal immediacy/distance.Third, because the learning context can affect learners' motivation and the people who support (or not) their motivation, we examine these patterns across three language learning contexts, including ESL, HL, and ML contexts.In the province in which this study was conducted, the ESL context involves the acquisition of a high-vitality language (locally and internationally) in a location where there is ample opportunity to interact with English speakers, particularly in public domains.The HL context involves the acquisition of a language of lower vitality than English locally, with the opportunity for interaction with native speakers, particularly in private domains.The ML context involves the acquisition of a lower status language than English locally, and little opportunity to interact with native speakers.

Participants
One hundred and four people registered in language courses at a mid-western Canadian university agreed to participate in the study.These included 34 HL learners studying primarily French (17.6%),German (11.8%), and Cree (8.8%); 34 ML learners studying primarily French (38.2%),Spanish (17.6%), and Mandarin (14.7%); and 36 ESL learners, who originated from seven countries, including China (36.1%), Germany (13.9%), Iran (13.9%), and Korea (8.3%).The HL and ML students were mostly registered in language-focused courses (i.e., not literature or cultural studies courses) open to students regardless of their heritage-language status (i.e., none were enrolled in courses specifically designed for HL students).The ESL group consisted primarily of international students who were enrolled in an English-focused preparatory course in order to be admitted into the university.
A summary of respondents' demographic information is presented in Table 1.The groups did not differ in age (F(2,100) = 0.28; p = .759;partial eta 2 = 0.005), and although there were more females in the HL group than the ML and ESL groups, a chi-squared analysis indicated this distribution across groups was not statistically significant (χ 2 (2) = 4.79, p = .091).The sample represented students across educational levels: 10 HL learners, 14 ML learners, and 16 ESL learners (38.5%) were registered in graduate school, whereas 17 HL learners, 16 ML learners, and 13 ESL (44.2%) learners were enrolled in undergraduate programs.The mean age at which participants started learning the language did not differ between learner groups (F(2,71) = 1.71; p = .188;partial eta 2 =.046), nor did the average years spent learning the language (F(2,100) = 1.91; p = .154;partial eta 2 = .037).The groups did differ in their self-evaluation of linguistic competence (i.e., the mean of self-ratings of the ability to read, write, speak, and understand the TL; (F(2,101) = 8.23; p = .000;partial eta 2 = 0.140), such that the ESL learners rated themselves significantly more highly than did the HL learners, and the ML learners rated themselves midway between these two groups and did not differ significantly from either.

Materials and procedure
We adopted a "focussed essay" technique (MacIntyre & Gardner, 1991) to elicit the participants' motivational orientations and perceptions of support.A nine-item questionnaire was used to obtain qualitative information regarding respondents' reasons for learning the language, and how significant others had encouraged or discouraged this learning.Depending on the group, the participants were asked "What are your reasons for learning English/your second language/your heritage language?".This question was followed by eight questions regarding the extent to which other people encouraged or discouraged their language learning.The writing prompt for encouraging behaviour was "How has [significant other inserted here] helped you to learn your second language?Please describe your experience in detail, including some of the most helpful things they have done.".The writing prompt for discouraging behaviour was "Have you ever felt discouraged learning the language because of something [significant other] did?What happened?Why did you feel discouraged?"The significant others included; "your language teachers"; "your friends or classmates"; "your family members (e.g., parents, spouse, and/or children)"; and "members of the English/second/heritage language community".For each question, participants were given a full page (i.e., Letter (US)) to write a thorough response.The questionnaire also elicited demographic information.All participants wrote their answers in English.
Poster advertisements informed language students of the study and invited them to a designated location to fill out a questionnaire.All questionnaires were completed in a small group setting, with a research assistant available to answer questions.The respondents could take as long as they wished; most took less than one hour.After completing the questionnaire, participants were thanked, and given $10.00 CAD as a token of appreciation.The study protocol was approved by the university's research ethics board, and conformed Canadian government's Tri-Council Policy and the American and Canadian Psychological Associations' ethical guidelines.

Analytic strategy
Three independent coders coded the content of the qualitative data generated by the nine questions.The responses to the question regarding reasons for learning a language were coded as present or absent for each of SDT's five subtypes of intrinsic and extrinsic motivation (external, introjected, identified, and integrated regulation) and amotivation.Some reasons, although well articulated, could not be readily categorized within the SDT framework.For this set of responses, alternative themes were developed if multiple people (circa five people or greater) mentioned the same idea.For example, one "other" category, termed "integrative orientation" (Gardner, 1985), was used when it was not clear whether social integration into other language communities was desired for more or less selfdetermined reasons (e.g., "to communicate with people").Not all of the responses fit into the SDT or "Other" categories: some were coded as unclear or irrelevant, and these were not included in the data analyses.
It should be noted that respondents could have multiple reasons for learning a language, and these reasons could be coded in different categories (e.g., a respondent could provide a reason that was coded as intrinsic motivation and another reason that was coded as external regulation).If a respondent provided multiple reasons that were from the same category (e.g., multiple reasons coded as intrinsic motivation), they received only 1 "point" for that category.
Responses to the questions about the influence of significant others' sociocommunicative behaviour on respondents' motivation were analyzed in a similar manner.With regard to how significant others encouraged learners, the categories of "autonomy-support", "informational feedback", and "interpersonal immediacy" were used.With regard to how significant others discouraged learners the categories of "control", "negative feedback", and "interpersonal distance" were employed.When the socio-communicative behaviour could not be identified using the SDT framework, it was classified as "other"; if multiple people reported similar "other" responses (circa five people or greater), an alternate theme was articulated.
At least two people coded the data or subsections of the data.Cohen's kappa was used to estimate inter-rater reliability between coders.Initial kappa coefficients ranged from .28 -.75, indicating minimal to moderate levels of agreement (McHugh, 2012).The majority of the discrepancies could be readily resolved following clarification of the constructs by the first author and through discussion among the coders.Following revisions, kappa was calculated a second time and ranged from .76 -.98, indicating moderate to almost perfect consistency in the raters' coding.
Once the data were reliably coded, separate chi-squared (χ 2 ) analyses were conducted for each category to determine whether the distribution of valid responses differed depending upon the learner group.In the case of a significant χ 2 , follow-up analyses examined the standardized residuals for the distribution of responses across cells.A summary of these results is provided in Tables 3-7.The complete set of coded responses are available at http://sites.psych.ualberta.ca/IClab/research/past-projects/.

Reasons for learning another language
The results of the content analysis of why one was learning another language showed that 73.0% of the 241 responses could be categorized within the SDT framework (see Figure 1 and Table 2 for a summary of the results).Note: Percentages sum to ~100% within each language group.Some "other" responses could be grouped into an alternate (not clearly SDT-related) theme.These included responses that either reflected the integrative orientation, that is, a desire to interact with the TL community (e.g., "Enables me to understand Canadian culture and people and to communicate with people"; "Trip/travel to Ireland"), or indicated that the language (usually English) was widely used ("First, it is the most popular language in the world"; "Nowadays, the world is connected by English, so we should study English right now!").Other miscellaneous responses were more ambiguous (e.g., "Since it is offered in my country, I studied it").Orient.

World
Lang.

% of Learners
Motivation Type HL ESL ML Overall, external and identified regulation were most frequently cited orientations (cited by approximately half of all respondents), followed by intrinsic motivation and the integrative orientation (cited by approximately a third of all respondents).Only a very small percentage reported integrated or introjected regulation, and no one expressed amotivation.
The χ 2 tests indicated that some reasons differed depending on the language learning context.
Integrated regulation was mentioned more frequently by HL learners and less frequently by ESL learners than would be expected by chance, and external regulation was mentioned more frequently by ESL learners and less frequently by HL learners.ML learners reported intrinsic motivation marginally more frequently, and ESL students marginally less frequently than would be expected by chance.ESL respondents were significantly more likely to comment on the popularity of the language than the other groups.

Encouragement and discouragement from significant others
Examples of responses reflecting the SDT forms of encouragement and discouragement are presented in Table 3. Examples of "other" responses are incorporated into the following discussion.Teachers.Of the 164 valid responses to the question regarding how the teacher encourages learners, 70.7% could be categorized within the SDT framework, and the remainder were classified as "other" (see Table 4).The "other" responses included several reports that when teachers used authentic materials, it encouraged their interest and engagement (e.g., "listen to French songs, movies, had debates, etc."; "In class, listening to radio, reading journals or books, watching movies, playing games, etc.").The most common forms of encouragement were "other" and informational feedback.The χ 2 analyses comparing the distribution of responses across the four substantive categories (i.e., autonomy-support, informational feedback, immediacy, and other) showed that some types of teacher encouragement differed depending on the learning context.These differences in distribution, however, were small.For example, informational feedback from teachers was reported more often by ESL learners and less often by HL and ML learners, but the standardized residuals indicated that the proportions did not exceed chance expectations.
Likewise, "other" types of encouragement were reported more frequently by ML learners and less frequently by ESL learners, but not at rates greater than chance.
Regarding discouragement, more than half (63.8%) of the 69 valid comments could be categorized within the SDT framework (see Table 4).There were two alternate themes regarding how the teacher discouraged students, which were defined as "poor speaking skills" and "poor teaching skills" (e.g., "All the teachers I had didn't speak English very well" and "She doesn't know how to teach.She always reads the name of the person who got the high or low mark", respectively).The most common form of discouragement was "other", followed by negative feedback and control.There were no significant differences in the pattern of responses across learner groups.
Friends/classmates.Approximately two-thirds (67.6%) of the 136 valid responses to the question regarding how friends and classmates encourage learners could be categorized within the SDT framework (see Table 5).The alternate themes included "doing fun activities with the learner" (e.g., "My friends took me shopping"), and "learner plays the role of teacher" (e.g., "Some have even asked me to teach them some phrases which encourages me").The most commonly reported form of encouragement was informative feedback, followed by "other" forms of encouragement.Although χ 2 analyses showed overall group differences in informative feedback, the standardized residuals indicated the differences in proportions were not large.There was a greater tendency for ESL learners to say they received informative feedback from their friends and classmates, and a weaker tendency for HL learners to do so.There were no other statistically significant differences between the learner groups.Approximately three-quarters (73.9%) of the 69 responses to the question regarding how peers discourage learners could be categorized within the SDT framework (see Table 5).
"Other" responses primarily focused on instances in which the learner felt that their peers had greater skill than they did (e.g., "They make me anxious or unconfident when their language skills are much more advanced than mine").The most commonly cited form of discouragement was negative feedback followed by "other" discouragement.Discouraging experiences involving peers did not differ between learner groups.
Family members.Half (55.5%) of the 146 valid responses regarding how family members encourage learners could be categorized within the SDT framework (see Table 6).Within the "other" category, the alternate themes included "providing material/financial support" (e.g., "My parents help me financially"; My brother gave me a book"); "facilitating opportunities to learn in an appropriate language environment" (e.g., "My wife encouraged me to go to North America and to take the TOEFL"; "They enrolled me in French immersion"), and "learner plays the role of the teacher" (e.g., "I can teach them what I learn--teaching is the best way to learn.").Of the four support categories, the most common response was "other" forms of encouragement; the χ 2 analysis indicated this tendency was marginally stronger among HL than ESL or ML learners.The three other types of encouragement did not differ between the learner groups.
Almost three-quarters (71.8%) of the 39 valid responses to the question regarding how family members discourage learners could be categorized within the SDT framework and 28.2% were categorized as "Other" (e.g., "they switch back to speaking English"; see Table 6)).Discouragement by family members was distributed fairly evenly across support types and learner groups, and did not differ between groups.
Target Language Community.Over three-quarters (80.0%) of the 125 valid responses to the question regarding how members of the TL community encourage learners could be categorized within the SDT framework, and 20.0% were categorized as "other" (see Table 7)."Other" types of support included providing material support" (e.g., "A guy gave me neat books and mags to read"), as well as "explaining the target culture and its people" (e.g., "They taught me a lot about Canadian culture in their own language and that was something I could not learn from translation").For all groups, providing informative feedback was one of the most important sources of encouragement (e.g., "explained words to me").X 2 analyses showed that the only difference in the distribution of responses was with regard to informative feedback: ESL learners were marginally more likely to say that community members encouraged them in this manner.Over three-quarters (77.8%) of the 81 valid responses regarding how members of the TL community discourage learners could be categorized within the SDT framework (see Table 7).The most common type of discouragement was "negative feedback" that undermined the learners' feelings of competence and provided no information about how to improve their language skills."Other" was the second most common type of response and included aspects of the language learning context being problematic (e.g., "I have trouble speaking in front of a large group and they stop listening after a while") and being intimidated by people who were better at the TL than the learner (e.g., "I feel intimidated by them because they know so much more than me").The X 2 analyses showed that the proportion who indicated that interpersonal distance was discouraging differed across learner groups, with ESL learners being marginally more likely and ML learners being marginally less likely to cite it (largely due to a greater number of comments from the ESL learners about discrimination, such as "Most of them use my accent to discriminate against me").

Discussion
This study examined the reasons why people learn new languages and how other people can support their motivation and learning.It further considered how the learning context can moderate these reasons and support from significant others.After discussing the results regarding the three study objectives, we address some limitations of this research and suggest some directions for future research.

Language learning orientations
The majority of responses to the question of why people were learning another language could be coded into the forms of regulation/orientations posited by SDT.Across all three groups, one of the more commonly cited orientations was that they were learning the language because it would help them to achieve an end that they valued (i.e., identified regulation).The fact that orientations consistent with identified regulation are common across university students is not surprising.At this stage in their development, it might be expected that students are taking courses that will help them to achieve important personal life goals.
Another unsurprising fact was that none reported amotivation, and only weak introjected regulation was evident.Again, as relatively autonomous university students, they had clear reasons for why they were learning the language and were not doing so because of feelings of obligation.
For the other forms of regulation, there was greater variation in their representation across the groups.For most ML and ESL students, the external contingencies of their university program were quite salient; however, whereas the ML students frequently also noted intrinsic motivation, the ESL students did not.For ESL students, the imperative to acquire at least a passing grade in their language course (for most, it was an entrance requirement in order to pursue university studies) overshadowed most other orientations.
Although the ML students were also affected by this requirement, they also had other, more self-determined reasons for learning the language, including a feeling of satisfaction in the learning process.
In contrast with the other two groups, the HL students seldom mentioned external contingencies as a reason for learning the language; instead, more so than the other groups, they indicated that the language was integral to their self-concept.Several respondents also mentioned that learning the language was enjoyable and/or they wanted to integrate into the target community, although not at a rate that was clearly greater than the other two groups.
The importance of the association between identity and HL learning has been consistently found in quantitative studies (Mu, 2015), and as expressed in the current study, it seems that this identity is an important motivator of learning.
It is noteworthy that integrated regulation was shown to play a prominent role in some learners' motivation.Integrated regulation has seldom been observed in studies of language learning orientations, arguably because they often focused on novice learners who have had little opportunity to internalize the language into their sense of self (Noels et al., 1999).This finding underscores the importance of considering the learning context for understanding language motivation.Had we not included these HL students whose lives have long implicated in the TL, we might have wrongly assumed that this form of regulation was irrelevant to language learners, and thereby brought into question the applicability of SDT in the language learning domain.

Support from significant others
Teachers.Consistent with previous SDT-informed studies of teachers' support of learners' motivation (Noels, 2001b;Chaffee et al, 2014;McEown et al., 2014), students felt they were most encouraged when their teacher provided a structured learning environment in which expectations were clearly articulated, the exercises were well designed to improve competence, and feedback was constructive and timely.This type of informational feedback is essential to developing a sense of competence in the language.Also important was the use of engaging, authentic materials.Although it is not evident how such materials might differentially affect learners' feelings of autonomy, competence and/or relatedness (likely all aspects are implicated), the tone of learners' comments indicated that these activities were "fun", and hence might directly encourage intrinsic motivation.ESL students were particularly conscious of informative feedback, perhaps because their achievement on grades and/or the entrance examination was so important for their future education.ML students, on the other hand, were particularly appreciated of authentic materials, perhaps because as foreign language students, they had less access to the target culture.
Friends/Classmates.Past research has demonstrated how classroom climate, which includes peer interactions, has the ability to influence the basic psychological needs for autonomy, competence, and relatedness (Joe et al., 2017).Our study not only confirms this finding, but extends it further by emphasizing the impact that peers have with regard to competence.The most commonly cited form of classmate encouragement in our study, informative feedback serves as a mechanism through which learners can develop competence in a TL.In particular, ESL learners were likely to say that their peers provided them with informative feedback.As the ESL learners in this study tended to be somewhat more advanced in their language skills than the other two groups (i.e., they had higher selfevaluations of linguistic competence than the other two groups), it may be that the classmates of advanced learners can be better relied on to provide accurate information.Additionally, if their friends include members of the TL community, of course, they are also an excellent source of information.On the other hand, negative feedback was the most common form of peer discouragement, thereby demonstrating how peers have the ability to both help or hinder competence.Such discouragement may not be intentional; several students reported that diversity in students' level of competence could be problematic, as for example, might be the case when heritage and modern language students are integrated into the same class (Beaudrie & Ducar, 2005).Rather surprisingly, peers did not serve as a large source of interpersonal immediacy for learners; while forming bonds with peers is certainly very important, more important for language learning motivation seems to be the feedback that peers provide.
Family.A diverse range of family members provided social support to language learners.Although parents were mentioned most often (~70% of the responses), spouses, siblings, children, in-laws, grandparents, uncles, and other extended family were also regularly mentioned.Although family members do provide psychological support of students' feelings of autonomy, competence, and relatedness, other forms of instrumental support were also often mentioned.Learners found the provision of financial and other types of material support, primarily from parents, to be particularly encouraging, as was their encouragement and facilitation of the learners' access to the TL community.These relatively tangible and practical forms of support do not seem to be construed as pressures, but instead as necessary resources that help the learner autonomously learn the language.HL learners were especially likely to comment on this form of parental support, perhaps because their family was better positioned to gain access to the TL community.Although the number of responses was fairly low, family members can also discourage language learning by undermining the learners' decision to pursue an activity they enjoy, criticizing the learners' level of competence, and lacking interest the learners' aspirations and activities.
TL Community.As expert speakers of the TL, members of the TL community are able to provide the necessary information for all learners to improve their competence.
Because the ESL students were living in an Anglophone region, they were more opportunities for them to receive feedback from the TL community than for the other groups.It is noteworthy that conversations with TL speakers are not always supportive, largely because TL speakers also undermine learners' sense of competence, either directly by pointing out the learners' limited competence or indirectly by virtue of the fact that they are inherently more competent in the language.The fact that learners found being with native speakers difficult in part because the setting limited comprehension (e.g., loud parties, large crowds, childlike voices), suggests that TL speakers could quite easily help the learner by avoiding such settings or making adjustments to mitigate the problem.Although only a few people commented on TL speakers being downright nasty (thereby creating interpersonal distance), it is noteworthy that ESL speakers were more likely to report this perception, including prejudicial comments and rude, discriminatory behaviour.
In sum, a large proportion of learners' orientations for learning another language could be accounted for using SDT's forms of regulation.The same was true regarding significant others' types of encouragement and discouragement.Teachers, peers, family members and members of the TL group were all implicated in learners' motivation across all three language learning contexts.That said, some significant others played more prominent roles in some contexts, and some aspects of encouragement/discouragement were more prominent in some contexts than others.

Limitations and Future Directions for Research and Language Education
There are several limitations to this study that point to directions for future research.
First, the focussed essay technique provides an opportunity to gain insight into a greater number of students' experiences than a face-to-face interview study would typically allow.
Future research using this technique however should address two drawbacks.First, unlike face-to-face interviews, there is no opportunity to follow-up participants' responses with probes to better facilitate interpretation.As a result, there were several instances in which it was unclear how to categorize responses according to our coding scheme.An important instance concerns the support of peers; the interview question did not differentiate between friends and classmates, but some research suggests that friends outside the classroom might be more influential than those in the class (MacIntyre, Baker, Clement, & Conrod, 2001).A follow-up meeting with participants in which they review the coding and/or answer the researchers' questions could circumvent this problem.Second, thematic categorization only indicates whether a theme is present or absent in a respondents' text; each theme has equal weight in the description of the responses.It does not provide information about the extent to which each presented theme was endorsed relative to the other themes identified in the same text.If the relative importance of each theme to the participant is of interest, a rating or ranking scale must be incorporated.
Despite these limitations, the fact that the majority of learners' responses to our questions about motivation and social support could be interpreted within the SDT framework points to the utility of this theory across language learning domains.Given that this is the case, several directions for future research are possible.First, this study focused on selfreports of students' experience of supportive and unsupportive behaviour; follow-up studies of significant others' perspectives would be insightful, particularly in terms of the correspondence between what learners and significant others believe they are doing in terms of support.In addition, observational studies of interactions between learners and significant others, focused on the actions and communication style that reflect autonomy-support, informational feedback, and interpersonal immediacy (or not) would also be important.
Second, the complexity of the dynamic relations between social support and learners' motivation must be appropriately modeled.For instance, because learners can interact with significant others who both support and thwart their basic psychological needs (Jang, Kim, & Reeve, 2016), both aspects should be examined.Given that motivation and support can vary over time (Noels, Vargas Lascano, & Saumure, 2019), longitudinal analyses of specific social interactions (e.g., between students and the teacher during an activity or class period) and/or of the development of a relationship, across the duration of a course) are necessary.Because social support is inherently an interpersonal process, dyadic analyses (including conversation and discourse analyses) are necessary (cf., Goodboy & Kashy, 2017), as well as multilevel modeling when considering social interactions across classes, programs, and/or schools.
Third, more studies of how the language learning context impacts the social dynamics of motivational support and the psychological dynamics of learners' motivation, using qualitative, quantitative and mixed methods, is necessary.We warmly welcome the return to studies of languages other than English, including auxiliary (e.g., Esperanto) and classical (e.g., Latin; Katz, Noels, & Fitzner, in press) languages, and we hope that comparative methods will be used to examine variations across languages, programs, nations, and so on.
No doubt the early work of Lambert (1978), Gardner (1985), Clément (1980) and other scholars who emphasized the important role of the social milieu outside the classroom for language learning motivation can provide a useful foundation to this "new" direction (Ushioda & Dornyei, 2017).
With regards to language education, our research findings underscore that the teacher, although important, is not the only person who affects learners' self-determination and motivation.Certainly if teachers adopt an autonomy-supportive orientation, their students are more likely to be self-determined and actively engaged; and there are several resources for the interested instructor to learn how to develop such skills (Noels, 2015;Reeve, 2011).
However, teachers might also consider how the attitudes and actions of other people affect learners' engagement.There is likely little a teacher can do to modify these influences (with perhaps the exception of classmates), but possibly conversations with students about the role that others play in their learning experience might be a way to more self-reflective and self-determined learning.

Conclusion
The findings of the present study contribute to a large body of research that highlights the role of agency and self-determination for motivated engagement in language learning (Noels et al., in press), and it extends this research by highlighting the important influence of the language learning context on psychological and social dynamics involved in motivation.Although several studies have highlighted the important role of the teacher for supporting learners' autonomy, competence, and relatedness, this study demonstrates the complementary roles also played by peers, family members, and the TL community in fostering motivation.By taking the context into greater consideration, it is hoped that a more comprehensive understanding of language learning motivation can be developed.

Figure 1 .
Figure 1.Percentage of learners who indicated each reason for learning a language.

Table 1
Demographic Information for the Learner Groups

Table 2
Example Quotations for Each Type of Encouragement and Discouragement.
"My boss always criticizes me and suggested that I should just open my ears better and it will all be clear."

Table 3
Reasons for Learning Another Language: Distributions, X 2 squared analyses, and examples as a function of theme and learner group a The "%" represents the proportion of people in the learner group who indicated this reason for learning.Percentages may sum to greater than 100% because respondents could indicate multiple reasons.

Table 4
Teachers: Distribution of Responses Regarding Encouragement and Discouragement as a Function of Support Type a The "%" represents the proportion of people in the learner group who indicated this form of encouragement or discouragement.Percentages may sum to greater than 100% because respondents could indicate multiple reasons.

Table 5
Friends/Classmates: Distribution of Responses Regarding Encouragement and Discouragement as a Function of Support Type aThe "%" represents the proportion of people in the learner group who indicated this form of encouragement or discouragement.Percentages may sum to greater than 100% because respondents could indicate multiple reasons.

Table 6
Family: Distribution of Responses Regarding Encouragement and Discouragement from Members of the Family as a Function of Support TypeThe "%" represents the proportion of people in the learner group who indicated this form of encouragement or discouragement.Percentages may sum to greater than 100% because respondents could indicate multiple reasons. a

Table 7
Target Language Community: Distribution of Responses Regarding Encouragement and Discouragement from Members of the Target Language Community a Function Support Type The "%" represents the proportion of people in the learner group who indicated this form of encouragement or discouragement.Percentages may sum to greater than 100% because respondents could indicate multiple reasons. a