The ontology of power

There have been quite a few posts on the concept of power over the years in this blog (link). This continues to be an intriguing subject for me. Fundamentally the question of the moment is this: how does "power" fit into a social ontology of the kinds of things that exist in the social world? Does "power" exist, or is it a term that encompasses a range of other social things and relations? At one point I characterized power as an "influence" concept (link), analogous to "is humorous," "is rude," or "is charismatic". In each case the concept is defined in terms of a capacity possessed by the agent to influence certain kinds of behaviors by others around him or her.  Along these lines we might define power as a feature of an individual as follows:

Caesar is powerful =df Caesar has access to social levers of coercion that permit him to coerce certain kinds of behavior by others.

Those levers might include --

• command of military or paramilitary forces
• relations with conformant legislators positioned to enact legislation and regulation
• relations with conformant private citizens positioned to coerce their dependents to behave this way or that

On this approach, "power" is a dispositional concept attaching to individuals and invoking access to certain kinds of instruments of coercion.

What is "coercion"? This is a topic that Robert Nozick picked up early in his career (“Coercion” in Philosophy, Science, and Method: Essays in Honor of Ernest Nagel). We might define coercion very simply in terms of an actor's ability to influence the terms of choice that confront another actor. The Godfather put it best: "I'll make you an offer you can't refuse." Here is a possible definition:

x coerces y =df x creates a choice situation for y in which the costs of choosing other available options are unacceptably high, leaving only O_n+1 to be chosen.

(Here is an article on coercion in SEP; link.)

We can then broaden this concept to groups in society by postulating that certain groups have greater access to the levers of coercion than others. So we might say that "capitalists have more power than workers" because capitalists have access to the lever of unemployment, whereas workers have access only to the power to withhold their labor at substantial personal and familial cost. And we might say that organized crime lords have more power than shopkeepers because the criminals have greater ability to use and threaten violence against the shopkeeper than the reciprocal.

We can also broaden the concept by grouping the levers of coercion into different families: economic coercion, physical coercion, community coercion (shunning within a religious community, for example), … This allows us to identify the sources of power within a given society. Individuals and groups who are favorably positioned with respect to these different categories of instruments of coercion are more powerful than others.

What this account leaves out is the range of "soft" power associated with framing and ideology. This is a different avenue of social influence. It is possible to impel individuals or groups towards certain kinds of actions by shaping their beliefs and cognitive-affective frameworks. Using the media to shift the terms of debate over a policy -- taxing the rich at higher rates or providing universal health insurance -- is a way of exerting power without coercion.

A few features of "power" emerge from these observations. First, power is a relational concept. An individual possesses power over other individuals (relation 1) and does so in virtue of the relations he bears to other persons and institutions (relation 2). Second, power is the activity side of structure: structures and sets of social relations constitute the environment within which individuals are empowered to exercise power over others. The structures perdure, and individuals act within their elements to exert their will over others. And third, power is distributed over many individuals and groups within society at a time. But it is fluid and subject to fluctuations as structures and individuals change. Dick Cheney was powerful at one point in time but not at other points. And what changed was not Cheney but the levers of power and social networks to which he had access.

This seems to imply that power is not itself a "thing" within a reasonable social ontology. It is rather a relational characteristic that exists at various social locations depending on the connections those locations have to the levers of influence over individuals and groups.

(We might speculate that there is an art to exercising power that means that not everyone is equally adept at the activity. Being well situated with respect to the levers of influence is a necessary condition but not a sufficient condition for being powerful.)

This account lines up well with some aspects of Steven Lukes' analysis of power in his classic book, Power: A Radical View (link). It also aligns well with Michael Mann's definition of social power in The Sources of Social Power: Volume 1, A History of Power from the Beginning to AD 1760 and elsewhere: "My earlier work identified four primary 'sources of social power' in human societies: ideological, economic, military, and political" (Fascists, 5).