Geography Reference
In-Depth Information
if it is to serve as the basis of a meaningful ontology of regional growth and development.
The stakes here may be high. Path dependence may help explain why regional growth
disparities persist; it may help explain why particular industries and technologies develop
in certain locations but not in others; and it may help us to understand why some regional
economies are better able to adapt over time than others. Our previous paper (Martin
and Sunley, 2006) was intended as an initial exploration of such issues. In that paper, we
explored the sorts of processes that could give rise to path-dependent economic develop-
ment in a geographical - and specii cally, regional - setting, and sought to argue that
many of those processes are themselves place-dependent. The aim here is to build on
that discussion and to focus much more specii cally on how far and in what ways path
dependence can serve as an
evolutionary
concept for studying the economic landscape.
The thrust of our argument is that this task is problematic. For one thing, we i nd that
despite its declared emphasis on the importance of history, path dependence theory as
formulated by its leading architects - David, Arthur, and others - retains elements of
equilibrium thinking, which we contend is in tension with the idea of path dependence as
an evolutionary concept. We are sceptical that the evolution of the economic landscape
ever tends towards any equilibrium state: to be sure, the economic landscape is charac-
terised by self-organisation and order, but these are not the same thing as equilibrium.
We argue, therefore, for a wider view of path dependence that allows for patterns and
trajectories of development that do not approach or reach an equilibrium state, and
that do not require an equilibrium interpretation. This leads on to a discussion of how
to characterise such paths and their emergence in a regional context. Our conclusion is
that a rethinking of the notion of path dependence may be called for if it is to function
as a core concept within an evolutionary perspective on the economic landscape. To this
end, economic geographers could benei t from looking at the wider discussions of path
dependence and historical change that have taken place in recent years across the social,
political and policy sciences, and at related ideas in evolutionary ecology.
2. The basic model of path dependence and its geographical application
One of the abiding intentions behind Paul David's development and proselytisation of
the concept of path dependence has been to persuade economists - and especially neoclas-
sical economists - to move beyond traditional equilibrium modelling and to 'take history
seriously'. This endeavour resonates strongly with evolutionary economists, some of
whom, such as Hall (1994), have elevated the notion of path dependence to the status of
a 'i rst principle' of evolutionary economics. David (2001, 2005, 2007) himself argues that
path dependence is a property of a 'wide array of processes that can properly be described
as evolutionary', including economic processes. But in what sense, precisely, does path
dependence function as an evolutionary concept? Neither David nor Arthur, the primary
exponents of path dependence ideas, has set out a fully specii ed model of economic evo-
lution based on path dependence, nor discussed how the concept relates to other key ideas
in evolutionary economics. Nevertheless, it is possible to infer from their various writings
the sort of model of economic evolution implied by their depiction of path dependence.
According to Sydow et al. (2005), the basic model of path dependence developed by
David and Arthur implies a three-stage model of the historical (and, we might add, geo-
graphical) development of a technology, industry, institution or organisation. In the i rst,
'pre-formation' phase, as they term it, considerable scope and variety exist for exploring
