Complex Dynamic Systems Theory as a meta-theory for language education

If you’ve been following this blog long enough, you know of my long-standing interest in Complex Dymamic Systems Theory (CDST), a way of looking at teaching and learning as interconnected, constantly evolving processes. Earlier this month, Juup Stelma and I published an article in TESOL Journal, entitled Revisiting complex dynamic systems theory: Empowering language teachers and teaching. You can read the article¹ or download a pre-print by clicking the buttons below. In this post, I would like to discuss why we wrote the article, and share some thoughts about what TESOL can gain from a complexity² perspective. What I hope to manage, with this post, is to explain why I think that a complexity perspective, when properly deployed, can shift the dynamics between people who have traditionally dictated how we should teach and people who actually teach languages.

---

How can Complex Dynamic Systems thinking help TESOL?

A persistent challenge in language education is the disconnect between teaching practice and the theory that purportedly informs it. There are many reasons for this, some practical and some political, and it is not my intention to discuss these at length here (but if you’d like to find out more, you could take a look at this post). Two reasons that stand out, however, are the following: (a) language education theory is too fragmented; and (b) language education theory is too specialized.

TESOL as a synthesis of disciplines

I have written elsewhere that I view language education as a practical endeavour informed by theoretical knowledge from three domains: applied linguistics, language education psychology, and language pedagogy. Such an interdisciplinary knowledge base comes with a challenge, though: even when we view the informing disciplines from the same vantage point (that of a language education professional), they are still distinct bodies of knowledge. Linguists (even applied linguists) work within their specific disciplinary background, they draw on their theories and they use terminology that is different from that used by psychologists and education specialists. The same, of course, applies to the other disciplines as well.

Paradoxically, in TESOL, deeper knowledge widens the theory-practice gap in ways that are not seen in other, more unitary fields.  

What this means, to me, is that even if a teacher wanted to deepen their knowledge of the informing theory, they would likely find it hard to do this in a balanced way. Professional development courses are typically designed to help us specialize in one of the informing disciplines of language education (or most likely in a niche within one of these disciplines). For instance, a teacher attending a postgraduate course might develop a very detailed understanding of Second Language Acquisition, or they might specialize in critical pedagogy or in motivation theory. But as they deepen their knowledge in any of those areas, they will move away from the synthetic outlook that must inform the practice of language teaching. Thus, somewhat paradoxically, deeper knowledge widens the theory-practice gap in ways that are not present in other, more unitary fields.    

Complexity as a meta-theory

This is where I believe that complex dynamics systems theory can play a useful role. I would argue that, despite differences in disciplinary traditions, linguistics, psychology, and pedagogy share the same epistemological and ontological assumptions. This means that they have a shared understanding of what things are, how we can learn about them and –I wish to put forward– how we can combine them. Regardless of where we position ourselves in the constellation of disciplines, we all have to deal with similar phenomena and make sense of them in similar ways. I would also like to suggest that Complex Dynamics Systems Theory (CDST) gives us the epistemic tools to make sense of these connections. In other words, CDST works best as a meta-theory (a ‘theory of theories’) that provides structure and unity to what would otherwise be a fractured universe of knowledge.

This is, perhaps, clearer when shown through examples. For instance, linguists, psychologists and education theorists will all observe how important starting points are in shaping future outcomes. We will variously talk about the richness or poverty of input during the sensitive period of language development, about early formative experiences, and about the effects of parental influence in early childhood –but what we are all describing are different manifestations of what CDST calls sensitivity to initial conditions. Similarly, we will all struggle with deciding which factors are directly relevant to whatever it is that we are studying and which factors are contextual –what CDST calls the boundary problem. You can extend this list to include things like patterns of stable behaviour (attractor states), change that is neither entirely predictable nor completely random (non-linear change), dynamic stability, sudden re-organisations etc.

How we might use Complex Dynamic Systems Theory in TESOL

The main point of my argument is that an outlook informed by Complex Dynamic Systems Theory can help us address the theory-practice gap in TESOL. We cannot answer this challenge just by arguing why theoretical knowledge is not useless (as some anti-intellectuals would like to have us think). What I am suggesting is that, while we keep learning about one or another field –as dictated by our strengths, needs and interests–, a complexity perspective helps us retain the conceptual anchoring that tethers our developing understandings to other disciplines and to teaching itself.  

Some early challenges

In some of our previous work, several colleagues and I have tried to show how the tools of complexity theory can become relevant to teaching. In 2015, we³ held the first Manchester Roundtable on Complexity and ELT. There, we tried to show how teachers can transfer tools from Complex Dynamic Systems Theory to the description of phenomena that we see in everyday TESOL practice. I suppose that one could call that a good start, and it was certainly the beginning of rewarding professional connections. However, the uptake of our work (published in this article) has been underwhelming.

I believe that part of the reason for this is that we have been pitching complexity in the wrong way. We were successful in showing similarities across disciplines, and between theory and practice. However, we were less successful in explaining them, and we were perhaps too enthusiastic as we showed so many parallels that we overwhelmed intake. Still, I do think that we did manage to make a case for the potential of CDST, and also to raise wome awareness of the differences between how applied linguistics and language education used this potential.

Some unhelpful outlooks

A second challenge in making productive use of CDST in language education revolves around a somewhat unfortunate misconception about the theory, which has kept it from becoming mainstream in TESOL. Because of its origins in STEM, some people have tended to think that they need to emulate the research methods from those disciplines when studying social phenomena. This has led to a proliferation of research that tries to document and mathematically analyse patterned behaviour. Research in this type of “restricted”⁴ complexity often looks robust and sometimes it is also useful in understanding how regularities emerge. For the most part, however, it has tended to be irrelevant to practice, sometimes epistemically superficial, often inexplicably unconcerned with agency, and at times disturbingly unethical in its methods.⁵ 

There is, however, a different way to capture the potential of CDST in language education –one that is sometimes derisively called a ‘soft’ option. I occasionally use ‘complexity’ or ‘complexity perspective’ to differentiate this outlook from more mathematically-oriented formulations of CDST.

This perspective entails appropriating the terms and concepts of the theory in order to generate descriptions of language learning that are native to language education. These descriptions need to be holistic, so that they can capture the entirety of the phenomena that are relevant to our needs (e.g., language development, cognition and feelings, curricula and teaching practices), and they also need to be synthetic so that they can capture how these phenomena interact to produce emergent outcomes (e.g., developing linguistic proficiency).  

Our article

This is where the article that Juup and I wrote becomes relevant. Our starting point is that there’s a plethora of factors that shape language learning activity, some inside the learners and some around them. These factors are so many and they interact in such unpredictable ways that it would be futile to even list them, let alone describe their web of connections.

What is important, and feasible, is to describe the large picture they create –the dynamics of TESOL. Different dynamics (configurations) will produce different opportunities for teaching and learning, some more constraining and some more open-ended. What we tried to do, then, by drawing on several years of research, was to make a list of such dynamics and to describe the activities with which they tend to be associated.

Dynamics of TESOL

In the article (and also in our 2021 book, The Intentional Dynamics of TESOL), we propose viewing language education as a complex system that shows evidence of four properties, historicity, adaptiveness, non-linearity, and emergence. Depending on which property is dominant, types of activity (dynamics) emerge, as follows.

Normative activity

Because complex systems are shaped by their history, they will often reproduce past patterns of behaviour and resist change. In TESOL we can see normativity in action when teachers and learners continue to engage in grammar drills and the teaching of meta-language, even though the ineffectiveness of such practices is evident.

Contingent activity

Complex systems are resilient and adaptive (in fact, some theorists call them complex adaptive systems). This means that they tend to preserve their structural properties even when they are perturbed. In the context of language learning, we can see adaptivity when teachers and learners make contingent changes to their behaviour. Some examples of adaptivity in language classrooms include adjusting to the arrival of a new student or the introduction of a new textbook, or dealing with unexpected behaviour in class.  

Creative activity

A third feature of complex systems is that their activity is non-linear. What this means is that they contain within them the potential to generate new structure, just like a seed contains within it the potential to produce a plant, rather than a larger seed or a set of many seeds. Creativity is easy to see in language education, both in teacher activities and in learner output, as long as the normative pressures are not too strong. It is the creativity of complex systems that we leverage whenever we face the dilemma of teaching students how to generate meaning structures through simplified language.

Purposive activity

There is an axiom in complexity theory, that a complex system is what it does. In this sense, if we can nudge a system to do something over enough time, it will eventually settle into a new configuration that is optimised for this activity. Examples of such purposeful activity are few in language education. This, of course, reflects how hard it is to dislodge systems from their preferred states when strong normative pressures are in place.

However, this is a process that we can see in new efforts, and in small-group activity, where relatively small investments of effort can have a profound effect on how a system behaves in the future. It is also a crucial component of any system that includes people, because –whether or not we exercise control over all the choices open to us– humans have agency.

How is CDST relevant to TESOL?

One point that we make in the article is that teachers can benefit from being aware of these four dynamics; and that such situational awareness is useful if one wants to shape one’s praxis. We believe that there are two empowering elements in doing so. The first one is that such a theoretical outlook is more relevant and more realistic for teachers, compared to in-depth knowledge of specific domains in the informing disciplines. The second is that theoretical awareness of the dynamics that shape TESOL shape can help teachers better understand what is possible in their context and this can, in turn, make them more agentive.

Such a complexity-informed description of language education constitutes a shift in the knowledge landscape of TESOL

To extend this further, I would argue that such a complexity-informed description of language education constitutes a shift in the knowledge landscape of TESOL. I suggested, above, that much professional development activity tends to be oriented towards building deeper knowledge about specific niches. This results in a fragmented professional community, where knowledge and practice do not interface with each other; and paradoxically this creates a weaker profession that treats theoretical knowledge with both awe and contempt.⁶ I would like to think that the perspective put forward here is more democratic, in that it is equally accessible to everyone in language education, and more grounded in our professional needs.

Conclusion

In a book written in 2008 (Complex Systems and Applied Linguistics), Diane Larsen-Freeman and Lynne Cameron wrote about a need for conceptual work that will make complex systems theory ‘native’ to applied linguistics. If we view language education as a discipline distinct from applied linguistics –a view that I find increasingly appealing– then, by extension, we need to nativise CDST for our purposes as well. I cannot suggest that what we have written in our article has finally discovered this holy grail, but I do think that such a theory will not be very different in outlook from what we have described.

---

Footnotes

1. I am very grateful to the Hellenic Academic Library Association, who paid to make the article available as an Open Access publication. ↩︎
2. Complex Dynamic(al) Systems Theory, Complex Systems Theory, Complexity? Although people have used all the above terms to describe similar phenomena, with or without differences in nuance, recent usage has tended to prefer ‘Complex Dynamic Systems Theory’ (CDST). I defer to this convention but also use ‘complexity’ to describe a ‘soft’ approach to the use of the theory. ↩︎
3. Juup Stelma, Lynne Cameron, Susan Dawson, Sarah Mercer and I. ↩︎
4. The distinction between ‘generalised’ and ‘restricted’ complexity was made by Edgar Morin (2006). Generalised complexity is a paradigm shift in the way we think about the social world. This is more-or-less what I mean by ‘soft’ CDST or a ‘complexity perspective’. The latter (i.e., restricted complexity) involves using existing tools such as timelines, experiments, simulations etc. to describe complex phenomena without challenging the existing epistemic paradigm. ↩︎
5. As an example of such work, I was once almost bullied into taking part in a study where students at a university course would be asked to download an app that would send them reminders several times a day, at which point they would need to record their mood and rate their willingness to study vocabulary or grammar. I indelicately refused, because such a study would normalise surveillance culture while only producing trivial findings. ↩︎
6. Which is, I am afraid, a much more dangerous mindset than might be immediately obvious. By combining a sense of epistemic powerlessness and contempt for the institutions that use knowledge to challenge authority, this outlook creates the kind of education that paves the way to totalitarianism. ↩︎