Deleuze and complexity science in second language acquisition

While preparing a short presentation on the usefulness of some of Deleuze’s concepts for pedagogy/didactics , I re-read an assignment I wrote for my didactics class and found it a lot clearer than I remembered it. Therefore, hoping that perhaps someone else (particularly teachers of language or anyone interested in Deleuze, chaos theory or complexity science) might be interested, I will publish the main part of the theoretical elaborations below. Comments are very welcome! (in English or Swedish)

I have, like many others, always been highly skeptical of grammar theory, and the usefulness and correct form of grammar teaching in the classroom have been less than obvious to me. The first theoretical elaboration one might begin from this position could of course be the anti-grammar movement based on Krashen’s ideas. The main assumption made by Krashen is that grammar can be acquired “naturally” from free interaction and meaningful input. The language classroom should then, from this perspective, not consciously focus on language or grammar form, but rather on the practical and self-motivated use of the language, the communicative aspect. Krashen has, among other things, focused much on free reading as a way of acquiring language patterns.

I do not, however, subscribe to this anti-grammar view of second language acquisition. At the same time, I do not feel content with the form of grammar exercises that have been promoted in many of our seminars at Linköping university. These so-called problem-based exercises are supposed to be a less theoretical way to teach grammar than what we used to do in the old days, before Krashen. Though, as much as these exercises are said to be communicative, I can not help but feeling that they still impose a boring and rote task on the students, which are likely to do the exercises in an often mechanical fashion rather than gaining any deeper or lasting understanding. To me, the obvious hierarchy of teacher-imposed theoretical learning of grammatical rules has in this case just been exchanged for a micro-fascism where the students’ immediate communicative desires are still subjugated by the need to solve abstract and irrelevant teacher-imposed tasks.

So, how can Deleuze and chaos/complexity science help us with our predicament?

Consider one of the most common ways to view knowledge: as a tree. We often imagine the mind of a child as being something close to a seed, small, empty, in need of the parent’s nurture and watering. Slowly, the child’s competence in various areas grow, upwards, linear, in a quite predictable way forming a strong body or stem, then the heavier branches (categories of knowledge such as “social studies” or “English”) then dividing into smaller branches (such as “the American electoral system” or “food-related vocabulary”). This view could perhaps be said to form the foundation of the compartmentalization that we see in our schools, with few, if any, connections between the different “branches” (i.e. subjects). (see Pedagogiska magasinet 2/2010 for more on this topic.) In the contemporary debate, knowledge is seen as quantitative, as building blocks, or perhaps as some kind of liquid being used to “water the trees” (or, even worse, to “fill the buckets”) i.e. teach the students.

An alternative to this view is the image of the rhizomatic network, an illustrative concept shared by post-structuralist philosophers Gilles Deleuze and Felix Guattari as well as neuroscientists trying to understand the mechanisms of the brain. Seen this way, the brain is a dynamic and vivid network that is constantly changing and learning by making new connections between the nodes in the network. Learning does not have to start from the bottom of the tree trunk, forming a uniform base for specialized knowledge that can, perhaps, be allowed in the uttermost branches, many years into his/her education, after the student has gained insight in the structured systems of knowledge forming the lower branches; rather, knowledge is seen as a network where connection can, should and have to be made in all kinds of directions, between high and low, between different subjects, between the simple and the complex.

This view could be connected (aha!) to the sociocultural theories of Vygotskij, seeing interaction between students as central to learning, allowing for unexpected connections forming new kinds of knowledge, rather than focusing on a single teacher giving exactly the same type and amount of water and fertilization to everyone. Moreover, this can be connected to one of the basic assumptions of connectionism: that complexity (and subsequently what we perceive as intelligent behavior) emerges from the interaction between simple units cooperating and reacting to their environment. As Nick C. Ellis puts it: “Closer scrutiny [in this case: of the extremely unpredictable patterns of movement of an ant] reveals that the control decisions are both simple and few in number. An environment-driven problem solver often produces behavior that is complex only because a complex environment drives it.” (Ellis 1998:643)

Diane Larsen-Freeman examines the usefulness of the closely related chaos and complexity sciences. One of the most interesting concepts, and one that is central to these sciences, is nonlinearity. That a system is nonlinear simply means that it does not develop in a predictable and linear way, where x amount of force put into the system necessarily corresponds to the same amount of change in the system. One concrete example for teachers: going through a set of grammatical rules on the white board does not necessarily mean that all the students will remember everything you said for all eternity… As an alternative to this absurd belief (not held by many, although many tragically believe that students should remember everything, and that failure to do so is a question of faulty morale), Larsen-Freeman posits the nonlinear system:

A nonlinear system is one in which the effect is disproportionate to the cause. Conversely, in a linear system a cause of a particular strength results in an effect of equal strength. When a spacecraft is nudged into orbit by firing its thruster rockets, a linear system is responsible. Nonlinear systems can also sometimes exhibit linearity, however, at other times, they may react in a way that is all out of proportion to the cause. A rolling pebble, for example, can trigger an avalanche. (Larsen-Freeman 1997:143)

From this perspective, we can combine the perspective of automatic acquisition advocated by Krashen with the occasional theoretical grammar lesson and, more importantly, with grammar-oriented feedback on the texts and verbal production of individual students. The argument goes like this: if it is basically correct, as Krashen assumes, that students do acquire language structures in a free and self-motivated fashion, then this should be the main driving factor behind acquisition. However, you could (rightly) argue that students will make unnecessary mistakes and take longer to understand basic patterns without any guidance at all. This is where the theories from chaos/complexity science enter. If the knowledge of language patterns is seen as a nonlinear rhizomatic network, and if students will seek understanding in a free manner, then they will arrive at a level of understanding that is “almost there”, i.e. they might notice certain patterns and rules without really being able to formulate the general rule and thereby letting their knowledge “fall into place”. This is where the teacher steps in, giving input that is similar to the final pebble that starts the avalanche. One example of this could be a presentation on the genitive – a pattern that most will have noticed but few Swedes fully understand even after three years in upper-secondary school, and if this problem is believed to be very common in the group and if it is believed that very many of the students are at a similar level regarding his particular problem, this could be made with the whole class gathered together. At other times, with more unevenly distributed mistakes or incomplete understanding (such as different irregular nouns) it is probably a better idea to tutor the individual student and give her/him feedback when or soon after mistakes are made.

• Ellis, Nick C. (1998): Emergentism, Connectionism and Language Learning. Language learning 48:4
• Larsen-Freeman, Diane (1997): Chaos/complexity science and Second Language Acquisition. Applied Linguistics vol 18 no 2