other language may be enough to make the system switch and in this sense
lead to a change of state of the system. In other words, the grain of sand is
analogous to the trigger and the switch to an avalanche. The specificcon-
figuration of interacting variables will be very idiosyncratic: all the individual’s
experiences are reflected in her present state, and what may be a trigger for
one individual may have no effect at all for another individual.
In physics there is a world of research on phase transformations, such as
the change of ice into water or gas into afluid state. A change of the lan-
guage system that leads to a switch can be seen as a state phase transition
from Lato Lb. Triggering can have the effect that at all levels or just on a
minor local level the system changes phase. Following ideas on SOC, the
impact from a specific trigger can not be predicted, but the effects are
assumed to follow a power law function which states that over longer peri-
ods of time and many iterations, there will be many small changes, and a
small number of larger ones. The essence of this line of thinking is that the
larger ones are not simply outliers, but part of a normal pattern and there-
fore do not need a special explanation. Or, to use the common terminology
for this, they follow a power law.
Criticality plays an important role in CDST models in various disciplines.
So far it has not been applied to language. It presents a useful model for
specific types of CS. The model as presented leads to smooth switches that
are internally generated. So, following this model, it does not imply the com-
plete activation of one language and the complete deactivation of another one.
Languages just differ slightly in activation, which makes switching light.
In this section I have tried to show that some of the core concepts of
CDST, the criticality of systems, also applies to aspects of language use,
exemplified here with respect to CS. It could be argued that the notion of
“restructuring”(McLaughlin 1990) is an example of SOC. In her discussion
of this concept, Lightbown (1985: 17) already linked ideas of CDST and criticality
with restructuring in language learning:
Restructuring occurs, because language is a complex hierarchical system
whose components interact in non-linear ways. Seen in these terms, an
increase in error rate in one area may reflect an increase in complexity
or accuracy in another, followed by overgeneralization of a newly acquired
structure, or simply by a sort of overload of complexity which forces a
restructuring, or at least a simplification in another part of the system.8.5 The analysis of language variation
Inter-individual and intra-individual variation are typical in the process of
language development. Variability is seen as a necessary condition for
development to take place.“It is a basic tenet of CDST that for change to
occur, stable patterns must become unstable in the endogenous environment
Trends III 97