Page - (000139) - in Knowledge and Networks

132 energy stems from the way power and class struggles function in the organizational society: in collective efforts to hoard opportunities and to saddle others with con- straints while still encouraging or obliging everyone to compete for these same opportunities and resources. The clearest way to comprehend the origins of the energy for rotation in OMRT is to understand that it is used to concentrate power in a stratified organizational society, a society made of superimposed levels of agency. Such a society spends a great deal of energy catching up in status-related competi- tion imposed from above, self-imposed from below, or both. That struggle is not so much about catching up with the Joneses next door as it is about adjusting to top- down constraints on maintaining or enhancing one’s status. The promise of sharing power and status takes the power differentials generated by the structure of organi- zational society and turns them into a source of energy. Of course, decentralization is followed by recentralization. But each step in this catch-up cycle is what pro- duces the energy for OMRT. These OMRT are intrinsically multilevel, and the only way to understand them is to develop models depicting the dynamics of multilevel networks. These develop- ments will be at the heart of future explorations in the social sciences. The mesoso- cial order and the multilevel dimension of social phenomena show that systems of superimposed interdependencies (one interorganizational, the other interindividual) create dynamics specific to each level. But because levels are partly interlocked, dynamics across levels drive each other. Drawing on Simmel’s (1908/2009) ideas about social circles and Breiger’s (1974) “dual” approach to the coconstitution of individuals and groups in society, sociologists have begun to look at the dynamics of multilevel structure and their consequences for societies (Lazega & Snijders, 2016). Articulation of distinct levels of action for that purpose can be partly accounted for, beyond bipartite structures, with a method called structural linked design, which brings together networks of different levels by using individuals’ affiliation ties, be they single or multiple (see Fig. 7.1). Statistical analyses of linked-design data (Wang, Robins, Pattison, & Lazega, 2013; 2015) show that two levels are not just superimposed but highly intertwined without being necessarily rigidly nested. That relationship implies that changes in ties at one level contribute to changes in ties at the other level even if the capacity to force changes at the other level varies with socioeconomic attributes of the actors. At each level actors attempt to structure the contexts of their interactions and have to manage the attendant contextually imposed constraints by trying to redesign their opportunity structures. In this approach each complete network is examined separately and then combined with that of the other level by means of information about each individual’s membership in the first network (interindividual) and in one of the organizations of the second network (interorganizational). Work undertaken so far within this framework has shown that dual or multiple positioning in super- imposed systems of interdependencies makes it possible to formulate and test pre- cise hypotheses about the relation between members’ position in the structure and individual achievements, especially when this positional reckoning is based on E. Lazega