Page - (000200) - in Knowledge and Networks

194 returns” (Arthur, 1994), although they have also become more receptive to explana- tions other than decreasing unit costs or network effects (David, 2001). Third, this path-dependent process is likely to lead to a lock-in, that is, a situation in which actors are stuck with former choices because an alternative course of action is no longer feasible. Building on the works of David and Arthur as well as on that in political science (e.g., Mahoney, 2000; Pierson, 2000; Thelen, 1999), the ascendant theory of orga- nizational path dependence (Sydow, Schreyögg, & Koch, 2009) highlights the importance of imprinting processes, the notion that history already matters in the preformation phase of a path. In keeping with earlier contributions and with much thinking in economic geography (see Henning et al., 2013), this theory also under- lines the importance of self-reinforcing processes triggered by one or more events or actions that narrow down alternatives in the formation phase, including coordina- tion, complementarity, and particular kinds of learning effects. Finally, like the original conception by David and Arthur, the theory of organizational path depen- dence stresses the lack of any realistic alternatives in the lock-in phase. Nevertheless, it asserts the importance of considering agency also in this last phase because agents do not simply have to take the path; they are capable of making on-path changes and, under specific circumstances, of shaping, leaving, or even breaking the existing path. Although this theory is fairly new, it has already been applied explicitly to the analysis of a nascent cluster (optics), for which the relevance of technological, insti- tutional, and organizational factors for regional path dependence was pointed out by Sydow et al. (2010). More significant, however, the theory of organizational path dependence is consistent with recent developments in the theory of regional path dependence (in the fields of economic geography and regional studies, see espe- cially Henning et al., 2013; Martin & Sunley, 2006). Table 10.1 gives an overview of the most important features of this theorizing in four important streams of literature. Path-breaking activities, which do not necessarily imply the creation of a new path, can be different in nature and degree, but there is a minimum requirement: Since the process of becoming path dependent has been framed as progressively eliminat- ing the scope of decision making, this minimum condition is the effective restoration of a choice situation—the insertion of at least one alternative course of action. However, open- ing the window for an alternative is necessary but not sufficient. The new alternative has to be a superior one (Arthur, 1994), because implanting an inferior one would not constitute a real choice. (Sydow et al., 2009, p. 702) Of course, what is perceived as a real choice by the actors or outside observers depends heavily on context. In reality, it is less a question of choice or no choice than of degree. A more realistic understanding of path-breaking would therefore be geared to increasing the scope of choice. A potential means of path-breaking in this sense is platforming, which tries to (re)introduce diversity, related variety in particular, into a developmental process, including a regional knowledge path. In the case of forming a new path, a certain amount and composition of such related variety is required. However, both are hard J. Sydow and F. Koll