Geography Reference
In-Depth Information
cluster is said to be 'institutionally rich' if it includes numerous institutions of ering a
variety of resources, high levels of interaction among these institutions, collective struc-
tures that emerge as a result of this interaction, and a mutual awareness of the necessity
of developing common objectives and unii ed action (Sydow and Staber, 2002). The evo-
lutionary interest in institutions is not so much directed at understanding institutions as
formal organizations, but is motivated by a concern for the meanings and social disputes
that create, transform, and reproduce specii c institutional arrangements.
Complexity theory has also provided a number of useful insights for the evolutionary
analysis of clusters (Martin and Sunley, 2007). It views clusters as open and complex
adaptive systems, driven by non-linear, self-organizing, and co-evolutionary dynamics,
with emergent properties based on spontaneous interaction among system components
(McKelvey, 1997). One of the arguments taken from this theory is that clusters are
most robust, innovative, and adaptive if they exist at the 'edge of chaos', a position that
balances order and chaos, control and change, and cooperation and competition. The
evolving balance between robustness and fragility, between staying the same and chang-
ing into something else, has yet to be explored more fully in research on clusters. The
most successful clusters are not necessarily, as is often argued in this literature, those
that are most innovative, in the sense that they generate a continuous stream of new
ideas. Constant change and continuous mutations may make it impossible for the actors
to apply new discoveries, to the extent that the exploitation of discoveries requires some
degree of stability to retain useful knowledge, create common understanding, and estab-
lish accountability. Successful clusters may be l exible, but they also contain stabilization
dynamics without which the benei ts obtained from selected variations would dissipate
and change would be random.
The principle of self-organization states that the cluster 'tunes itself' spontaneously
towards an order, without an overall plan and central direction. However, what exactly
this means in particular instances is left unclear especially in analyses that remain stuck
at a normative level, with arguments about complex systems being 'ef ective' if they are
'somewhere in the middle' between too much and too little complexity, or if they provide
an 'optimal mix' of homogeneity and heterogeneity. It also skirts the so-called structure-
agency problem. What kind of
spontaneity
is possible in a social system that would
normally require individual actors to respond to and modify the social pressures of the
environment of which they are part? How is a spontaneous
order
possible if individuals
are to transform their own relationships in contexts that are ontologically distinct from
agency?
The above perspectives have added to our understanding of a range of issues concern-
ing cluster evolution. For example, they have highlighted the potential adaptive utility
of structural and behavioral diversity, the role of extra-cluster linkages, the dynamics
of organizational foundings and failures, and the role of institutions as rule and con-
vention setters, supporting selection and retention. By emphasizing the variability of
cluster performance across time and space, they have contributed to the understanding
of issues that are at the core of economic geography. And by showing - albeit often
indirectly - that these issues involve social and cultural frames, they have pushed a view
of cluster evolution as an essentially human enterprise, thus falling squarely within the
purview of human geography. On the whole, however, researchers who have drawn on
these perspectives have not been keenly interested in the
specii cs
of human agency. By
