Larsen-Freeman’s seminal article inApplied Linguisticsin 1997. As she men-
tions, it took a while for this line of thinking to be seen as an important
development. Seeing language as a complex adaptive system and second
language development as a dynamic process has profoundly shifted our
understanding of both. A target language is no longer seen as a static target
representation, to which learners increasingly conform. In its place, language
and its use are mutually constitutive; they determine each other. It is in
situated, co-adaptive, discursive encounters that learners obtain relevant and
accessible exemplars from which they can learn. Language use, change and
learning take place on multiple timescales and levels of complexity. They are
temporally and contextually dependent processes.
This view has substantial implications for methodology. One is a new
appreciation for variation in the data, which is seen as providing information
rather than clouding the“real data”. Because individuals differ in their
developmental paths, longitudinal case studies are encouraged; they provide
adifferent picture of development than group studies and cross-sectional
studies, which tend to disregard variation by focusing more on central ten-
dencies. Rather than isolating a single variable, researchers look for patterns
among interacting factors, whereby some factors have more influence at
certain times than others. The interconnectedness of embedded subsystems
in a language means that changes in one system may lead to changes in other
systems.
Because of different initial conditions and the uniqueness of each learner,
the same, shared input may lead to different, individual outcomes. Because
of this non-linearity, development trajectories may not be predictable,
although they should be explicable retrospectively. Simple causal statements
give way to contingent statements, and rules give way to patterns. CDST is
not a theory of language. Theories of language that are theoretically con-
sistent with it are Cognitive Linguistics and Systemic Functional Linguistics.
Both underscore the dynamic nature of language and understand language as
a socio-cognitive construct, deriving from embodied experience. Also
aligned with CDST is an emergentist or usage based approach to language
development. Karlfried Knapp sees as one of the significant trends the move
from seeing language as“a more or less stable formal system to viewing it as
a dynamic, adaptive process-oriented system whose acquisition is usage-
based”. This view is endorsed by many informants, including Zoltán Dör-
nyei, Jean-Marc Dewaele, Aneta Pavlenko and John Schumann. Diane
Larsen-Freeman also expresses her worries about this new perspective:“It is
difficult to convey the power of this new approach to conceiving familiar
phenomena. Ironically, we are limited by our language itself in reflecting its
dynamaticity.”Rod Ellis is yet to be convinced of the relevance and impact
of CDST. In his view, the interconnectedness between components should
include a link between the social, the cognitive and the linguistic aspects of
language use. So far, no study has shown that convincingly. The question is
what the definition of a theory is. Is CDST a theory, as some of its proponents
88 Trends III