2
Sociology is a highly fractured discipline, containing simultaneously a number of theories
and perspectives that are contradictory or even incommensurable. In recent years, however, one
theoretical viewpoint—network theory—has attained a high degree of acceptance discipline-
wide, from social movements to comparative-historical to social psychology to cultural
sociology. If 21
st
century American sociology shares any substantively meaningful common
ground it is that “networks matter.” Of course, different scholars and different subfields vary in
the importance they accord to social networks. But in practice this variation in importance occurs
on a scale from “total” to “very.” Either the substantive content or the geometric structure of
networks are held to influence, if not cause, everything from religious conversion (Lofland and
Stark 1965), to finding a job (Granovetter 1995), to personal identity (McFarland and Pals 2005),
to mercantile dominance of Florence in the early 15
th
century (Padgett and Ansell 1993).
Networks are everywhere.
As other scholars have pointed out, network theory has many obvious attractions (see
Emirbayer and Goodwin 1994). It provides a largely observable mechanism (interactions
between real people), the proffered mechanism is truly social (not grounded in personality or
idiosyncratic preferences), and the formal nature of network explanations makes them
generalizable across substantive areas. Concepts like centrality, closure, brokerage, and density
are useable in widely diverse substantive inquiries. Though network theory is not without
problems or complications, these factors, among others, help explain the appeal of network
theory across the discipline.
The goal of this paper is certainly not to dismiss the causal importance of networks.
There is ample reason to believe that both the shape and content of social relations have a
profound effect on their members. Instead, the goal of this paper is to problematize the formation
Convention