The collective attempt of 17 established L2 researchers to synergize their work despite their sometimes-fundamental theoretical starting points has resulted in a valuable SI in The Modern Language Journal (Michel et al., 2025), in the tradition of The Douglas Fir Group (2016) and other collaborative achievements, like the Modern Language Journal 2019 Supplement. These initiatives were conceived to offer a broader perspective on second language learning and teaching, moving beyond simplified approaches that focus on either cognitive or social aspects of language acquisition. Such (over-) simplified perspectives offer the advantages of theoretical transparency and provide opportunities for systematic research using well-established methodological techniques. However, the reality of language development may be too diverse, fuzzy, dynamic, and multidimensional to be represented by a single “neat and clear” framework. Therefore, our task as researchers in the field of second language development is to strike a balance between the complexity of real-world phenomena and our ability to answer relevant research questions about the principles and mechanisms of the developmental process. From this perspective, the collaborative synergetic endeavor has successfully generated a productive framework that promises to move the field forward. The supportive commentaries in the Synergies SI, as well as the subsequent philosophical commentaries, clearly move in that direction. Whether one chooses to focus on similarities or differences across theories is a matter of preference. The resulting debate is timely, hopefully productive, and perhaps even urgent, but it does not warrant a sense of crisis, as suggested by Lantolf et al. (2025). In this short contribution, I will elaborate on the need for a holistic perspective, with particular emphasis on a Complex Dynamic Systems Theory (CDST) framework. One of the challenges raised in critical discussions of CDST, including the paper by Lantolf et al. (2025), concerns the interconnectedness of the phenomena that could be described as “factors affecting SLA.” From a holistic perspective, however, development cannot be conceptualized in terms of discrete factors that affect a process, but rather in terms of “coupled subsystems” that collectively constitute the process itself. In contrast to modular models that assume input–output relations between relatively independent components, CDST conceptualizes development as emerging from a complex network of tightly interdependent subsystems. These subsystems are not external influences on the system; collectively, they are the system. Since language development is a rather abstract process, the coupling of subsystems is best illustrated using a more concrete analogy. The human digestive system is an example of such a multi-stage network: the different organs, such as the mouth, stomach, intestines, liver, and pancreas, each functioning as an open system embedded within broader regulatory mechanisms, including neural and hormonal control. Although each component can be studied in isolation, such analyses cannot yield a comprehensive understanding of digestion as a coordinated process (see Mackie, 2019). Yet, while complex systems may indeed consist of “an infinite number of changing components,” some degree of reduction is unavoidable if scientific understanding is to be achieved. Scientific models are necessarily reductive representations of complex phenomena, and hardly any researcher would argue that science must mirror the full complexity of reality to explain it. The critical question is not whether models simplify, but whether a given simplification adequately captures the phenomena of interest without stripping away essential features of interacting and evolving systems. Reduction becomes problematic when core characteristics are replaced by assumptions of linearity, population homogeneity, or additive effects that obscure interaction and change over time. At the same time, relatively simple models can be powerful tools for identifying underlying mechanisms, provided that they meaningfully represent developmental processes (Caspi, 2010; Van Geert, 2003, 2023; Verspoor Verspoor Freeborn et al., 2023). Another point of criticism concerns the predictability of second language development. In response to the claim that prediction is difficult, if not impossible, Lantolf et al. (2025) argue that “if this were the case, it is hard to imagine how life would be possible, since survival depends on the ability to predict changes in the environment, which is only possible if changes repeat specific laws of the world (Toomela, 2020).” At a very global level, development in groups of learners may indeed be predictable in terms of broad tendencies or stages, much like first language development progresses from babbling to first words and eventually to sentence-level speech, or how labial sounds and open syllables typically precede velar sounds and coda consonant clusters. In second language development, however, such stage-based regularities are far less robust. Variation in learning conditions, sociocultural contexts, age of onset, and individual learner characteristics substantially reduces the predictability of uniform developmental stages, even when certain general patterns can be identified. Crucially, this type of stage-based prediction is not the primary focus of CDST-informed research or of other holistic approaches to development. Rather than emphasizing what develops, these approaches are primarily concerned with how development unfolds over time, that is, with the processes and mechanisms through which change emerges (see van Geert Molenaar, 2015). As a result, meaningful insights into developmental processes require research designs that capture intraindividual dynamics and can only be obtained through (multiple) dense case studies. Although through the use of analytic techniques such as cluster analyses, groups of learners who show similar behavioral patterns may emerge from the data (Peng et al., 2022), the developmental process itself remains fundamentally individual. From a CDST perspective, predictability therefore resides not in fixed sequences of outcomes, but in patterns of interaction, variability, and change over time. Referring to the three ruptures mentioned by Yurevich (2009) as quoted in Lantolf et al. (2025), the number of theories is large, and a possible source of fragmentation. However, it was precisely this diversity that motivated the Synergies project. By bringing together scholars working from a range of theoretical perspectives, the project deliberately foregrounded points of convergence rather than divergence, with the aim of identifying shared assumptions and complementary insights across frameworks. With respect to the “diagonal rupture,” the critique concerning the gap between theory and practice is well taken. There remains an unfortunate distance between theoretical accounts of second language development and their direct applicability to teaching practice. At the same time, this gap is characteristic of much foundational research. The primary goal of theory-driven research is to develop an empirically grounded understanding of developmental processes; the investigation of pedagogical implications typically follows as a subsequent, though essential, step. Many of the central research questions addressed in the Synergies SI originate in challenges encountered in instructional contexts, and explicit attention to pedagogical implications is built into each contribution. The diversity of theoretical perspectives also leads to a wide diversity of pedagogical implications. This is not surprising, given the equally wide range of instructional contexts in which language learning takes place. Teaching is not a uniform practice, nor can it be reduced to a single notion of “the classroom.” Instructional contexts vary substantially across dimensions such as language typology, learning goals, proficiency levels, educational settings (e.g., second vs. foreign language contexts), and learner characteristics, ranging from young children in formal schooling to adult professionals in business contexts, and refugees acquiring a language for immediate survival. Moreover, language learning frequently occurs beyond formal classroom settings. In such a heterogeneous landscape, it is unrealistic to expect unified pedagogical prescriptions. While calls for “meaningful professional standards or coherent preparation programs” (Lantolf et al., 2025, p. 999) may be appropriate in idealized contexts, the rejection of “one-size-fits-all approaches to language teaching” (Atkinson et al., 2025, p. 58) is both empirically and pedagogically justified. In their commentary, Lantolf et al. emphasize the diversity of methodologies used in the SI and argue for the need to establish shared methodological foundations to advance the field. This goal is broadly supported here, with the qualification that methodological coherence should begin with a clear distinction between product-focused and process-focused research questions. Research methods can only address the questions they are designed to answer. Cross-sectional group studies, employing appropriate statistical techniques and systematic qualitative analyses, are well-suited to examining relationships among variables at a single point in time. However, questions concerning developmental change require longitudinal designs that trace individual trajectories over time. Given the dynamic, context-sensitive, and individually variable nature of second language development, both the feasibility and desirability of a single, general theory of SLA remain open to question. If any theoretical framework is likely to accommodate this diversity, it would be a holistic, process-oriented meta-theoretical approach of the kind advanced in the Synergies SI (Atkinson et al., 2025). Rather than relying on population averages or global outcome measures, such approaches take the individual learner as the primary unit of analysis and seek to understand development through dense longitudinal data that capture change as it emerges. Progress in second language research depends on sustained dialogue about theories, methods, and applications across researchers using differing theoretical assumptions, from other disciplinary traditions, and adjacent fields, as well as within the field of SLA itself. Scholarly exchange of this kind is essential for advancing our understanding of complex phenomena such as language development. The Modern Language Journal provides a very valuable forum for open and constructive debate. The aim of the Synergies SI was not to establish full consensus on language learning and teaching, nor to advance a single, unified theory of SLA. Rather, its purpose was to examine the extent to which scholars working from diverse perspectives might converge on a set of fundamental assumptions and to explore these points of convergence in dialogue with seemingly incompatible theoretical positions. Both the contributions to the Synergies SI and the accompanying commentaries suggest that this objective has been at least partially realized. The new commentary by Lantolf and colleagues contributes productively to this ongoing exchange by critically engaging with the assumptions, implications, and limits of the perspectives represented in the SI. Future discussions will hopefully promote mutual understanding of our theories, perhaps agree to disagree, and, who knows, move toward the ultimate synergy of theories of second language learning and teaching, but it remains essential for the advancement of the field.
Wander Lowie (Fri,) studied this question.