Networks, Power In

A network is the sum of connections between individuals or organizations in a given locale or sector of activity. It is more than a one-off or one-shot interaction necessary to do a single piece of business, but is something that iterates and continues in place for a long time. Sometimes the exercise of power needs no network, like the confidence man who approaches someone on the basis of no prior knowledge; this lack of a prior network helps the con take place. More usually, individuals have relationships with people they know over a reasonably long period of time, [Page 445]and these relationships are structured by that history, which is also influenced by the bonds of common knowledge or friendship. There are various kinds of network structure, which may, for example, be close, sparse, or centralized, with various kinds of pathways between the different actors. A network may contain or embed social values that influence how the social actors relate together and make decisions.

The relationship between power and networks is subtle. Networks are a form of social cooperation that helps ensure that collective action takes place. They provide the knowledge and trust to ensure that benefits are shared out fairly and help solve problems based on mutual knowledge and exchange. Network members can signal to each other to act, thereby confronting the collective action problem. The closeness of the bonds among members can help build up trust. Also the expectation that an actor will meet the other actor again (rather than act imperiously, as in a one-shot game) prevents conflict from becoming destructive or people taking their gains and walking away.

Networks also provide resources that individuals need in order to act. Mark Granovetter claims that networks can allow individuals to access information and even if the links are not strong, where they bridge across different networks individuals gain collective power. Informal links are often more important that organizational ones (the “strength of weak ties”). Some networks will have greater resources than others; for example, extensive middle-class networks of acquaintances may be more effective in accessing resources such as employment than, for instance, a dense working-class network made up of family members and neighbors who do not know people outside their close circle. As a result, bridging networks are associated with better policy outcomes because of their better access to information. Of course, these networks could contain untrusting members who just have to work together, but this is probably uncommon, as networks need a lot of trust to work effectively or else they break apart.

Actors do exercise power in networks to get what they want—they use the network to dominate others, usually by taking advantage of the fact that the structure of the network determines the information flow. Formal network analysis, which concerns the measurement of networks, helps understand how this works. Networks are composed of nodes and ties: nodes are the actors, while ties are the links between them that may or may not be present or present to a different degree. Nodes and ties may be arranged in different patterns and have mathematical properties as determined by graph theory. Of importance for the study of power is that it is possible to denote a score for each actor in the network created by summing the ties to others. The most common score is network centrality, which is useful from a power perspective because the more central an actor's location, the more influential he or she is likely to be. Because information flows through that central actor, the network the network can be structured to his or to her own advantage. For writers on government, this notion of centrality is often called nodality, and in the view of Christopher Hood and Helen Margetts it is one of the key tools governments use to get things done. There are different measures of centrality. The most common is degree centrality, which counts up the links one actor has to the rest. This might be thought to be too simple, for what matters is the extent to which an actor dominates others who have weak ties rather than simply the number of ties the actor has. As Phillip Bonacich writes, “It is advantageous to be connected to those who have few options. Power comes from being connected to those who are powerless” (1987, p. 1171). Thus, for Bonacich centrality is defined as the sum of an actor's connections to other points, weighted by the centrality of each of these other points, though in practice it is highly correlated with simple or what is called Freeman centrality. There is a lot of research that links centrality to power in networks, such as in banking and in the study of community power. There are other measures of network structure, but they are less used in studies of power.

...