Willingness to communicate in a second language and the promise of complex dynamic systems theory: A narrative review

Abstract

Willingness to communicate in a second language (L2 WTC) is a speaker’s voluntary engagement in communication using a target language. WTC has undergone several conceptualisations over the past twenty years or so. The aim of the current article is to present a narrative review of the major conceptual developments in research on L2 WTC. First, the article discusses the strengths and limitations of the major conceptualizations of L2 WTC, i.e. MacIntyre et al. pyramid model, Wen and Clement Chinese conceptualization, and Kang’s situational model of L2 WTC. Second, the article presents the basic features of the complex dynamic systems theory (CDST) and discusses how it serves as a meta-theory with immense explanatory power to encompass the complex, dynamic and non-linear behaviour of L2 WTC. Finally, corresponding to a CDST construal of L2 WTC, the paper discusses some of the methodological developments and possible directions for future research. The article aims to contribute to language teachers’ and teacher educators’ awareness of the complex and dynamic nature of L2 WTC and provide future researchers with an alternative theoretical framework and corresponding methods to study L2 WTC.

Keywords

Willingness to communicate complex dynamic systems theory second language pedagogy classroom research teacher education

Modern approaches to second and/or foreign language learning insist upon learners’ willingness to make authentic use of target language for meaning making. Willingness to communicate in a second language (L2 WTC henceforth) is defined as a learner’s proactive exploitation of opportunities to communicate in L2 with a specific interlocutor(s) in a specific context (MacIntyre et al., 1998). Much of the research into L2 WTC has been characterized by an ontological, epistemological and methodological orthodoxy (Baker and MacIntyre, 2003; Donovan and MacIntyre, 2004; Dornyei and Ryan, 2015; Khajavy et al., 2016; MacIntyre et al., 1999, 2001). Ontologically, studies have maintained a separation between mind, body and world; in other words, they have examined psychological, contextual and physical variables in isolation from one another. Epistemologically, L2 WTC has been conceptualized as a static and linear phenomenon; and methodologically, most studies have employed quantitative tools to examine the correlations between variables (Dornyei and Ryan, 2015; MacIntyre and Legatto, 2011; Weaver, 2009). Moreover, previous research has lavished an overwhelming focus on psychological variables, especially desire to communicate with a specific person, self-confidence, and motivation, and personality, and presented a static, linear and componential view of variables in isolation from contextual and physiological variables underlying L2 WTC. While studies adopting static and linear approach and quantitative methodology have rendered useful insights into the trait-like behaviour of L2 WTC, however, there is still a need for a meta-theoretical framework which encompasses psychological, contextual and physiological aspects, adopts innovative methods, including both quantitative and qualitative, which might allow an in-depth examination of the complex, dynamic and non-linear nature of L2 WTC and the interaction amongst underlying variables.

It is in this scenario that complex dynamic systems theory (CDST henceforth) has gained increasing popularity amongst scholars in the field of individual difference (ID) variables in particular (de Bot and Larsen-Freeman, 2011; Dornyei, 2009; Dornyei et al., 2015; Dornyei and Ryan, 2015; King, 2016; Larsen-Freeman and Cameron, 2008). Most significantly, research on motivation has taken the lead in exploring the complex and dynamic nature of L2 motivation (Dornyei, 2014; Dornyei and Ryan, 2015; Waninge et al., 2014). This theoretical shift to the CDST paradigm has also impressed scholars of L2 WTC and has inspired some of the empirical work as well (Syed, 2016; MacIntyre and Legatto, 2011; MacIntyre and Serroul, 2015; Peng, 2016; Wang, 2019; Wood, 2016). However, the field is still dominated by a positivist paradigm (Dornyei and Ryan, 2015; Larsen-Freeman, 2019; MacIntyre, 2012, 2020). In Larsen-Freeman (2019, p. 73) words.

It is fairly common practice with studies of individual differences, such as those having to do with motivation or attitude, to treat all members of a group so designated as if they were homogeneous, and thus overlook their particularity. Then, too, ideologies can lead to deficit views of learning and can demean language users.

Moreover, despite its increasing popularity among language scholars in ESL and EFL contexts (Hashimoto, 2002; Khajavy et al., 2016; Kim, 2004; MacIntyre et al., 2002; Mystkowska-Wiertelak, 2021; Yashima et al., 2016a) willingness to communicate in a second language (L2 WTC henceforth) has been underexplored in the context of Pakistan. This article, therefore, aims to inform language scholars and language pedagogues in English as a second or foreign language (ES/FL) contexts in general and Pakistan in specific. The article first discusses the theoretical and empirical developments in L2 WTC highlighting the strengths and limitations of previous frameworks; second, it presents CDST as an alternative approach to examine the complex, dynamic and non-linear nature of L2 WTC. Finally, it discusses some future directions in L2 WTC research for novice researchers.

A note on methodology of this review

The present study is an historical overview of the conceptual and methodological developments in L2 WTC research. This review is close to what Macaro (2020) terms as narrative/traditional review which involves a review of the existing literature with a purpose to provide readers with an overview of extant literature on a topic; display a scholarly consensus or conflict; highlight and fill a research gap (p. 230). In a similar vein, the present review illustrates the major conceptual and methodological trajectory of L2 WTC research with a purpose to inform teachers and teacher educators, and provide directions to future researchers. The study acknowledges that narrative reviews inherently suffer from bias due to lack of information about the process and method of the selection of the studies reviewed as well as a lack of systematicity and depth in reviewing each study (Macaro et al., 2018). On our part, however, we have tried our best to minimize the bias by reporting on the cited literature with intellectual honesty.

The present study reviewed L2 WTC literature expanding between 1998 and 2020 with a specific focus on the major conceptual developments in L2 WTC research. The question this study aims to answer is: What conceptual developments L2 WTC research has gone through for the past two decades? The keywords used for searching relevant literature included: Willingness to communicate in a second/foreign language, willingness to communicate in English, factors affecting willingness to communicate, a dynamic systems approach to willingness to communicate in a second/foreign language, and complex dynamic systems theory. Studies for in-depth review were screened through reading the titles and abstracts. Based on the screening of abstracts (sometimes even a complete article), studies were included and excluded according to the criteria presented below:

Studies included:

Conceptual papers on L2 WTC

Studies that report on empirical data on L2 WTC in a second or foreign language

Conceptual papers and books on CDST/DST in applied linguistics

Studies into WTC that explicitly adopt a CDST perspective

Studies published in English language only

Some of the key studies authored or co-authored by the key authors in L2 WTC research, such as Peter D. MacIntyre, Clement Richard, Kang

Some of the key studies on the concept CDST authored/co-authored by Diane Larsen-Freeman, Zoltan Dornyei, and Kees de Bot.

Studies published in journals indexed in Web of Science journals

Studies excluded:

Studies reporting on WTC in L1.

Finally, thirty-one (n = 31) studies were shortlisted for the current review.

Figure 1.

MacIntyre et al. (1998) pyramid model.

Developmental trajectory of L2 WTC

L2 WTC is defined as a speaker’s voluntary decision to participate in a conversation involving the use of target language at a moment (MacIntyre et al., 1998). It is regarded as the most immediate predictor of L2 use and promoting L2 WTC has been designated as arguably an important goal of language pedagogy. First conceptualised as a situational variable, L2 WTC has undergone a conceptual evolution over the past two decades; from situational (MacIntyre et al., 1998) to process-oriented (Cao and Philip, 2006; Kang, 2005) to ecological (Cao, 2009; Peng, 2016) to dynamic (MacIntyre and Legatto, 2011; Peng, 2016; Wang, 2019; Wood, 2016) variable. One of the reasons L2 WTC has suffered from dichotomies is due to ontological approaches and methodological choices made by researchers to study the construct. Ontologically, L2 WTC has been conceptualized as a psychological variable immediately predicted by a set of other psychological variables, such as desire to communicate with a specific person and state self-confidence. The variables exist in a linear relationship with one another. For example, MacIntyre et al. (1998) conceptualized L2 WTC as a situational variable and immediate predictor of L2. According to MacIntyre et al.'s (1998) pyramid model, L2 WTC possessed dual characteristics of both a trait and a state variable (Figure 1).

The pyramid comprises six layers: The top three layers, such as situated antecedents, behavioral intention, and communication behavior, while the bottom three layers include motivational propensities and affective-cognitive context, and social and individual context. It is interesting to note that the model exposits change in variables on different timescales. For instance, the bottom layer represents relatively stable, trait-like and enduring variables that change over a long period of time, while the top three layers display variables that tend to change from moment-to-moment. MacIntyre (2020) highlighted two inconsistencies between the pyramid model and L2 WTC research: while the pyramid model conceptualized L2 WTC as both trait and state-like variable, it ‘did not address the measurement of state WTC and was silent on how it fluctuates overtime’ (p. 120); second, it did not describe how the trait-like factors, such as personality and culture, and state-like variables, such as self-confidence and desire to talk to a person, interact at any given moment.

In addition, from a CDST perspective, the variables existed in a linear relationship with other variables whereby it was directly predicted by desire to communicate and state self-confidence and indirectly predicted by trait-like variables, such as motivation and personality and social variables. Thus, the model postulated an unsurpassable disjuncture between psychological and contextual variables. Studies influenced by the pyramid model necessarily adopted a linear perspective and made methodological choices which eventually reinforced the trait-like view of L2 WTC rather than a situational one. For example, for a long time studies into L2 WTC used questionnaires originally developed to tap into L1 WTC.

Besides linear, cause-and-effect nature, the pyramid model suffered from a western-bias in the sense that while it specifically covered learners’ L2 WTC in the Canadian context, it did not offer insights into the nature of L2 WTC in other contexts, such as Asian. This task was fulfilled by Wen and Clement (2003) who offered a Chinese conceptualization of L2 WTC. According to Wen and Clement (2003) model, L2 WTC in a Chinese context displayed a significant influence on cultural context. Distinguishing between learners’ desire to communicate (DC) in L2 and willingness to communicate (WTC), the model implied a non-linear relationship between variables whereby any combination of motivational, affective, societal and personality variables could interrupt learners’ desire to communicate to transform into L2 WTC. Nevertheless, learners’ L2 WTC displayed strong inclination towards cultural factors, such as respect for teachers and other-directedness. While the model offered valuable insights into the nature of Chinese L2 WTC, it was limited in a number of ways: i) it only postulated psychological and contextual variables but missed out on linguistic factors, for example, perceived competence and code switching; ii) it failed to explain how these variables interacted and co-constituted L2 WTC.

Another model built upon the previous two models was presented by Kang (2005) in her seminal article. However, in contrast to her predecessors, Kang (2005) adopted a process oriented approach to examine the nature of WTC of four Korean English language learners in a university context. Using classroom observations and stimulated recall interviews, the study found that learners’ L2 WTC not only fluctuated from situation to situation but also from moment to moment within a conversation due to the interaction between psychological (excitement, responsibility and security) and contextual factors (i.e. topic, interlocutor and conversational context). In relation to the previous models postulating a static behavior of WTC, this model offered a dynamic view of WTC in a classroom context. Based on her empirical study, Kang (2005) presented a model which showed a dynamic and (implicitly) non-linear relation between underlying variables affecting L2 WTC. However, the model was marked by a linear relationship between situational and ultimate L2 WTC. That is, while the situational WTC was exposed to the interference of psychological and contextual variables, the ultimate WTC automatically followed its situational variant and was relatively static and stable. Thus, the process of learners’ development from situational to ultimate WTC was irreversible.

Kang’s (2005) model inspired a range of studies using ecological and sociocultural approaches to explore the underlying variables and the nature of their interaction (Cameron, 2020; Cao, 2009; Kang, 2005; Peng, 2014; Suksawas, 2011). These studies made a radical transition from the pyramid model in the sense that they conceived of L2 learning as a process rather than product and instead of relying exclusively on surveys and trait-oriented questionnaires, researchers employed a variety of qualitative methods, such as classroom observations and interviews, to explore the factors determining L2 WTC. For example, a significant contribution of ecological approach to understanding L2 WTC was the complexification of the context. Previous research accounted for only the macro-context as a relatively stable variable exerting a distal impact on L2 WTC through motivation and desire to communicate (Cetinkaya, 2005; Clement et al., 2003; Kim, 2004). Ecological approach, however, reconceptualised context as a dynamic variable and classified it into four mutually interacting layers, such as micro-(i.e. classroom), meso-(i.e. institution; university), exo-(i.e.family), and macro-layer (i.e. socio-cultural; economic) (Cao, 2009; King, 2016; Mercer, 2016; Peng, 2014, 2016). The studies showed that it’s not only the macro-layer that influences learners’ classroom communication; rather, all these four layers exert influence on learners’ communicational behavior. However, while ecological approach tells us how macro-, exo- and meso-layers exert their influence on an individual learners’ L2 development, it does not suggest how the micro-layer affects the learners’ behaviour in other layers of the context. For example, from the perspective of ecological theory we know that a learners’ socio-economic status and cultural upbringing (macro-layer) affects his/her communication behaviour in L2 classroom (micro-layer), but we are not informed as to how the learners’ classroom behaviour affects his socio-economic and cultural behaviour.

Complex dynamic systems theory: An alternative framework to understand L2 WTC

Ever since Larsen-Freeman (1997) introduced chaos and complexity approach to applied linguistics in her seminal article, complex dynamic systems theory (CDST) has attracted significant attention from scholars and researchers working on individual difference variables, such as motivation and L2 WTC. CDST offers a paradigmatic alternative to the conventional approaches to language and language pedagogy. First and foremost, CDST challenges the dichotomy, in fact, trichotomy between mind, body and world, that is, a segregation between learners’ psychology, physiology and context. Instead, CDST embraces all the three aspects and regards them as interdependent constituents of a whole rather than isolated, independent entities (Larsen-Freeman and Cameron, 2008). Previously L2 WTC was conceived of as a variable displaying both trait and state features whereby the underlying variables stood in a linear, cause-and-effect relationship with one another (MacIntyre et al., 1998; Wen and Clement, 2003). That is, L2 WTC was seen as directly determined by desire to communicate with a specific person or group and state self-confidence. CDST has offered an alternative framework to conceptualise L2 WTC.

According to CDST, L2 WTC is not only a complex but also a dynamic variable; that is, L2 WTC is constantly undergoing flux as a result of interaction between underlying variables. It is noteworthy that change in L2 WTC occurs at multiple timescales, such as seconds, minutes, hours, days, weeks, months, years, simultaneously. A broad distinction has been made between timescale and time window. The total time period in which a system is studied is said to be as a time window. For instance, learners’ L2 WTC studied over a semester wherein the semester is the time window. Timescales, on the other hand, refer to the study of weekly changes in her L2 WTC within that semester (de Bot, 2015). The two timescales are interpenetrating; that is, the overall L2 WTC of the learner in a semester depends on the ebbs and flows in her L2 WTC on a monthly, weekly and daily basis respectively. While changes on a longer timescale are determined by trait-like variables, such as personality, trait-like self-confidence and motivation, moment-to-moment variations are effected by state-like factors, such as interest (or a lack thereof) in topic, behaviour of interlocutor, mood swings, state-like motivation, classroom environment (Ali, 2017; Cameron, 2020; Syed and Kuzborska, 2019; Wang, 2019).

One of the important features of dynamic systems, in this case L2 WTC, is its sensitive dependence on initial conditions. Initial conditions refer to the psychological and/or physical state of a learner before and/or at the time of entering the class (Verspoor, 2015). It has been suggested that detailed information about a learner’s initial conditions is necessary to develop learners’ L2 WTC (de Bot and Larsen-Freeman, 2011). Research has shown that L2 WTC at a specific moment depends on a learner’s personality, level of proficiency in L2, trait-level motivation, mood, past experiences of the class, perception of the teacher, lack of preparation for class, physiological state, such as feeling hungry, experiencing headache or toothache, and even his dress-up before or at the time of entering the class on a particular day (Cao, 2014; Cetinkaya, 2005; Elwood, 2011; Oz, 2014; Syed and Kuzborska, 2019, 2020; Wang, 2019).

Unlike previous frameworks, CDST conceives of interaction of variables as nonlinear. Nonlinearity refers to the idea that change in L2 WTC at a given moment is not always proportional to the input. Sometimes, a minor change in an underlying variable occasions dramatic changes in learners’ L2 WTC, while at other times changes in a number of variables make little difference in learners’ L2 WTC (Syed and Kuzborska, 2020; Wang, 2019). For example, more teaching on the part of a teacher does not always result in more learning on the part of learners (Larsen-Freeman and Cameron, 2008). Higher motivation, to use another example, does not always cause higher L2 WTC as there are other variables, such as anxiety, self-perceived communicative competence, desire to communicate, that might hamper students’ L2 WTC. Variables underlying L2 WTC also demonstrate co-adaptability by adapting to the changes in other variables.

Lastly, from a CDST perspective, development of L2 WTC is characterised by attractor and repeller states. Attractor states represent preferable, stable states which L2 WTC enters as a result of interaction of facilitating or debilitating factors respectively (Larsen-Freeman and Cameron, 2008; de Bot et al., 2007; MacIntyre and Legatto, 2011; Waninge et al., 2014). Stability in CDST does not mean absence of change, or ‘stasis’, in the system. Rather, stability refers to a specific pattern of development in a system at a specific time. What might appear to be stability at a macro timescale, for instance, might be viewed as variability at a micro timescale. Thus, CDST is characterized by stable variability and variable stability (Hiver, 2015; Larsen-Freeman, 2020). Repeller states, on the other hand, refer to a pattern of development where L2 WTC tends to stay in a state for only a brief time. For instance, a student might not be willing to participate in a class discussion because he was not given an opportunity to speak a minute ago; but a few minutes later the class started discussing another topic which raised his L2 WTC once again. His L2 WTC is said to be in a repeller state as it did not stay low for too long. While previous models were either inherently incapable of accounting for conceiving L2 WTC in its complexity or merely unnoticeably indicated towards its complex nature, CDST provides a set of conceptual tools which facilitate an explanation of the complex behaviour of L2 WTC.

CDST and learner agency

Scholars have called for a reappraisal of CDST in order to account for the role of learner agency (Hoorie, 2015; Karimi-Aghdam, 2016; Larsen-Freeman, 2019). They argue that the current conceptualisation of CDST overplays the role of the interaction of underlying variables, while underplays the role of learner agency in determining L2 WTC. According to CDST, learners’ L2 WTC is determined by the underlying sub-structure, that is, interaction between underlying psychological, contextual and linguistic variables. It tends to imply that learners are passive subjects of the structural changes rather than active agents capable of shaping their own L2 WTC by intervening in the process of communication (Hoorie, 2015; Karimi-Aghdam, 2016; Larsen-Freeman, 2019). For instance, talking on CDST's beach ball analogy, Hoorie (2015) argues that while the beach ball enters and leaves attractor and repeller states depending entirely on the constellation of external factors, human agents can exercise their agency to resist entering into these states. Therefore, it has been proposed to conceive of learner agency as a dialectical phenomenon subject to both internal, i.e. free will, and external factors, i.e. environment.

Larsen-Freeman (2019) offers a comprehensive definition of learner agency as ‘optimizing conditions for one’s own learning and choosing to deploy one’s semiotic resources to position oneself as one would wish in a multilingual world’ (p. 71). She argues that learner agency is relational, emergentist, spatio-temporally situated, heterarchical and multidimensional phenomenon. That is, agency emerges and is achieved by learners in relation to ecological affordances that exist in a heterarchical relationship to each other whereby no single variable is more powerful than any other to determine learner agency. An illustration of learner agency is when some students proactively exploit opportunities to make use of L2, while others prefer to wait for opportunities to arise (Syed, 2016). Recently, Pennycook (2016) has advanced a posthumanist approach that extends the concept of agency to phenomena including places and artefacts. He argues that humans exist in the world with nonhuman beings, for example, animals, artefacts, nature and places. Thus, while humans exercise their agency in/on both the human and nonhuman worlds, the latter also exert their agency on humans and their actions. Recent studies has also shown how learners’ L2 WTC is shaped not only by the interlocutors but also by the distance between them and their interlocutors and the artefacts, such as book, as well as the quality of their shirt and shoes (Syed and Kuzborska, 2020; Wang, 2019; Wood, 2016).

Similarly, Karimi-Aghdam (2016) argues that the existing framework of CDST (he uses the term contextual DST) emphasises a bottom-up emergence of L2 and understates the role of cognition and learner agency in L2 development. Conceptualising interaction of variables that co-construct L2 use, contextual DST involves a teleological view of language development in the sense that it implies a direct path from language use to language development. In contrast, Karimi-Aghdam (2016) proposed a dialectical dimension to DST (Dialectical DST, DDST henceforth) which holds that L2 development is subject to changes at two developmental levels concurrently: micro-level (i.e. cognitive) and macro-level (i.e. dialogic interactions at social level). As per Karimi-Aghdam’s (2016) DDST, L2 development occurs as a result of reciprocal or circular causation between top-down causal forces (i.e. emergent mental forces) and bottom-up forces (i.e. environmental and pedagogical). Thus, L2 development is not only an effect of micro-causal forces but also serves as a macro-causal force affecting changes in the learning environment. Most notably, DDST does not consider L2 development as merely a bi-product of social interaction but a rather complex and purposive phenomenon subject to a dialectical relation between social dialogic interaction and human agency. Thus, a dialectical construal of DST enhances its explanatory power by removing the implicit conceptual lacunae and social dialogic bias. These developments have contributed a great deal to the rigour of CDST as a meta-theory and have opened up ways for a nuanced understanding of learners’ L2 WTC.

Some recent methodological developments

This dialectical construal of CDST expands our understanding of the dynamics of L2 WTC but also serves as a super-theory to explain the dynamics of dynamics that is CDST. This understanding has implications for novice researchers in the domain of language learner psychology in particular and language pedagogy in general. Adopting CDST approach will allow researchers to obtain an in-depth and holistic understanding of learners’ moment-to-moment changing L2 WTC, including learners’ styles, aptitudes, attitudes, and learning strategies, and the complex non-linear interaction of underlying factors determining ebbs and flows in learners’ communication behavior inside classroom. Owing to non-linear nature, the exact behaviour of L2 WTC is difficult to predict at a given moment. Therefore, it has been suggested that L2 WTC’s dynamic behaviour can only be studied retrodictively, that is, by studying the patterns of change in L2 WTC retrospectively (Dornyei, 2014; Larsen-Freeman, 2015).

Moreover, scholars have argued that correlational and cross-sectional research despite certain advantages, is not adequate to research L2 WTC from a dynamic perspective (Larsen-Freeman, 2020; MacIntyre, 2020). Thus, calls have been made to give-up ‘the bifurcated research agenda’ (Larsen-Freeman, 2015) and adopt more innovative methods of examining dynamic systems, such as L2 WTC. While cross-sectional studies allow us to understand changes in L2 WTC on a global scale across a group of participants, they do not inform us about how small changes occurred in individual learners’ L2 WTC at a particular time and what determined those changes. Therefore, ‘a view from below’ is needed, such as the use of individual longitudinal case studies, where the ‘messy little details’ lie (Larsen-Freeman, 2020: 5).

This demand was first addressed by MacIntyre and Legatto (2011) who designed a method, called idiodynamic, to examine the moment-to-moment changes in L2 WTC of L2 learners (n = 6) in a Canadian university context. Idiodynamic method is a retrospective method that involves obtaining video-recordings of learners’ L2 use while performing a task (or a series of tasks). As soon as they are done with the task, the learners are then asked to watch the video and rate the level of their L2 WTC throughout the task using software designed specifically for this purpose. After that, an interview is conducted with each participant in which they are asked to watch the video again along with the graphic data of their ratings to elaborate on the factors which determined his/her L2 WTC during the task (MacIntyre, 2012). The method has proved remarkably successful in capturing moment-to-moment fluctuations along with the factors underlying those variations and has inspired studies in various contexts already (e.g., MacIntyre and Serroul, 2015; Wood, 2016).

Using a CDST approach, MacIntyre and Serroul (2015) investigated the moment-to-moment variations in French L2 learners (n = 12) in a Canadian university context. The study employed the idiodynamic method involving French proficiency questionnaire, eight oral communication tasks followed by participants’ self-ratings and stimulated-recall interviews. In another study, Wood (2016) examined L2 WTC and fluency of Japanese English L2 learners (n = 4) in a university context in Canada. The study adopted an exploratory multiple case study while the data was obtained in three stages: 1) participants were given tasks, i.e. describing a busy street scene in L2; each participant’s performance was video-recorded; 2) participants watched their videos immediately after the task and drew graph depicting variations in their L2 WTC during the task that they had performed. Finally, participants were interviewed to explore the factors causing variations in participants’ L2 WTC during the task. These studies have paved the way for novice researchers to better comprehend the dynamic behaviour of learners’ WTC.

Larsen-Freeman (2020) proposes micro-developmental longitudinal studies that require denser data collection intervals. Past few years have witnessed the emergence of such studies adopting a variety of methods, including longitudinal multiple-case mixed method studies (Ali, 2017; Cameron, 2020; MacIntyre and Legatto, 2011; Mystkowska-Wiertelak, 2021; Peng, 2014; Saidi, 2018; Syed, 2016; Syed and Kuzborska, 2019; Wood, 2016; Yashima et al., 2016b) to capture the complex and dynamic nature of L2 WTC in various contexts (Ayers-Glassey and MacIntyre, 2019; Zarrinabadi and Tanbakooei, 2016). For instance, in her doctoral study Cameron (2020) conducted a longitudinal case study (18 months) to investigate L2 WTC of Iranian adult immigrant students (n = 10) in New Zealand, and their teachers (n = 10) combining DST and Ecological approaches. The study employed classroom observations, questionnaires, stimulated recalls and semi-structured interviews. Results showed that L2 WTC was influenced by external factors (such as, the teacher, texts and methods used, and class activities), and internal factors (such as, self-perceived English-speaking competence, confidence, anxiety, motivation, and personality).

Similarly, a comprehensive study by Yashima et al. (2016b) examined the influence of learner (n = 21) variables and contextual factors on L2 WTC in a Japanese context over a semester. The study adopted quite a complex research design including questionnaires to examine learners’ trait-level L1 WTC, anxiety, and language proficiency, classroom observations of learners’ communication in pairs, groups, and whole-class settings; learners completed reflective journals on the reasons why they were willing or unwilling to speak L2 in the discussion; questionnaires on participants’ level of anxiety were also administered after every class session; participants were administered questionnaires with a purpose to elicit their reflections on their participation in whole-class discussions throughout the semester. Finally, stimulated recall interviews were conducted with four of the 21 participants to dig deeper into the factors that affected their L2 WTC. Results of the study revealed that learners L2 WTC fluctuated concurrently on two timescales, that is, over the semester and from moment to moment within group discussion. Moreover, their WTC over the semester was affected by enduring variables, like personality and language proficiency, while at the momentary level it was influenced by state-like variables, including interlocutors’ behaviour and group dynamics (see also Syed and Kuzborska, 2019). However, studies explicitly adopting a CDST framework are still very rare.

Pedagogical implications

CDST perspective can help teachers to develop a better understanding of learners’ ID profile, for example, L2 WTC, motivation, emotions and personality. It will enable teachers to manipulate variables through instructional intervention to provide a conducive learning environment that strengthens factors that facilitate L2 learning. Biedroń and Pawlak (2016) suggest that teachers may explore learners’ styles by using a learning style survey originally designed by Cohen et al. (2001). This would enable them to make informed decisions with regard to selecting topics, designing tasks and adopting materials that might enhance learners’ L2 WTC inside class.

The study also highlights the importance of learner agency which has long been neglected in research in applied linguistics (Larsen-Freeman, 2019). It is, therefore, recommended to future scholars to examine the relation between learner agency and L2 WTC inside the classroom. That is, researchers can investigate how L2 WTC and agency reinforce/resist one another; for instance, how learners exercise their agency to avoid a negative attractor (or a repeller) state by making L2 use. Additionally, research has mainly focused on L2 WTC in speaking skills in classroom settings, while WTC in other skills, such as reading, writing and listening, especially in other settings, such as university canteen, social media, i.e. Facebook, Twitter, Snapchat, Instagram, WhatsApp and other digital contexts have been paid less attention to (Zarrinabadi and Tanbakooei, 2016). It is, therefore, recommended that scholars can explore learners’ L2 WTC in digital settings as Research has shown that students frequently engaging in virtual intercultural interaction in digital settings demonstrate higher willingness (Lee and Lee, 2019).

More specifically, due to the dismal situation of Covid-19, almost all colleges, schools and universities in the country have moved their classes online. While higher education has made online classes mandatory for students, a number of technical as well as pedagogical concerns have been raised by students and teachers. Based on my colleagues’ as well as my own observations, students’ oral participation in online classes has been lower than their level of participation inside a physical classroom. It, therefore, necessitates the exploration of the complex and dynamic nature of factors underlying students’ L2 WTC in online classes to make students’ as well as teachers’ experience of online learning and teaching effective rather than self-defeating.

Conclusion

The purpose of this review was to present a brief overview of some of the important conceptual and methodological developments in research on L2 WTC over the past two decades. L2 WTC is an immediate predictor of L2 use. Given that L2 use is a necessary condition for L2 learning, fostering L2 WTC becomes an important goal of language pedagogy (MacIntyre et al., 1998). While previous approaches to L2 WTC lent important insights into the stable, trait-like behaviour of L2 WTC, they lacked ‘consideration of the here-and-now experience of WTC’ (MacIntyre, 2020: 127). CDST provides a metatheoretical lens to examine the fluid and non-linear process of L2 WTC construction or destruction on a moment-to-moment timescale. Novice researchers can make use of a methodological toolkit, combining quantitative measures and idiodynamic method, video-/audio-recorded classroom observations, learners’ journals, StRs and even think-aloud protocols, to study the complexity, moment-to-moment variations and the relations between constellations of facilitating and debilitating variables underlying L2 WTC in a variety of communication settings, such as inside/outside classroom and digital/non-digital settings.

Supplemental Material

sj-pdf-1-rie-10.1177_00345237211031261 - Supplemental material for Willingness to communicate in a second language and the promise of complex dynamic systems theory: A narrative review

Supplemental material, sj-pdf-1-rie-10.1177_00345237211031261 for Willingness to communicate in a second language and the promise of complex dynamic systems theory: A narrative review by Hassan Syed, Shumaila Memon, Zulfiqar Ali Chachar, Sharique Zameer and Tanweer Shah in Research in Education

Footnotes

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.

ORCID iD

Hassan Syed

References

Ali

(2017) Investigating Pakistani university students' level of willingness to communicate (WTC) in English as a foreign language: A case study of students from the Shah Abdul Latif University, Khairpur, Sindh (SALU). Doctoral Thesis, Northumbria University, UK.

Ayers-Glassey S and MacIntyre PD (2019) Measuring willingness to communicate in first and additional languages: An annotated bibliography. Cape Breton University. Retrieved on 29 June 2021 from https://www.researchgate.net/profile/Samantha-Ayers-Glassey

Baker S and MacIntyre P (2003) The role of gender and immersion in communication and second language orientations. Language Journal 53(1): 65--96.

Biedroń

Pawlak

(2016) The interface between research on individual difference variables and teaching practice: The case of cognitive factors and personality. Studies in Second Language Learning and Teaching 6(3): 395–422.

Cameron

(2020) An ecological investigation of the willingness to communicate (WTC) in English of adult migrant learners from Iran in a New Zealand tertiary classroom. Doctoral Thesis, Auckland University of Technology, New Zealand. https://openrepository.aut.ac.nz/handle/10292/13146

Cao Y and Philip J (2006) Interactional context and willingness to communicate: A comparison of whole class, group and dyadic interaction. System 34: 480--493.

Cao Y (2014) A socio-cognitive perspective on second language classroom willingness to communicate. TESOL Quarterly 48(4): 789--814.

Cao

(2009) Understanding the notion of interdependence, and the dynamics of willingness to communicate. PhD Thesis, ResearchSpace@ Auckland, New Zealand. https://researchspace.auckland.ac.nz/bitstream/handle/2292/5584/02whole.pdf?sequence=2

Cetinkaya

(2005) Turkish college students’ willingness to communicate in English as a Foreign language. Available at: https://etd.ohiolink.edu/!etd.send_file?accession=osu1133287531&disposition=inline (accessed 19 October 2015).

10.

Clement R, Baker S and MacIntyre PD (2003) Willingness to communicate in a second language: The effects of context, norms, and vitality. Journal of Language and Social Psychology 22: 190--209.

11.

Cohen AD, Oxford RL and Chi JC (2001) Learning style survey. Retrieved on 29 June 2021 from http://www.mrspsela.com/uploads/2/1/8/7/21879712/_____learning_style_survey_all_students.pdf

12.

de Bot

(2015) A History of Applied Linguistics: From 1980 to the Present. London: Routledge.

13.

de Bot

Larsen-Freeman

(2011) Researching second language development from a dynamic systems theory perspective. In: Verspoor

De Bot

Lowie

(eds) A Dynamic Approach to Second Language Development. Amsterdam/Philadelphia: John Benjamin Publishing Company, pp.55–85.

14.

de Bot

Lowie

Verspoor

(2007) A dynamic systems theory approach to second language acquisition. Bilingualism: Language and Cognition 10(1): 7–21.

15.

Donovan

MacIntyre

P D

(2004) Age and sex differences in willingness to communicate, communication apprehension, and self-perceived competence. Communication Research Reports 21(4): 420–427.

16.

Dornyei

(2009) The Psychology of Second Language Acquisition. Oxford: Oxford University Press.

17.

Dornyei

(2014) Researching complex dynamic systems: ‘Retrodictive qualitative modelling’ in the language classroom. Language Teaching 47(1): 80–91.

18.

Dornyei

MacIntyre

Henry

(2015) Introduction: Applying complex dynamic systems principles to empirical research on L2 motivation. In: Dornyei

MacIntyre

Henry

(eds) Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.1–10.

19.

Dornyei

Ryan

(2015) The Psychology of Language Learner Revisited. London: Routledge.

20.

Elwood

(2011) Enriching structural models of L2 willingness to communicate: The role of personality, ego, permeability, and perceived distance. Available at: gradworks.umi.com/34/57/3457882.html (accessed 9 November 2015)

21.

Hashimoto Y (2002) Motivation and willingness to communicate as predictors of reported l2 use: the Japanese ESL context. Second Language Studies 20(2): 29--70.

22.

Hiver

(2015) Attractor states. In: Dornyei

MacIntyre

Henry

(eds) Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.20–28.

23.

Hoorie

(2015) Human agency: Does the beach ball have free will? In: Dornyei

MacIntyre

Henry

(eds) Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.55–72.

24.

Kang

(2005) Dynamic emergence of situational willingness to communicate in a second language. System 33(2): 277–292.

25.

Karimi-Aghdam S (2016) A Dialectical Reading of Dynamic Systems Theory: Transcending Socialized Cognition and Cognized Social Dualism in L2 Studies. Language and Sociocultural Theory 3(1): 55--82. doi:10.1558/lst.v3i1.27547

26.

Khajavy

G H

Ghonsooly

Fatemi

A H

, et al. (2016) Willingness to communicate in English: A microsystem model in the Iranian EFL classroom context. Tesol Quarterly 50(1): 154–180.

27.

Kim SJ (2004) Exploring willingness to communicate (WTC) in English among Korean EFL (English as a foreign language) students in Korea: WTC as a predictor of success in second language acquisition. Retrieved September 2, 2013, from https://etd.ohiolink.edu/ap:0:0:APPLICATION_PROCESS=DOWNLOAD_ETD_SUB_DOC_ACCNUM:::F1501_ID:osu1101267838,attachment

28.

King

(2016) Introduction to the dynamic interplay between context and the language learner. In: King

(ed.) The Dynamic Interplay Between Context and the Language Learner. Hampshire: Palgrave Macmillan, pp.11–28.

29.

Larsen-Freeman

(1997) Chaos/complexity science and the second language acquisition. Applied Linguistics 18(2): 141–165.

30.

Larsen-Freeman

(2015) Ten ‘lessons’ from complex dynamic systems theory: What is on offer? In: Dornyei

MacIntyre

Henry

(eds) Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.11–28.

31.

Larsen-Freeman

(2019) On language learner agency: A complex dynamic systems theory perspective. The Modern Language Journal 103: 61–79.

32.

Larsen-Freeman

(2020) Chaos/complexity theory for second language acquisition/development. The Encyclopedia of Applied Linguistics. Epub ahead of print 15 June 2020. DOI: 10.1002/9781405198431.wbeal0125.pub2

33.

Larsen-Freeman

Cameron

(2008) Complex Systems and Applied Linguistics. Oxford: Oxford University Press.

34.

Lee

(2019) Affective factors, virtual intercultural experiences, and L2 willingness to communicate in in-class, out-of-class, and digital settings. Language Teaching Research 24(6): 1–21.

35.

Macaro

(2020) Systematic reviews in applied linguistics. In: Mckinley

Rose’s

(eds) The Routledge Handbook of Research Methods in Applied Linguistics. New York: Routledge, pp.230–239.

36.

Macaro

Curle

Pun

, et al. (2018) A systematic review of English medium instruction in higher education. Language Teaching 51(1): 36–76.

37.

MacIntyre P, Babin P and Clement R (1999) Willingness to communicate: antecedents and consequences. Communication quarterly 47(2): 215--229.

38.

MacIntyre

(2012) The idiodynamic method: A closer look at the dynamics of communication traits. Communication Research Reports 29(4): 361–367.

39.

MacIntyre

(2020) Expanding the theoretical base for the dynamics of willingness to communicate. Studies in Second Language Learning and Teaching 10(1): 111–131.

40.

MacIntyre

Babin

Clement

(1999) Willingness to communicate: Antecedents and consequences. Communication Quarterly 47(2): 215–229.

41.

MacIntyre

Baker

Clement

, et al. (2001) Willingness to communicate: Social support and language-learning orientation of immersion students. Studies in Second Language Acquisition 23(3): 369–388.

42.

MacIntyre PD, Baker SC, Clement R, et al. (2002) Talking in order to learn: Willingness to communicate and intensive learning programs. The Canadian Modern Language Review 59 (4): 589--607.

43.

MacIntyre

Clement

Dornyei

, et al. (1998) Conceptualizing willingness to communicate in a L2: A situational model of L2 confidence and affiliation. The Modern Language Journal 82(4): 545–562.

44.

MacIntyre

Legatto

(2011) A dynamic system approach to willingness to communicate: Developing an idiodynamic method to capture rapidly changing affect. Applied Linguistics 32(2): 149–171.

45.

MacIntyre

Serroul

(2015) Motivation on a per second timescale: Examining approach-avoidance motivation during L2 task performance. In: Dornyei

MacIntyre

Henry

(eds) Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.109–138.

46.

Mercer

(2016) The contexts within me: L2 self as a complex dynamic system. In King

(ed.) The Dynamic Interplay Between Context and the Language Learner. Hampshire: Palgrave Macmillan, pp.11–28.

47.

Mystkowska-Wiertelak

(2021) Fluctuations in willingness to communicate during a semester: A case study. The Language Learning Journal 49(1): 1–12.

48.

(2014) Big five personality traits and willingness to communicate among foreign language learners in Turkey. Social Behavior and Personality: An International Journal 42(9): 1473–1482.

49.

Peng

(2014) Willingness to Communicate in Chinese EFL University Classroom: An Ecological Perspective. Bristol: Multilingual Matters.

50.

Peng

(2016) The context-sensitivity of the self-concept and willingness to communicate in Chinese EFL classroom: A case study. In King

(ed.) The Dynamic Interplay Between Context and the Language Learner. Hampshire: Palgrave Macmillan, pp.84–103.

51.

Pennycook

(2016) Posthumanist applied linguistics. Applied Linguistics 39(4): 445–461.

52.

Saidi

SBB

(2018) Willingness to communicate in English among Malaysian undergraduates on campus: An identity-based motivation perspective. http://etheses.whiterose.ac.uk/22546/1/Siti%20Saidi_Final%20thesis%20submission_311218.pdf (accessed 9 March 2021).

53.

Suksawas

(2011) A sociocultural study of EFL learners’ willingness to communicate. http://ro.uow.edu.au/cgi/viewcontent.cgi?article=4429&context=theses (accessed 7 October 2015).

54.

Syed H and Kuzborska I (2019) Understanding the nature of variations in postgraduate learners' willingness to communicate in English. Cogent Education 6(1): 1--22. DOI: 10.1080/2331186X.2019.1606487

55.

Syed H (2016) Dynamics of variables underlying Willingness to Communicate (WTC) in English (L2): A case study of postgraduate business students at a university classroom in Sukkur, Pakistan. PhD thesis, University of York.

56.

Syed H and Kuzborska I (2020) Dynamics of factors underlying willingness to communicate in a second language. The Language Learning Journal 48(4): 481--500. DOI: 10.1080/09571736.2018.1435709

57.

Verspoor

(2015) Initial conditions. In: Dornyei

MacIntyre

Henry

(eds), Motivational Dynamics in Language Learning. Bristol: Multilingual Matters, pp.1–10.

58.

Wang

(2019) The dynamics of willingness to communicate: A study of Chinese high school students’ English language learning during communicative task performance. PhD Thesis, ResearchSpace@ Auckland, New Zealand. Available at: file:///C:/Users/Hassan%20Shah/Downloads/whole%20(1).pdf

59.

Waninge

Dornyei

De Bot

(2014) Motivational dynamics in language learning: Change, stability, and context. The Modern Language Journal 98(3): 704–723.

60.

Weaver

(2009) Trends in research on willingness to communicate. In: Stoke

(ed.) JALT2008 Conference Proceedings. Tokyo: JALT, pp.53–66.

61.

Wen

Clement

(2003) A Chinese conceptualisation of willingness to communicate in ESL. Language, Culture and Curriculum 16(1): 18–38.

62.

Wood

(2016) Willingness to communicate and second language speech fluency: An idiodynamic investigation. System 60: 11–28.

63.

Yashima

Ikeda

Nakahira

(2016a) Talk and silence in an EFL classroom: Interplay of learner and context. In: King

(ed.) The Dynamic Interplay Between Context and the Language Learner. Hampshire: Palgrave Macmillan, pp.84–103.

64.

Yashima T, MacIntyre PD and Ikeda M (2016b). Situated willingness to communicate in an L2: Interplay of individual characteristics and context. Language teaching research 22(1): 115–137.

65.

Zarrinabadi

Tanbakooei

(2016) Willingness to communicate: Rise, development, and some future directions. Language and Linguistics Compass 10(1): 30–45.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB