Introduction

This work provides a chronological and thematic account of empirical studies and position papers on second language (L2) writing scholarship from a complex dynamic systems theory (CDST) perspective. As a theoretical framework, CDST was formally introduced into applied linguistics research by Diane Larsen-Freeman in 1994 (Larsen-Freeman, Reference Larsen-Freeman1994). However, more than a decade passed before CDST-L2 writing studies emerged in the literature, with Larsen-Freeman (2006)Footnote ^* frequently cited as the first related publication. Initially, scholarship focused primarily on the quality of linguistic output (e.g., measures of complexity, accuracy, and fluency, or CAF) in North American and European contexts. Since these early foci, studies have expanded to cover a range of constructs and contexts that employ increasingly sophisticated and diverse research methods (for a recent collection of studies, see Fogal & Verspoor, Reference Fogal and Verspoor2020). In this time, a CDST approach to L2 writing research has matured alongside a general CDST view of language change that has contributed, through empirical studies, to understanding the nonlinear, adaptive, context dependent, and complex and dynamic nature of L2 development (see Hiver et al., Reference Hiver, Al-Hoorie and Evans2021, for an overview).

Alongside emergentist approaches to language change such as usage-based linguistics and connectionism, a complexity theory framework emphasizes the interconnected nature of the human experience and, in the present context, the dialogue informing the science that endeavors to study it. Accordingly, generating a thematic timeline of scholarly activity associated with CDST and L2 writing is no straightforward task. A CDST perspective requires, for instance, that one acknowledge straightaway that any thematic strand—including one linked to a metatheory like CDST—is necessarily comprised of an inter-and intra-disciplinary foundation that predates and informs the integration of CDST and L2 writing scholarship. Accordingly, like other threads of applied linguistics research, much of what is listed below is closely connected to previous work. For example, such precursors include a vast body of scholarship on corpus linguistics (see Flowerdew, Reference Flowerdew, O'Keeffe and McCarthy2010) and L2 writing (see Manchón & Matsuda, Reference Manchón and Matsuda2016) that, through works such as Wolf-Quintero et al. (Reference Wolfe-Quintero, Inagaki and Kim1998) and Norris and Ortega (2009*), have inspired a notable collection of CDST studies exploring the CAF triad. In addition, researchers in this domain have also drawn from CDST scholarship outside of applied linguistics. For example, Thelen and Smith (Reference Thelen and Smith1994), Davis and Sumara (Reference Davis and Sumara2006), Morin (Reference Morin2008), and Byrne and Callaghan (Reference Byrne and Callaghan2014) have all had a direct impact on how L2 writing scholars approach writing phenomena through a CDST lens (influencing, for example, epistemology and methods). Moreover, expansive discussions about CDST and its (a) place and utility as a comprehensive theory of language development (de Bot et al., 2007*; Herdina & Jessner, Reference Herdina and Jessner2002; Larsen-Freeman, Reference Larsen-Freeman, Ortega and Han2017; Larsen-Freeman & Cameron, 2008*), (b) related methods (Hiver & Al-Hoorie, Reference Hiver and Al-Hoorie2016, Reference Hiver and Al-Hoorie2020; MacIntyre et al., Reference MacIntyre, MacKay, Ross, Abel, Ortega and Han2017; Verspoor et al., 2011*), and (c) pedagogical value (Larsen-Freeman, Reference Larsen-Freeman2012; Levine, Reference Levine2020; Mercer, Reference Mercer and Hall2016; Verspoor, Reference Verspoor, Ortega and Han2017) have all impacted relevant L2 writing research. Accordingly, as influence can be drawn from an expansive network that emphasizes the interconnected nature of a CDST framework, a comprehensive account of related relationships is beyond the scope of the present context.

Moreover, even within the sub-discipline it is impractical to trace the extensive body of scholarship on CAF; for example, considering the need to account for nascent interests in other features writing scholars have recently pursued (see, for instance, attention to agency in Byrnes, 2020a*, Nitta, 2020*, and Nitta & Baba, 2018*; a focus on individual differences in Huang et al., 2021*, Lowie & Verspoor, 2019*, Nitta, 2020*, and Nitta & Baba, 2018*; or attention to written corrective feedback in Fogal et al., 2020* and Lee et al., 2021*). As such, given the vast number of publications under the wider CDST umbrella, only those that were explicitly CDST- and L2 writing-focused in their orientation are included below, with few notable exceptions (see Norris & Manchón, 2012* and Norris & Ortega, 2009*). Further, complementary approaches (e.g., sociocultural or ecological frameworks) are not presented, nor are works published in languages other than English. Rather, the timeline draws from widespread L2 contexts (e.g., Canada, China, Indonesia, Italy, Japan, Thailand, the Netherlands, South Korea, and the United States), across a range of second and foreign languages (e.g., Chinese, Dutch, English, Finnish, and German) and is comprised primarily of journal articles and book chapters, representing both seminal works and studies by emerging scholars. Collectively, the selections below (a) highlight the scope of scholarship despite the area's recent (and CAF-focused) beginnings, (b) emphasize the momentum and interest this approach presently enjoys, and (c) underscore the utility of the framework for researching L2 writing.

These areas of attention have also been underpinned by two distinct motivational profiles driving research: the first investigates L2 writing data to study writing development (e.g., Baba & Nitta, 2014*; Fogal, 2019*; Macqueen, 2012*), and a second explores L2 writing performance to better understand general language development (e.g., de Bot et al., 2007*; Huang, et al., 2020*; Larsen-Freeman, 2006*). By way of a brief historical overview, the literature within these threads reveal early (and ongoing) attention to variability as a prominent feature of analysis, with initial and continued attention concentrated on patterns of interaction and variability at inter- and intra-individual and group levels across time as characteristics of dynamic systems and markers of development (see early works by Verspoor, Lowie and colleagues). This early attention to variability was developed largely through research on the CAF triad (see studies by Verspoor and colleagues as well as a collection of works by Housen and Bulté) and primarily in the European context. While said foci remain an important feature of CDST-L2 writing studies (e.g., Huang et al., 2021*), later publications expanded beyond these areas in something of a shift away from learner output to research that takes a broader, less granular approach to systems thinking than initial trends indicated (i.e., moving beyond syntax-level interactions that chart how, for example, accuracy and fluency co-develop) to include, for instance, context as a meaningful feature of CDST research (e.g., Hepford [Reference Hepford, Fogal and Verspoor2020] on social ecosystems; Nitta [2020*] on study abroad; Byrnes [2020a*, Reference Byrnes and Manchón2020b] and Fogal [2020a*, Reference Fogal, Fogal and Verspoor2020b] on curriculum as a construct of investigation).

These foci were complemented by new or renewed attention to research methods germane to investigating complex systems and L2 writing (i.e., methods that are systems oriented and process—rather than product—focused: e.g., biographical retrodiction [Nitta, 2020*], system mapping [Fogal, Reference Fogal, Fogal and Verspoor2020b, 2021*], a multilevel synchrony method [Rosmawati, Reference Rosmawati, Fogal and Verspoor2020], renewed attention to phase shift analyses [Hepford, Reference Hepford, Fogal and Verspoor2020], and mixed method approaches [Baba, Reference Baba, Fogal and Verspoor2020; Gilmore & Gánem-Gutiérrez, Reference Gilmore, Gánem-Gutiérrez, Fogal and Verspoor2020]). These and other methods (see MacIntyre et al., Reference MacIntyre, MacKay, Ross, Abel, Ortega and Han2017; Verspoor et al., 2011*) are needed as working with CDST concepts (e.g., co-adaptation, emergence, and attractor states—see Larsen-Freeman & Cameron, 2008*, for a rendering of these concepts) require researchers to reconceptualize conventional approaches to related studies (see MacIntyre et al., Reference MacIntyre, MacKay, Ross, Abel, Ortega and Han2017). For example, high frequency data collection points help shed light on adaptive, developmental processes, while accounting for context enables a systems-orientated, relational understanding of L2 developmental phenomena—both core foci of CDST research. Such efforts have led to advances in L2 writing scholarship, including complementary and at times novel perspectives on (a) task-based learning and task-based repetition (see a collection of studies by Baba and Nitta and Verspoor, Lowie, and colleagues), (b) dynamic usage-based linguistics (Huh et al., 2018*; Verspoor & Smiskova, 2012*; Verspoor et al., 2012*; Vyatkina et al., 2015*), (c) multilingualism (Cenoz & Gorter, 2011*; Herdina & Jessner, Reference Herdina and Jessner2002; Huang et al., 2020*), (d) the pedagogical potential of CDST-inspired findings for students and instructors (Huh et al., 2018*; Jackson & Park, 2020*; Nitta & Baba, 2014*; Tin, 2011*; Verspoor, Reference Verspoor, Ortega and Han2017) and curriculum development (Byrnes, 2020a*, Reference Byrnes and Manchón2020b; Fogal, Reference Fogal, Fogal and Verspoor2020b, 2021*), and (e) psychological factors (e.g., self-regulation and agency) informing writing development (Han & Hiver, 2018*; Jackson & Park, 2020*; Nitta & Baba, 2015*, 2018*; Nitta, 2020*; Wind & Harding, Reference Wind, Harding, Lowie, Michel, Rousse-Malpat, Keijzer and Steinkrauss2020). These themes have helped complement and frame current understandings and research designs, respectively, into L2 writing development, and have forged new paths forward (Bulté & Housen, 2020b*; Byrnes, 2020a*; Huang et al., 2021*). In addition, this research framework has captured the attention of young scholars and increasingly informs research projects at the graduate level.

This timeline traces how CDST-L2 writing research has arrived at its present state and outlines recent arguments for moving it forward. Accordingly, the timeline serves not only to describe the past and present, but also to outline how prominent voices in the field envision its continued development. This work, then, will help those familiar with the scholarship to consolidate their thoughts, and will assist readers eager to join the conversation to understand how this relatively new approach to L2 writing research has established itself in its present form. Such a historical and thematic perspective can serve as a helpful precursor for charting a path forward. To support this endeavor, the following codes are employed to assist readers with tracking development and change in the field.