Sixty Years of Language Motivation Research

This article offers a historical analysis of the major themes that the language motivation field has examined in its 60-year history. The discussion starts by briefly reviewing the social-psychological and the situated–cognitive periods. The former was primarily concerned with affective factors in intergroup relations, while the latter with learners in classroom contexts. The second half of the article surveys a number of emerging themes in the field to highlight major findings and potential future directions. These themes include the dynamic, affective, unconscious, and long-term aspects of motivation to learn English and other languages, as well as the implications of the pervasive presence of technology in daily life.


Introduction
One day, a Canadian graduate student met with his advisor to discuss possible thesis topics. During that meeting, the student remarked that he could not see how one could learn the language of another group if she or he does not like that group. At this point, the advisor said, "Hey man. There's your thesis!" The student was Robert Gardner with his supervisor Wallace Lambert at McGill University in 1956. That meeting gave birth to the field of second language (L2) motivation and was the primary instigator of decades of research (Gardner, 2001).
The purpose of this article is to offer a historical background of research into language motivation and then highlight a number of emerging themes that seem to hold potential for future research. This analysis builds on and expands previous efforts to understand the historical trajectory of the field (e.g., Boo, Dörnyei, & Ryan, 2015;Dörnyei, 2005;Dörnyei & Ushioda, 2011). These historical analyses have identified three phases that the field has gone through. The first is the social-psychological period, in which a common theme was the focus on the macroperspective language learning. That research was largely concerned with affective factors in intergroup relations. The field then progressed into a second phase, in which the scope of research was expanded to address learners in the microcontext of the classroom and the cognitive processes underlying language learning. The current, third phase is witnessing the emergence of various themes, such as the dynamic, affective, unconscious, and long-term aspects of motivation to learn English and other languages.
Previous surveys of the field may be characterized as back pointing, in which they have primarily focused on trends found in the past two phases without elaborating on the emerging trends in the current period. The present analysis therefore offers a more forward-pointing survey of these emerging trends. As reviewed below, some of these themes have been discussed separately in the literature, but they have not been synthesized and presented in one place to date. In addition, one of these themes, technology, has been largely overlooked in the motivation field despite the increasing technologization of learning and classroom instruction in many parts of the world. Before surveying these themes, the discussion starts with a brief overview of the first two phases.

The Social-Psychological Period
The social-psychological period was spearheaded by Gardner and associates in Canada (e.g., see Gardner, 1979Gardner, , 1985Gardner, , 2010. The fundamental basis of this research is the assumption that learning an L2 is different from other school subjects because L2 learning additionally requires openness to the L2 group and willingness to adopt features from it. This concept came to be known as integrative motivation. A number of 701976S GOXXX10.1177 1 The University of Nottingham, UK 2 Jubail Industrial College, Saudi Arabia researchers in the first half of the 20th century-including Arsenian, Marckwardt, Nida, Whyte, and Holmberg-challenged the then-dominant view that intelligence and aptitude are the primary factors in successful L2 learning. Building on work by these researchers, Gardner undertook a more focused investigation into the role of affective factors. Gardner (2010) classifies the history of his own research program into three phases. He calls the first phase ancient history. This phase dates from 1945 to 1972, and covers the above early researchers as well as Gardner's MA and PhD work and later research included in Gardner and Lambert (1972). The second phase, or early history, spans the 1970s and the early 1980s. During this period, Gardner and P. C. Smythe obtained funding to establish the Language Research Group at the University of West Ontario and conducted studies across Canada. Gardner calls the last phase of his research modern history, which describes work conducted in the 1980s. This is when Gardner and his graduate students continued research after the Language Research Group was disbanded. (As explained in more detail below, the 1990s marked the beginning of the cognitive-situated period.) During these three phases, Gardner and associates engaged in a very productive research program, which led to the development of the socioeducational model (for a review, see Gardner, 2010). This model postulates four different aspects of the learning process: social milieu (cultural and educational backgrounds), individual differences (IDs; intelligence, aptitude, motivation, and anxiety), acquisition contexts (formal vs. informal), and outcomes (linguistic vs. nonlinguistic). Most empirical research focused on the integrative motive, according to which language achievement is influenced positively by motivation and aptitude, and negatively by language anxiety. Motivation, in turn, is a function of integrativeness, attitudes toward the learning situation, and instrumentality. Gardner (2007) also identifies four stages of L2 development: elemental, consolidation, conscious expression, and automaticity and thought. From this perspective, "acquisition involves making the language part of the self" (Gardner, 2010, p. 7). Although Gardner's framework was arguably the dominant paradigm in the social-psychological period, there were also other frameworks active at the time. These include Clément's (1980) social context model, Giles and Byrne's (1982) intergroup model, and Schumann's (1978) acculturation model. However, as Dörnyei and Ryan (2015) observe, the common theme shared by all these approaches is their macro-level analysis of the interrelationship between social groups and contextual variables. Interest in more classroomoriented research had to wait until the second period.

The Cognitive-Situated Period
For decades, Gardner and associates have repeatedly insisted that the socioeducational model is dynamic rather than static (e.g., Gardner, 2010, p. 46;Gardner & Tremblay, 1994, p. 366;MacIntyre, 2002, p. 49) and that it "does not just link variables together but describes a process" (Gardner, 2010, p. 59). Nevertheless, Gardner has also acknowledged that It is not intended to provide explanations to individual teachers as to why or why not some of their students are more or less successful than others, or to give teachers advice on how to motivate their students, or to provide reasons to students to help them understand their own success or lack thereof. It is a model to account for general relationships in a parsimonious and testable structure that is subject to verification and replication. (Gardner, 2010, p. 26) In addition to not being classroom-friendly, Gardner's model was criticized from other perspectives. For example, Dörnyei (1994bDörnyei ( , 2005 explains that mixing motivational intensity (i.e., effort) with the abstract mental phenomenon of motivation will lead to conceptual ambiguity (see also MacIntyre, 2002, p. 49). In addition, with the spread of World English as a decentralized global language, the idea of integrating with native speakers from Anglophone countries started to became less and less meaningful (e.g., Coetzee-Van Rooy, 2006;McClelland, 2000). Some integrative and instrumental orientations have also become hardly distinguishable (Lamb, 2004).
Starting from the 1990s, these accumulating issues led language motivation research to shift into a new phase, which was described as the cognitive-situated period Dörnyei & Ushioda, 2011). Because Gardner's integrativeness was conceived of as an affective factor, Dörnyei (2010) tries to uncover its cognitive underpinnings through reinterpreting it into the ideal L2 self. The attempt to reinterpret integrativeness in cognitive light gave birth to the L2 Motivational Self System (L2 MSS, Dörnyei, 2005Dörnyei, , 2009a. The L2 MSS consists of the ideal L2 self, the oughtto L2 self, and the L2 learning experience, thus preserving the original tripartite conceptualization of the integrative motive. The L2 MSS is based on two "parent" theories, self-discrepancy theory (Higgins, 1987), and possible selves theory (Markus & Nurius, 1986). On one hand, learners experience uneasiness if their perceived level of proficiency is discrepant from the level they aspire to achieve (i.e., actual-ideal discrepancy) or from the level they think they are expected to meet (i.e., actual-ought discrepancy). This psychological uneasiness may serve as a motivator to reduce this discrepancy by improving L2 proficiency. On the other hand, possessing an elaborate vision of a desired possible self may intensify one's motivation because the imagined self becomes an experiential reality that the individual can see and hear. The closest parallel to this ideal self-image is the L2 native speaker. Learners, therefore, could draw from their past knowledge of L2 speakers to envision for themselves a desired future that would have a motivational effect in the present. According to a survey of language motivation research spanning about a decade, Boo et al. (2015) observe that the L2 MSS is currently the dominant framework in the field. In another survey spanning about two decades, Sugita McEown, Noels, and Chaffee (2014) also observe that an increasing number of researchers have used Self-Determination Theory (Deci & Ryan, 1985) as their primary theoretical framework or have drawn from some aspects of it.
In addition to these two theoretical frameworks, some researchers have continued trying to realign language motivation with educational psychology and to make use of other non-L2-specific theories, such as social cognitive theory (e.g., Mills, Pajares, & Herron, 2007) and attribution theory (e.g., Williams, Burden, Poulet, & Maun, 2004). This type of innovation necessitated also expanding the methodological repertoire, which has led to an increasing number of qualitative investigations as well as practical applications of language motivation research into classrooms (Boo et al., 2015). In fact, because this research has been so concerned with classroom processes and with making motivation research more teacher-friendly-as opposed to the focus on larger picture in the social-psychological period-it might be appropriate to describe this phase as the educational period.

The Current Period
According to Dörnyei and Ryan's (2015) historical analysis, the language motivation field is currently in its third phase. Dörnyei and Ryan characterize this period by the shift to sociodynamic perspectives. Indeed, perhaps the most salient characteristic of this phase is the growing emphasis on the dynamic nature of motivation and its temporal variation. However, a number of other themes are also emerging, and so "we need to take care not to portray researchers . . . as part of a coordinated, focused movement, when it was more the case of various diverse concerns emerging at a similar time" (Dörnyei & Ryan, 2015, p. 80). This makes it hard to give this period a single monolithic title. Such titles usually emerge in retrospect, especially when the field is ready to move to a new phase.
Looking at developments in the field from a broader perspective, some commentators (e.g., Boo et al., 2015;Sugita McEown et al., 2014) have also pointed out some limitations in previous research. For example, both of the previous periods coincide in characterizing motivation as a conscious process in which learning English-rather than other languages-is examined within a relatively short duration and using rather "simplistic" research designs (Boo et al., 2015, p. 156). The samples investigated also tend to be tertiary students, while different theoretical frameworks have tested participants from different age groups and from different cultural backgrounds, thus making cross-theoretical comparisons problematic (Sugita McEown et al., 2014). In addition, most of this research has overlooked the increasing technologization of everyday life and its impact on language learning nowadays. These themes are discussed in more detail next.

Dynamic Motivation
The observation that, unlike L1 learners, L2 learners vary substantially in how successful they are in their language proficiency prompted research into IDs. Researchers identified several ID factors that could potentially account for this variability, such as aptitude, motivation, learning styles, learning strategies, and anxiety (e.g., Dörnyei, 2005;Skehan, 1989). This approach can be intuitively summarized as finding out "why, how long, how hard, how well, how proactively, and in what way the learner engaged in the learning process" (Dörnyei & Ryan, 2015, p. 6, emphasis in original). Despite the intuitive appeal of this approach, recent theoretical analysis suggests that it may no longer tenable. In fact, because this approach rests on a number of problematic assumptions, Dörnyei (2009b) describes it as the "individual differences myth." One of these problematic assumptions is that these IDs are clearly identifiable (Dörnyei, 2009b). However, close examination of some of the traditional IDs shows that the borders between them are fuzzier than first assumed. For example, motivation is traditionally viewed as an affective factor and has been contrasted with the cognitive nature of aptitude, but, in reality, most influential motivational theories draw heavily from cognitive research and have cognitive components. Another example is anxiety, which is sometimes treated as a motivational component, sometimes as a personality trait, and sometimes as an emotion (MacIntyre, 2002). Similarly, aptitude involves not only cognitive aspects but also affective and conative dimensions (Robinson, 2007). The fuzzy distinction between the different IDs calls into question their modular view.
Another problematic assumption is that IDs are relatively stable (Dörnyei, 2009b). However, it is now increasingly recognized that, far from being stable, IDs are highly sensitive to contextual and temporal variation. From context to context and from time to time, the different IDs fluctuate, and consequently, their effect on language learning correspondingly fluctuates. A "motivated" learner may be less motivated on the next day, or at the next task. Even fluid intelligence and language aptitude are increasingly viewed as malleable. Fluid intelligence, which used to be assumed fixed and genetically predetermined, interacts with the environment, and therefore it is "dynamic rather than static and modifiable rather than fixed," which makes it "trainable to a significant and meaningful degree" (Sternberg, 2008, p. 6791). Neither is language aptitude independent of context, and instead, it is sensitive to task and situation specificity (Robinson, 2007). Therefore, the observed association between a cause and its effect may be substantial at one occasion but negligible at another (Ellis & Larsen-Freeman, 2006). "People do what their immediate situations tell them to do rather than what their long-standing internal traits might prompt them to do" (McAdams, 2006, p. 12).
An alternative to the modular view of IDs is that learner characteristics are dynamically changing both in response to context and time, and as a result of their interaction with each other. Dörnyei (2009b, p. 231) suggests that dynamic systems theory may be a viable approach that can do justice to this complex conceptualization. The adoption of a complexity theory perspective in language motivation may be seen as officially inaugurated by the publication of a recent anthology edited by Dörnyei, MacIntyre, and Henry (2015). The contributors to this volume drew from a variety of data-analytic techniques to examine the dynamics of motivation such as idiodynamics, latent growth modeling, and trajectory equifinality analysis. Dörnyei and Ryan (2015, p. 102) anticipate that this new perspective has the potential to keep language motivation researchers busy for the next decade.

Affect and Emotions
Dörnyei's attempt to reinterpret integrativeness in cognitive terms was part of a more general trend in the language motivation field. Starting in the early 1990s, researchers tried to adopt cognitive constructs following the then-trendy approach in educational psychological research (e.g., Dörnyei, 1994a). This general shift from affective-based models to cognitive-based models seems to implicitly reflect the view that affect is a "post-cognition" phenomenon, in which affect is a mere result of cognition (e.g., Muncy, 1986): Once we understand the cognitive processes involved, we can then deduce the affective outcomes. Although this view was dominant after the cognitive revolution, some recent research has cast doubt on it. For example, Pessoa (2008) argues that, at the neural level, the view that cognition and emotion are separate entities does not hold; in many cases, the two contribute jointly to behavior (see also Okon-Singer, Hendler, Pessoa, & Shackman, 2015). The situation is no different when it comes to language learning. As Schumann (1997) explains, "from a neural perspective, not only are various affective processes interrelated, but affect and cognition are also intimately intertwined" (p. 238).
An important class of this affective dimension is emotions, due to their close connection to identity and adjustment in language learning (e.g., Noels, Pon, & Clément, 1996). In fact, it might be argued that emotions are "true" affects, while the affective-based models in the first period may actually be cognitive to some extent. Despite this, emotions have not received adequate attention in the second language acquisition (SLA) field in general and in language motivation in particular. Dörnyei and Ryan (2015) blame this on the cognitivist roots of SLA, as well as the irregular, fluctuating nature of emotions. This is why Dörnyei and Ryan (2015) describe emotions as "the greatest omission" (p. 9) among individual difference variables, leading the field to suffering from an "emotional deficit" (p. 10). Emotions were also described as "fundamentally important motivators" (MacIntyre, Mackinnon, & Clément, 2009, p. 47) and as "the fundamental basis of motivation" (MacIntyre, 2002, p. 45), because motivation without emotion remains cold cognition that lacks potency. Swain (2013) also considers emotions the elephant in the room, maintaining that cognition and emotion are at least interdependent and at most integrated and inseparable.
Despite the fact that emotions have long been kept "in the shadows" of language learning discussions in favor of other variables (Garrett & Young, 2009, p. 209), actual classroom experience indicates that ability and attitudes alone are not sufficient to support motivation (MacIntyre, 2002). Classrooms in general are a cause of emotional turmoil for many people, and the language classroom in particular can be an especially emotionally loaded experience (Dörnyei & Murphy, 2010). As an illustration, MacIntyre (2002) gives the example of embarrassment. It is hard to imagine a language learner who has not been in an embarrassing situation in the language classroom, and depending on the intensity of embarrassment, the learner might resort to withdrawal or reticence as a form of emotional defense (King, 2011). It does not help that curriculum designers find it easier to focus on rigid activities that involve little emotional investment, which puts further burden on the teacher (Dewaele, 2015). To deal with this situation, Kramsch (2009) recommends that teachers try to detect aspects in the syllabus that can be subject to emotional arousal, such as love or hate, to encourage more learner investment in the lesson.
Although early research into emotions in language learning was limited to the detrimental effects of negative emotions, most notably anxiety (e.g., Horwitz, Horwitz, & Cope, 1986), the recent interest in different emotions has been dubbed as the affective turn in SLA (Pavlenko, 2013). For example, Gregersen and MacIntyre (2014) argue that neither are negative emotions always bad nor should researchers overlook positive emotions. This position has materialized most clearly in the publication of three anthologies in the same year (Gabryś-Barker & Gałajda, 2016;Gkonou, Tatzl, & Mercer, 2016;MacIntyre, Gregersen, & Mercer, 2016), marking the inception of the positive psychology movement in SLA. The contributors to these anthologies discuss various topics relating emotions to language learning, including empathy, hope and hardiness, enjoyment, flow, eudemonic happiness, and love. Considering the controversial status of positive psychology (e.g., Coyne & Tennen, 2010), the next few years would reveal the extent to which this movement is accepted in mainstream L2 motivation.

Unconscious Motivation
Despite the abundance of perspectives on identity, one idea has stood the test of time: William James's (1890) distinction between the I-self and the Me-self. Whereas the I-self constitutes the willful and volitional aspects of motivation, the Meself reflects automatic and unconscious motivators. As Roeser and Peck (2009) review, these two systems are separate but functionally interdependent, and both systems have to be accounted for in explanations of motivation. To date, most research on language motivation has, at least implicitly, assumed that the learner is a rational individual who is able to recognize and articulate what motivates him or her (Al-Hoorie, 2016a). Recently, however, there has been a resurgence in the interest in unconscious motivation in mainstream motivational psychology, as an increasing number of psychologists are starting to realize the importance of unconscious motivators (see Al-Hoorie, 2015). For example, in their review of the emerging themes in mainstream motivational psychology, Ryan and Legate (2012) document their surprise that the interface between conscious and unconscious motivation was the most frequently cited area to hold potential for future research. The language motivation field has also reached a level of maturity that allows it to start exploring issues related to unconscious motivation and to catch up with other SLA subdisciplines where unconscious processes have become a stable topic of investigation. To be more specific, "our field is ready to expand into exploring these areas because it seems evident that language globalisation has created a linguistic landscape that is characterised by both powerful positive trends and strong negative undercurrents" (Boo et al., 2015, p. 156).
Adopting an unconscious angle of human motivation does not have to be at odds with the current frameworks in the field. For example, neither possible selves theory nor self-discrepancy theory, the two parent theories of the L2 MSS, would disapprove of unconscious processes. In terms of possible selves theory, Markus and Nurius (1986) discuss the possibility of the unconscious activation of both positive and negative possible selves. In describing the effects of unconscious activation of possible selves, Oyserman (2013) similarly asserts that "these effects are automatic and do not require that people make a conscious choice as to how to think about themselves" (p. 185; see also Oyserman, 2015, p. 44). In a special issue marking the centennial of the publication of James's (1890) The Principles of Psychology, Markus (1990) contributes with an article titled "On Splitting the Universe," in which she endorsed James's distinction between the I-self and the Me-self and stressed its relevance today.
Self-discrepancy theory also accommodates unconscious processes: self-discrepancy theory does not assume that people are aware of either the availability or the accessibility of their selfdiscrepancies. It is clear that the availability and accessibility of stored social constructs can influence social information processing automatically and without awareness. (Higgins, 1987, p. 324) Neither do the behavioral consequences have to be conscious (Higgins, 1989, p. 98). In fact, self-discrepancy theory does not assume that a future self-guide is a stable individual difference variable (Higgins, 1998), but that situational variability can unconsciously induce the motivational effect independently from the nature of the learner's self-guides. In one study, for example, Higgins, Roney, Crowe, and Hymes (1994) use an ostensibly unrelated task to activate either the ideal or ought selves of their participants. Although the participants were not aware that their ideal or ought selves were activated, this activation was still successful in unconsciously shaping their performance on a subsequent free recall task. Gardner's integrative motivation also allows some room for such unconscious conceptualizations. In Gardner's (2010) words, integrativeness is not a conscious decision on the part of the individual and . . . individuals may not be aware of it . . . The rationale underlying integrative motivation is that emotional factors can influence behavior, sometimes in ways that are not even perceived by the individual concerned. (pp. 223-224) Although current theoretical frameworks are not, in principle, at odds with investigations into unconscious phenomena, researchers in reality have relied predominantly on self-report questionnaires and interviews (Ushioda, 2013). Expanding language motivation research to include implicit processes may enrich the field and open up numerous potential pathways. As an illustration, motivational psychologists have examined the implicit dimension of many well-known constructs. Examples include implicit attitudes (Petty, Fazio, & Briñol, 2009), implicit prejudice and stereotypes (Levinson & Smith, 2012), implicit motives (Schultheiss & Brunstein, 2010), implicit self-concept (Briñol, Petty, & Wheeler, 2006), implicit self-determination (Keatley, Clarke, Ferguson, & Hagger, 2014), and implicit self-regulation (Koole, McCullough, Kuhl, & Roelofsma, 2010). It clear that language motivation researchers would benefit from exploring "the other side" of their constructs as well (e.g., Al-Hoorie, 2016a, 2016b).

Long-Term Motivation
Language motivation does not emerge in a vacuum, but develops through interaction with various events over one's life history. The life history of each learner plays a major role in whether she or he decides to take up learning the language and whether she or he sees a reason to persist in it. This requires researchers to look at the bigger picture of the ecology of motivation. As an illustration, educational psychologist Avi Kaplan describes how a similar notion emerged in his research: Exploring various methodologies, [Kaplan] conducted a narrative interview study with 10 undergraduate students, aiming to understand the processes that led them to adopt different achievement goal orientations toward their studies in college. To his surprise, when asked about their experiences, choices, and engagement in college, the students spoke relatively little about the characteristics of the learning environment, their self-efficacy, or their attributions for success and failure. Instead, they elaborated on growing up in their hometown, their high school experiences, their dilemmas concerning careers and relationships, their family, ethnicity, friends, and the peer groups they belonged to or wanted to belong to. These students mentioned schoolwork specifically when the material seemed to be relevant to who they thought they were and who they considered or wanted to be. [Kaplan] had the insight that, to a large extent, these students' achievement goals in college were based in their identity and identity formation processes. (Kaplan & Flum, 2009, p. 73) Although long-term investigations require longitudinal designs, longitudinal studies are a minority in the language motivation literature (Sugita McEown et al., 2014). This status quo creates gaps in our knowledge. For example, despite their popularity in the literature lately, L2 self-guides have been typically examined cross-sectionally, and so little is known about how they initially develop or their evolutionary trajectories over time. For example, Dörnyei (2009a) suggests that L2 self-guides might not be appropriate for presecondary learners (see also Lamb, 2012), but there is "virtual absence" (Boo et al., 2015, p. 156) of systematic research into the motivation of younger learners.
An interesting exception to the lack of research on longterm motivation is the recent development of the notion of Directed Motivational Currents (DMCs, Dörnyei, Henry, & Muir, 2016). DMCs may be described as flow-like experiences (cf. Csikszentmihalyi, 1975) that extend over diverse tasks unified by an overall goal. In other words, DMCs are unique motivational surges that span over longer timescales and that are not necessarily enjoyable in themselves-as pleasure is derived from the end goal that is external to the activity. DMCs therefore occur when there is a clear vision of that goal as well as an identifiable factor triggering the launch of motivation. After their launch, DMCs are then maintained by ongoing behavioral routines and progress checks. Eventually, DMCs decline and motivation goes back to its normal levels. In explaining the rationale behind the term Directed Motivational Currents, Dörnyei et al. (2016) explain that "Both motivational and ocean currents represent a formidable flow of energy, carrying the life-forms caught up within them unimaginable distances" (p. xi).
In a first empirical study of the concept of DMCs, Henry, Davydenko, and Dörnyei (2015) conduct interviews with language learners who had experienced DMCs. The results showed that DMCs are characterized by a salient facilitative structure, involvement of identity investment goals, and positive emotionality. Further investigation into this positive emotionality has revealed that participants attribute it to the feeling that their entire identity was being transformed in the process (Ibrahim, 2016a). Other empirical research showed that DMCs can also be experienced by a group of individual working on a project (Ibrahim, 2016b) and may be intentionally induced by teachers (Muir, 2016). These results also suggest that the dynamics of such motivational surges may be different from regular, and even high, motivation. To date, most research in this area has been qualitative, and so only future research would tell how generalizable these findings are.

Languages Other Than English (LOTEs)
The SLA field is concerned with "the processes by which school-aged children, adolescents, and adults learn and use, at any point in life, an additional language, including second, foreign, indigenous, minority, or heritage languages" (The Douglas Fir Group, 2016, p. 19) as well as sign languages (Woll & Adam, 2012). This is one factor why there has been a growing interest in multilingualism in other SLA disciplines, which has amounted to a multilingual turn (e.g., Conteh & Meier, 2014;May, 2014). In fact, in recent discussions, there have been calls to go beyond multilingualism. For example, some authors have drawn from Mikhail Bakhtin's notion of heteroglossia to refer to the simultaneous use of multiple language forms and signs and the ensuing conflicts among them (e.g., Bailey, 2012), whereas critical applied linguists Makoni and Pennycook (2012) question the traditional view of languages as discrete entities rather than fluid and dynamic acts of identity. Makoni and Pennycook use the term lingua franca multilingualism to describe this view.
The L2 motivation field has not actively engaged with these debates. Just as its name suggests, L2 motivation research has mostly looked at learning an L2, typically English. For example, Boo et al.'s (2015) survey shows that research into the motivation to learn English was by far more predominant than research into all other languages combined. Similarly, in the context of heritage language learning, Comanaru and Noels (2009) point out that little work has been done to examine the motivational and affective profiles of these learners. In the context of indigenous languages, Ball (2009) also reports that she conducted a literature search on the difficulties experienced by indigenous language learners, but she could not find a single study that satisfied her search criteria. To quote Leeman and King (2015), research on minority languages "remains marginalized, underfunded, and often an after-thought" (p. 211).
Recently, some researchers argued that the dynamics of the motivation to learn (Global) English might be very different from those of learning LOTEs (Dörnyei & Al-Hoorie, in press). Because of the status of English, the motivation to learning it might interfere with the motivation to learn other languages. As an illustration, Dörnyei, Csizér, and Németh (2006) conduct a large-scale longitudinal investigation of language learning motivation-involving more than 13,000 learners over a period of 12 years focusing on five target languages (English, German, French, Italian, and Russian) in Hungary. The results revealed a fundamental restructuring of the different L2 learning dispositions. English maintained its high profile, but the other languages dropped steadily. Even the former lingua franca of the region, German, gradually became limited to only a selected few. Furthermore, Global English itself also displayed a marked shift over the decade: Although its popularity remained as strong as ever, the correlational link between motivation and the choice of English for language learning decreased, suggesting that the study of English is increasingly becoming a self-evident part of education rather than an L2-specific motivated decision.
Subsequent analysis also suggests that language learners develop distinct ideal L2 self-guides (see Dörnyei & Chan, 2013). It is therefore likely that the self-guides of the influential language would develop more strongly at the expense of self-guides related to other languages. Indeed, in a cluster analysis study, Csizér and Dörnyei (2005) identify different learner profiles and demonstrated that a positive disposition toward one language can clash with that of another. Similarly, research by Henry (2010Henry ( , 2011 also suggests that learners of different languages possess separate self-concepts and that the motivation to learn English can deplete the working selfconcept, thus disadvantaging LOTEs. All of this leads to the conclusion that the motivational basis of learning English may be quite distinct from that of learning LOTEs. Because most available research is Englishbiased, the available theories most likely reflect learning English rather than LOTEs. Paying more attention to LOTEs has the potential to deepen our understanding of the complexities involved in language learning motivation. One interesting outcome of researching LOTEs is that the role in integrative motivation seems to resurface. Integrative motivation might be a relevant concept in the context of LOTEs as there is usually a specific community out there that speaks the language and that is considered the "owner" of that language. In addition, many individuals who decide to take up learning a certain LOTE do so because they are already in the geographical area where the language is spoken or because they plan to move there. A person trying to learn Danish, for example, would most likely be thinking about a localized community rather than imagining themselves as a global citizen. The motivation to learn LOTEs is the subject of an upcoming special issue (Ushioda & Dörnyei, in press), which is likely going to stimulate further theoretical and empirical research in the future.

Technology and Motivation
In 2003, Carol Chapelle declared that the bond between technology and language use in the modern world should prompt all language professionals to reflect on the ways in which technology is changing the profession of English language teaching in particular, and applied linguistics as a whole. (Chapelle, 2003, p. 1, emphasis in original) Subsequently, computer assisted language learning (CALL) became mainstream in applied linguistics with various applications, especially in language testing (e.g., Chapelle & Douglas, 2006). Language motivation researchers also recognized the potential technology has. In an early study, Warschauer (1996) finds that students-regardless of gender, experience with computers, or learning context (second vs. foreign)-have positive attitudes toward using computers for writing and communication, and so the author recommended that teachers exploit this to enhance student motivation. Appel and Mullen (2002) also examine the effect of engaging in email exchanges on language learning. The authors devoted a section of their article to discuss the motivational implications of this strategy. Admittedly, however, although research into CALL has flourished over the years, there has been little overlap between the CALL and language motivation literatures, and these two disciplines have progressed largely independently.
To further complicate the scene, there has been an explosive growth in the use of technology in everyday life recently. In fact, due to the ease of access to online technology and its interactive nature, the popularity of TV and DVDs is starting to wane among young people (Henry, 2013). Nowadays, young people have access to a variety of social networking websites that facilitate exposure to the L2 such as Facebook, Twitter, Instagram, and Snapchat, as well as interactive 3D gaming such as Counter-Strike, Call of Duty, World of Witchcraft, The Sims, Second Life, and Unity (e.g., Collentine, 2011;Deutschmann, Panichi, & Molka-Danielsen, 2009;Gee & Hayes, 2010b;Peterson, 2010). Internet users can also easily read and contribute to discussion forums, blogs, wikis, and other online communities, all available in English (Kessler, 2009;Pinkman, 2005). Last, but certainly not least, there is the pervasive prevalence of smart phones, tablet computers, and wireless laptops (Stockwell, 2013). These gadgets also come with a bewildering amount of educational apps (Hirsh-Pasek et al., 2015). In Apple Store alone, there are currently more than 75,000 educational apps ready to download (Apple, 2017). This comes amid skepticism about the utility of these apps, and even concerns that they might, ironically, lead to screen addiction, increased aggression, depression, and anxiety (Kardaras, 2016). The inevitable conclusion from the changing face of today's language learning (cf. J. C. Richards, 2015) is that "a new type of student" is emerging (Henry, 2013, p. 138).
It is clear that language motivation research is still lagging behind these developments. At the same time, this rapidly evolving popular culture now constitutes a competition for classroom learning (Gee & Hayes, 2010a). As an illustration, Henry (2013) reports an authenticity gap that an increasing number of language learners are experiencing. That is, many learners are exposed to the L2 both inside and outside the classroom, but their experiences outside the classroom are much more stimulating. Contemporary digital gaming, for example, requires intense interaction, communication, and cooperation with real people (usually in English) to proceed in the game. These interactions may include devising sophisticated strategies and plans to carry out military operations to defeat opponents or monsters, which stimulates a lot of creativity and imagination and which can easily induce long flow experiences. These interactions also take place with players from all over the world, including native speakers of English, thus increasingly blurring the line between second and foreign language contexts. Native speakers are now virtually present, and so they no longer need to be present physically in the neighborhood for learners to have regular and meaningful interactions with them. These experiences dwarf the routine activities that learners do in the classroom, which now appear banal and trivial in comparison. A number of researchers have therefore called for building bridges between the classroom and these digital leisure activities to make classroom learning more motivating to today's learners (cf. Henry, 2013).
Today's technology has expanded what used to be imaginable for learners and teachers, opening the gate for imagined identities that are characteristic of the new world order (Darvin & Norton, 2015). Drawing from earlier work (K. Richards, 2006;Zimmerman, 1998), Ushioda (2011) argues that technology can be used to harness the learners' transportable identities. The notion of transportable identities refers to latent dimensions of one's identity that can be invoked in interaction, such as the teacher being a cat lover or the student being a fan of Manchester United. Ushioda explains that drawing from the identities learners develop through technology would encourage them to engage more genuinely in target-language communications in the classroom. Teachers are no longer limited to the handout they bring to the classroom everyday, but can now draw from topics they find interesting in discussion forums, live chats, blogs, wikis, podcasts, social networking sites, and video-sharing sites. This makes lesson planning take a new meaning. As this approach treats them as "people" rather than as abstract "language learners," students would develop a greater sense of autonomy and ownership of the activity, thus leading to more investment (Ushioda, 2011). This is certainly a promising area for future research.

Conclusion
This article has reviewed the three major phases that the language motivation field has passed through. The most recent phase is characterized by a number of diverse themes. In fact, Dörnyei and Ryan (2015) warn that the expanding scope of language motivation research may lead to fragmentation in which researchers in our field "will no longer speak the same language" (p. 102). This seems to be the natural evolution of academic fields as they mature (for an example in psychology, see Sternberg, 2005). The real danger is when different research strands use different jargons to describe very similar phenomena but with little overlap in their reference lists.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.