Geography Reference
In-Depth Information
17 Growth, development and structural change of
innovator networks: the case of Jena
Uwe Cantner and Holger Graf
1. Introduction
The notion of collective invention has been introduced in the literature by Allen (1983)
who provides evidence from the nineteenth-century iron and steel industry, where
innovative success was the result of the cooperative activities of several, dif erent
actors. Anecdotal evidence such as Allen's or the various studies on Silicon Valley (e.g.
Saxenian, 1994) or other success stories of regional innovation (e.g. Braczyk et al., 1998;
Cooke and Morgan, 1994, Keeble et al., 1999) was enriched by studies on the regional
dimension of knowledge l ows (e.g. Jaf e et al., 1993). Insights into the process of innova-
tion at the i rm level (e.g. Kline and Rosenberg, 1986) or the national level (e.g. Lundvall,
1992; Nelson, 1993) strengthened the view that the functioning of innovation systems is
the basis for innovation-based economic success.
Inspired by the seminal volumes by Lundvall (1992), Nelson (1993), and Braczyk et
al. (1998), a large number of investigations build on the systems view in relating innova-
tive activities and success. 1 As the systemic view of innovation is inherently dynamic and
deeply grounded in evolutionary theorizing, empirical studies that take the systems view
seriously should have two ingredients: heterogeneous, interacting actors and the dynam-
ics of these interactions. Unfortunately, many of the empirical studies usually fail to
account for both of these ingredients, because of the unavailability of appropriate data.
In studies based on aggregate data it is easier to observe the dynamics of the system as
many variables are available as time series. Many studies that account for the interactive
structure and the heterogeneity of innovative actors are based on interviews or surveys
that are only available at one point in time. This is not to say that system dynamics are
not considered important by these authors, but most of the arguments are based on theo-
retical reasoning rather than on empirical observations.
As a response to this unsatisfying situation, a number of researchers in economics and
economic geography started to employ social network analysis to study the interactive
structures in the innovation process. The network approach of ers a methodology that
accounts for heterogeneity and system dynamics. However, its empirical application is
subject to rather strong data requirements. Data need to be more or less complete in
terms of actors and relations and for dynamic studies they have to be available over
longer time periods. Information derived from patents or publications has these proper-
ties and in the application of social network analysis it provides a graphical representa-
tion of the actors and their relations in the system under investigation. For example,
networks of co-authorship are used to analyse the development of scientii c communities
(e.g. Barabasi et al., 2002, Moody, 2004), networks of co-invention can help us under-
stand the evolution of local clusters (e.g. Fleming et al., 2006), and citation networks
provide information about the l ow of knowledge (e.g. Sorenson et al., 2006). The struc-
370
 
Search WWH ::




Custom Search