Basics of Studying Power
by G. William Domhoff
What is Power?
"Power" is about being able to realize wishes, to produce the effects you want to produce. It is one of the basic dimensions of all human experience, whether at the interpersonal, group, or societal level. It is only one of several universal dimensions in the human experience, along with love, cooperation, esthetics, and intellectual curiosity, but it's the one that plays the lead role in all the problems discussed on this web site. It's the dimension that leads to bullies, rival gangs, enforced cooperation, hierarchy, ruling elites, ruling classes, and wars among rival nation-states.
In order to study power systematically, we need a basic definition. A good starting point is provided by one of the great philosophers of the 20th century, Bertrand Russell. It is very simple, but very profound: "power is the ability to produce intended effects" (Russell, 1938). The usefulness of this definition is shown by the fact that a sociologist who surveyed every major discussion of power before and after Russell wrote still ended up with Russell's definition, slightly altered: "Power is the capacity of some persons to produce intended and foreseen effects on others" (Wrong, 1995, p. 2). Wrong added the words "intended and foreseen" to rule out accidental success as a manifestation of power.
Russell's definition allows for the two kinds of power noted by social scientists, collective power and distributive power. Collective power concerns the capacity of a group to realize its common goals; it is the combination of organization, cooperation, morale, and technology that allows one group or nation to grow and prosper while another one falters. It's an essential form of power, but it is often taken for granted, and not talked about very much. It is what makes possible the existence of distributive power: if the group didn't have the collective power to grow and produce, there wouldn't be anything worth fighting over.
Distributive power, then, is about who has power over who and what. It is captured in a definition of power highly favored by sociologists, first provided by one of the founding figures of sociology, Max Weber, who wrote that "we understand by 'power' the chance of a man or a number of men to realize their own will in a social action even against the resistance of others who are participating in the action" (Weber, 1998; Wrong, 1995, p. 21). In a similar vein, an American political scientist who has written extensively on power and democracy, Robert A. Dahl, concludes his discussion of the concept by noting that definitions usually rest on the "intuitive idea" that "A has power over B to the extent that he can get B to do something that B would not otherwise do" (Dahl, 1957, p. 202).
From the perspective of distributive power, there is a domination-egalitarianism dimension that is potentially present in all human interactions and social structures, with dominators at one end and egalitarians at the other. Dominators think that there are inherent inequalities among individuals and groups that justifiably lead to extreme hierarchy and social stratification. They emphasize competition and say it's a dog-eat-dog world. Down through the centuries they have believed in male superiority, racial superiority, and the need for kingships and nobilities to control ordinary people. In the nineteenth century, they wanted to restrict the vote to property owners. More recently, they resisted feminism, racial integration, and the extension of civil liberties and democratic rights to everyone.Egalitarians, on the other hand, are those who want to minimize arbitrary authority and
hierarchy, and who believe that everyone should be accorded the same opportunity and respect in the political, economic, legal, and personal realms of life. They emphasize cooperation as the key to the good society. They deny that some individuals or groups are better than others, although they recognize that some individuals, but not groups, have unusual gifts for activities like art, athletics, music, or scientific research.
Beyond the distinction between collective and distributive power, Russell's definition of power has another advantage. It does not try to reduce the various types of power to any basic type that is said to have "ultimate primacy." Rather than seeing military force, economic ownership, political control, or religious institutions as the primary source from which all other forms of power develop, as the rival "grand theories" of 19th century sociology and politics tended to do, this definition allows for the possibility that the different types of power may combine in different ways in different times and places. As Russell put it in his wide-ranging historical and cross-national study of power:
This is the ideal starting point for power structure studies because it fits with the four-network theory of power proposed by sociologist Michael Mann (1986; 1993) after an analysis of power structures throughout Western history.
However, to say that power is the ability to produce intended and foreseen effects, and that no one form of power is more basic than any other, does not mean it is a simple matter to study the power of a group or social class. It is necessary to develop what are called indicators of power.
For research purposes, power is best thought of as an underlying "trait" or "property" of a collectivity. It is measured by a series of indicators, or signs, that bear a probabilistic relationship to it. This means the indicators do not necessarily appear each and every time the trait is manifesting itself. This way of thinking about such sociological concepts as "power" and "morale" was developed by social psychologist Paul Lazarsfeld (1966).
In a detailed discussion of the relationship between concepts and measurement in the social sciences, Lazarsfield showed that the traits of a social collectivity are similar in their logical structure to the personality traits utilized by psychologists to understand individual behavior and the disposition concepts developed by philosophers of science in order to explain the nature of scientific theory. Whether a theorist is concerned with "magnetism," as in physics, or "friendliness," as in psychology, or "power," as in sociology, the underlying structure of the investigative procedure is the same. In each case there is an underlying concept whose presence can be inferred only through a series of diagnostic signs or indicators that vary in their strength under differing conditions. Research proceeds, in this view, through a series of "if-then" statements. "If" a group is powerful, "then" it should be expected that certain indicators of this power will be present.
Unfortunately, the indicators of a concept such as power are never perfect. So it's important to have more than one indicator. Ideally, indicators will be of very different types so that the irrelevant components of each of them will cancel one another out. In the best of all possible worlds, these multiple indicators will point in the same direction, giving greater confidence that the underlying concept has been measured correctly. This point is most convincingly argued by five methodologists in social psychology and sociology:
Four Power Indicators
Working within this framework, there are four different power indicators that have been used by various researchers down through the decades. They can be called (1) who benefits in terms of having the things that are valued in the society? (2) who governs (i.e., sits in the seats that are considered to be powerful)? (3) who wins when there are arguments over issues? and (4) who has a reputation for power (i.e., who stands out in the eyes of their peers)?
Who governs? and Who wins? are the most widely accepted and used of the four indicators. The reputation for power indicator is the least used, especially at the national level of power.
In every society there are experiences and material objects that are highly valued. If it is assumed that everyone in the society would like to have as great a share as possible of these experiences and objects, then those who have the most of what people want are, by inference, the powerful. Put another way, the distribution of valued experiences and objects within a society can be viewed as the most visible and stable outcome of the operation of power within that social system.
In American society, for example, people seek to own property, to earn high incomes, and to live long and healthy lives. All of these "values" are unequally distributed, and all may be utilized as power indicators. In fact, strange though it may sound, differential infant mortality and adult death rates among groups and classes may be the most sensitive indicators of power in the realm of who benefits. No one wants their infants to die, and no one wants to die or see their loved ones die.
However, the primary focus with this type of indicator in most social science studies is on the wealth and income distributions, partly because this information is a little more accessible. This does not mean that wealth and income are the same thing as power, but that the possession of great wealth and income are visible signs that a group or class has power in relation to other groups or classes. There is a big difference between power and wealth, which relates to the whole idea of "power" as a "trait" of a collective, and to the use of indicators to measure concepts.
The argument for using value distributions as power indicators is strengthened by empirical studies showing that such distributions vary from country to country depending upon the relative strength of rival political parties and trade unions. In one study, it was found that the degree of inequality in the income distribution in Western democracies varied inversely with the percentage of social democrats who had been elected to the country's legislature since 1945. ("Social democrats" come from a tradition that began with a socialist orientation and then moved in a more reformist direction. For the most part, social democratic parties have only slightly more ambitious goals than the liberal-labor coalition in the United States; the left wing of the liberal-labor coalition in the United States would feel at home in a strong social democratic party in Western Europe.) The greater the social democratic presence, the greater the amount of income that went to the lower classes (Hewitt, 1977).
In another study, utilizing 18 Western democracies, it was reported that strong trade unions and successful social democratic parties were correlated with greater equality in the income distribution, progressivity in the tax structure, and a higher level of welfare spending (Stephens, 1979, Chapter 4). Thus, there is evidence that value distributions do vary depending on the relative power of contending groups or classes. In Western Europe, these findings suggest, the working class has greater power than it does in the United States. Several reasons for this difference are presented in various places on this website.
Power also can be inferred from studies of who occupies important institutional positions or takes part in important decision-making groups. If a group or class is highly over-represented or under-represented in relation to its proportion of the population, it can be inferred that the group is relatively powerful or powerless, as the case may be. For example, when it is found that women are in only a small percentage of the leadership positions in business and government, even though they make up a majority of the population, it can be inferred that they are relatively powerless in these areas. Similarly, when it is determined that a minority group has only a small percentage of its members in leadership positions, even though it comprises 10 to 20 percent of the population in a given city or state, then the basic processes of power--inclusion and exclusion--are inferred to be at work.
This indicator also can be shown to vary over time and place. The decline of landed aristocrats and the rise to power of business leaders in Great Britain has been charted through their degree of representation in Parliament, for example (Guttsman, 1969). Then, too, as women, African-Americans, Latinos and Asian-Americans began to demand a greater voice in the Unites States in the 1960s and 1970s, their representation in positions of authority began to increase (Zweigenhaft & Domhoff, 2006).
There are many issues over which groups or classes disagree. In the United States different policies are proposed by the corporate-conservative and liberal-labor coalitions on taxation, unionization, business regulation, and the environment. Power can be inferred from these issue conflicts by determining who successfully initiates, modifies, or vetoes policy alternatives. This indicator, by focusing on actions within the decision-making process, comes closest to approximating the process of power that is contained in the formal definition. It is the indicator most preferred by pluralists.
But it is still an inference to say that who wins on issues relates to the underlying concept of power than with the other types of empirical observations used as indicators. For many reasons, it is also the most difficult indicator to use in an accurate way. That's because some aspects of a decision process may remain hidden, some informants may exaggerate or downplay their roles, and people's memories about who did what often become hazy after they move on to other issues. These are just three among several problems with this indicator.
Who Stands Out/Has A Reputation for Power?
The fourth and final indicator, who stands out?, is derived from a person or group's reputation for being powerful, as determined by a series of interviews. The process begins with nominations by a cross-section of observers who are thought to be knowledgeable about the powerful on the basis of their occupational roles (e.g., reporter, administrator, fundraiser). The people nominated by this cross-section of observers are then interviewed and asked for their nominations, along with their opinions regarding the power of the other people on the original list. Any new nominees are then interviewed and asked for their opinions. The process ends, usually within three or four rounds, when the same names keep coming up and no new names are added to the list. This approach, called the reputational method, probably works best at the community or city level, where it is less expensive and time-consuming to apply (Hawley & Svara, 1972; Hunter, 1953). However, it also has been used with good results in two studies of national power in the United States, and in studies of Australia and Norway (Higley & Deacon, 1985; Higley, Field, & Groholt, 1976; Higley & Moore, 1981; Hunter, 1959).
All four of these power indicators have proven to be useful in a wide range of studies, but each of them has its strengths and weaknesses, as would be expected with any indicators in the social sciences. Not only do studies of who wins and loses on specific issues have risks, as pointed out just a minute ago, but the value distributions that determine "who benefits" may be in part determined by unintended consequences or by actions of the powerful to extend some benefits downward. In the case of the "who governs" indicator, people in leadership positions may sometimes have only formal authority and no "real power." As for the reputational method, bias might be introduced by the way in which the original cross-section of informants is chosen, by the wording of the questions asked in the interviews, by unwarranted newspaper publicity for some people, or by the strong leadership role that a person or group played on only one issue.
However, these potential weaknesses present no serious problem if the perspective of multiple indicators outlined earlier is utilized. This is because each of these power indicators does involve different kinds of information drawn from very different types of investigative procedures. A power analysis is most convincing if all four types of indicators triangulate on one particular group or social class.
But it isn't always possible to use all four indicators. Time and money might be limited. Or maybe there is no systematic information on the wealth and income distributions. Perhaps it isn't possible to gain access to decision-makers to interview them. You have to do the best you can with what is possible.
For more details on how to do an actual power structure study, see "How To Do Power Structure Research" in this site.
Basic Axioms about Power
Before plunging into systematic studies of power, it is worthwhile to step back and take a look at what we know about power in general. These are basic axioms that have been around for a long time. Systematic studies don't add much to them. The systematic studies just show how these axioms crop up again and again.
First, we know that power corrupts, and absolute power corrupts absolutely. This famous summary statement comes from a 19th century British observer, Lord Acton. It tells us what we see so often, that the powerful overreach, do dumb things, and often destroy themselves. Whether it is kings, presidents, or CEOs, or petty tyrants around the community, they get so full of themselves that they cut corners, cheat, treat people badly, or make such gross displays of their status that they alienate their followers or stir up an opposition. "The arrogance of power" is another expression for this basic idea. The ancient Greek concept of "hubris" (overbearing pride, presumption) also captures aspects of what too much power does to people, groups, and classes.
Although we are certain that power corrupts, no one knows for sure why this happens. Some hints may be found in studies of small cultic groups, where the charismatic leader only gradually realizes just how much power he or she has as the group gradually becomes increasingly deferential and obedient. So maybe it is just the interaction of leaders and led, in a context of common belief or purpose, that slowly convinces leaders that they are infallible and invincible. There's more to it than that, but that's the starting point (Lalich, 2004). When people tell us we are wonderful, all knowing, and all powerful, it is very difficult not to believe them. Especially if we are a little bit extra narcissistic and exhibitionistic to begin with.
The fact that power corrupts means that power congeals. People and organization try to hold on to it at all costs. Once there is a power structure, it is very hard to change it or dislodge it. Rare is the person who does not cling to power. That's one big reason why George Washington was so admired and trusted by his peers. When he stepped down as the first president, it led to a unique peaceful electoral transfer of power that was unprecedented in Western history.
So, the first and most basic axiom about power leads to the first and most basic axiom for those who don't want to be under the thumb of power structures: no matter what the society, and no matter how benign the new system you hope to create appears to be in theory, there have to be rival power centers to keep power from congealing. Big Government or Big Religion or Big Political Party can be as dangerous as Big Capital.
The second basic axiom concerning power is that the powerful always try to create an outside enemy, real or imagined, to bind the followers to the leaders. The human tendency to divide people into "us" and "them," which social psychology experiments suggest is readily triggered, makes this a very easy task to accomplish. It is almost as if, sadly, there is a need for the "other," that is, the person or group to be wary of.
In the real world, due to the early separation of the original human group into hundreds of groups that wandered off here and there across the world, only to bump into each other again much later and under varying circumstances, there are plenty of other ethnicities, religions, and nationalities to see as a dangerous enemy.
The third basic axiom is divide and conquer. If the followers are not faithfully bound to the leader by the dread of the outside enemy, then leaders can stay in power by favoring some followers and punishing others. A large group of followers is usually at a disadvantage to begin with because it is so hard for them to become organized, but the principle of divide and conquer makes if even more difficult.
Fourth, provide the followers with bread and circuses. There is a mundane version of this axiom that fits with sociological findings: make everyday life possible. Capitalize on the fact that people's everyday life can be compelling for many reasons: a love of friends and family, pleasure in work or artistic or athletic skills, and a desire for routines. If everyday life is possible, then people are less likely to try to challenge a power structure. Then add a little something new every now and then, the circuses, and the rulers are all set.
But it isn't always possible for dominators to provide the conditions that make everyday life and occasional circuses possible. Natural disasters, wars, economic breakdowns, and much more often disrupt everyday life, which puts the power structure at risk. Moreover, opponents of the power structure often find ways to disrupt everyday life on their own and thereby make a challenge of the power structure possible. Historically, of course, this disruption usually has been created by various forms of violence, but there is a good argument to be made that only strategic non-violence makes any sense for challengers to the power structure in a democratic society.
Fifth, and finally, it is important to remember that the powerful believe that the enemy of their enemy is their friend. This is the power version of the idea that "politics make for strange bedfellows." Only by understanding this axiom is it possible to realize that there is a rationale to the constantly shifting alliances that occur in human power struggles at any level from the personal to the international. When the issue is power, principles usually go out the window.
We won't be drawing on these axioms very much on this website. But it is important to know that they are often operating in some of the contexts we do analyze.
Dahl, R. A. (1957). The concept of power. Behavioral Science, 2, 202-210.
Guttsman, W. L. (1969). The English ruling class. London,: Weidenfeld & Nicholson.
Hawley, W., & Svara, J. (1972). The study of community power: A bibliographic review. Santa Barbara: American Bibliographic Center-Clio Press.
Hewitt, C. (1977). The effect of political democracy and social democracy on equality in industrial societies: A cross-national comparison. American Sociological Review, 42, 450-464.
Higley, J., & Deacon, D. (1985). The Australian National Elite in the 1970s and 1980s. Research in Politics and Society, 1, 97-127.
Higley, J., Field, G. L., & Groholt, K. (1976). Elite structure and ideology. New York: Columbia University Press.
Higley, J., & Moore, G. (1981). Elite Integration in the U. S. and Australia. American Political Science Review, 75, 581-597.
Hunter, F. (1953). Community power structure: A study of decision makers. Chapel Hill: University of North Carolina Press.
Hunter, F. (1959). Top leadership, U.S.A. Chapel Hill: University of North Carolina Press.
Lalich, J. (2004). Bounded choice: True believers and charismatic cults. Berkeley: University of California Press.
Lazarsfeld, P. (1966). Concept formation and measurement. In G. DiRenzo (Ed.), Concepts, Theory, and Explanation in the Behavioral Sciences (pp. 144-202). New York: Random House.
Mann, M. (1986). The sources of social power: A history of power from the beginning to A.D. 1760 (Vol. 1). New York: Cambridge University Press.
Mann, M. (1993). The sources of social power: The rise of classes and nation-states, 1760-1914 (Vol. 2). New York: Cambridge University Press.
Russell, B. (1938). Power: A new social analysis. London: Allen and Unwin.
Stephens, J. (1979). The Transition From capitalism to socialism. London: Macmillan.
Webb, E. J. (1981). Nonreactive measures in the social sciences (2nd ed.). Boston: Houghton Mifflin.
Weber, M. (1998). Class,status, and party. In R. Levine (Ed.), Social class and stratification: Classic statements and theoretical debates (Second ed., pp. 43-56). Lanham, MD: Rowman and Littlefield.
Wrong, D. (1995). Power: Its Forms, Bases, and Uses (Second ed.). New Brunswick: Transaction Publishers.
Zweigenhaft, R., & Domhoff, G. W. (2006). Diversity in the power elite (2nd ed.). Lanham, MD: Rowman and Littlefield.