Page - (000134) - in Knowledge and Networks

127 that depends partly on characteristics of the first two components. I define these terms metaphorically and loosely. The rotating body represents the learning organi- zation—the population of judges who switch places once a year in a circular system of places, as with a carrousel or “mobility in loops” (White, 1970, p. 380). The rota- tion axis represents a pecking order: a vertical differentiation between the judges and the emergent hierarchy of members with epistemic status. This axis can be pictured as the spinning top’s shaft, which supplies the angular momentum that keeps the spinning top erect. It allows learning to take place in a system that remains stable thanks to its movement. The fragile equilibrium created by the rotation move- ment represents the structural condition for learning collectively in the organization and depends on the stability of the rotation axis and the shape of the organization. Time is taken into account through rotation and speed. These members have the authority to know in the organization. Formal structure is summarized in rotation rules across intraorganizational boundaries and in status differences. Infrastructural stability, that is necessary for collective learning and social processes in gen- eral, comes from movement. The endogenous evolution of advice networks is characterized by three interre- lated moments. First, the centrality of members with high epistemic status varies over time. It initially tends to be reinforced. Pivotal members become ever more central in a Matthew effect (Merton, 1968): The more they attract advice-seekers, the more their reputation grows and, in turn, the more they are sought out. Among other members, there spreads the impression that turning to such a source for advice is safe and legitimating for their own knowledge claims, and that making this choice signals a rise in relative status. Concentration of epistemic authority thereby intensi- fies with the centralization of advice networks as learning comes to depend on a dwindling number of sources of authoritative knowledge. Second, in real-life organizations this centralization creates an overload for members with high epistemic status. These members tend to manage this overload by sharing some of their epistemic status, redirecting advice-seekers to other sources through recommendations. When advice from the handful of the supercentral advi- sors becomes inaccessible, irrelevant, inaccurate, untimely, or rare, members turn to these other advisers, creating new epistemic stars. Sharing epistemic status (a form of delegation) enlarges the number of advisers and lessens the centralization of the network. Third, the expansion in the number of central members with high epistemic sta- tus in the organization creates a problem of epistemic conflicts, consensus, and coordination among epistemic authorities. If it is easy to co-orientate with them, equilibrium is established. If not, conflicts between epistemic authorities trigger recentralization. When collective action eventually becomes endangered by an excessive number of epistemic leaders, some of them withdraw or retire, and others are disqualified in one way or another. As their numbers decrease, it becomes easier at the top to recreate consensus around a common definition of the situation, to give coherent social benchmarks for homogeneous judgments of appropriateness (Lazega, 1992). 7 Organized Mobility and Relational Turnover as Context for Social Mechanisms…