Geography Reference
In-Depth Information
space for contemplating the entire range of potentially relevant variations. They might
study, for example, the boundedness of ideas or populations of ideas in terms of the
processes through which distinctions are maintained (Lamont and Molnár, 2002). To
the extent that these boundaries are contested, the process of boundary construction is
subject to evolution. For cluster researchers, the issue then would be to examine how
actors tie boundaries together to create new distinctions around idea complexes underly-
ing, for example, shared identity, forms of governance, or collective strategy.
Associated with the view that ideas evolve in an ecological space in which ideas
compete for human attention (Heath et al., 2001) is the argument that populations of
ideas exist as polythetic combinations of ideas that are sui ciently related to be recogniz-
able as distinct ideational systems, such as those constituting a cluster's identity. This
raises the question of how to determine which of the many ideas l owing throughout
a cluster converge to a
dominant
cluster identity. The argument that clusters are dis-
tinguishable groupings of dominant ideas based on, for example, a core technology or
a local tradition, is problematic if there is a risk of mistaking dominance for adaptive
value. From an evolutionary perspective, what turns out to be dominant or valuable is
dii cult to predict, as individuals have a range of abilities they can use to accomplish a
variety of goals in a system that, as a whole, may not have a single principle for variation
and selective retention. Dif erent clusters may be organized around dif erent 'focal rules'
(Schelling, 1960), used for dif erent purposes, such as coordination, learning, or identity
maintenance. Some of these rules may support dominant ideas that turn out to have no
adaptive value.
Given that, at any point in time, clusters normally contain a mix of adaptive and
maladaptive variations, the evolutionary question concerns the origins of new variations
that might re-balance existing coni gurations. Network-theoretic arguments are typi-
cally invoked to theorize about the relational conditions that either enable or constrain
the generation of new variations. Networks carry information and, depending on how
they af ect the perception, interpretation, and enactment of ideas, they help transform
information into knowledge, while reconstituting the individual actor's mindset and the
cluster's collective mindset in the process. It has become something of an axiom to argue
that tightly woven social networks unduly constrain the remixing of routines (Staber,
2001), but the precise mechanisms involved in this process are not well understood. It
is not clear, for example, whether change is enhanced by a thick network of strong ties
connecting subsets of a cluster (bridging ties) or by a dense network of primary con-
tacts (cohesive ties). The evolutionary outcomes of network structure are a function of
the interaction between environmental conditions and the responses of cluster actors
(Glückler, 2007) who are motivated by evolving ideas about strategy, identity, technol-
ogy, cooperation, and so forth (Hite, 2005).
Under conditions of uncertainty, the new variations that emerge from this interac-
tion are essentially 'blind' with respect to the i tness of the system (Campbell, 1965), an
idea that is at odds with the (conventional) view of clusters as planned arrangements.
Evolutionary theorists question not so much the existence of rational choice and planned
interventions than the (often implicit) assumption that the consequences of actions are
closely tied to intentions. The 'wisdom' of an idea - that is, whether it is adaptive or
maladaptive - can be interpreted only retrospectively, once environmental conditions
are known. Bradshaw's paradox, which states that we 'need to know the biological
