Who Rules America?  By G. William Domhoff, University of California at Santa Cruz

Studying Power

Power Structure Research and the Hope for Democracy

by G. William Domhoff

This article provides an overview of power structure research since its sudden rise in the 1950s with the appearance of Floyd Hunter's Community Power Structure (1953) and C. Wright Mills' The Power Elite (1956). It explains how power structure researchers differ from pluralists, who emphasize individual preferences and voluntary groups, and from Marxists, who start with rival social classes and emphasize class conflict in all times and places.

Power structure researchers begin with organizations as the key to power, not voluntary groups or classes. But unlike the European "elite" theorists from the early 20th century, who claimed elites were inevitable for a mixture of psychological and organizational reasons, and favored the status quo or even more hierarchy, Hunter and Mills were small-d democrats who thought that greater equality and participation are possible.

At the same time, Mills and Hunter had profound differences with the mainstream "organizational theorists" in the social sciences in the 1950s and 1960s, who emphasized the positive aspects of organizations and tried to improve their functioning. More recently, some organizational theorists have focused on organizations as the instruments of the powerful few, but historically they were on the side of the big corporations and foundations who funded their work, striving to eliminate conflict within organizations, make them more efficient, and improve employee morale.

Background and Early History

Many areas of the social sciences and history are concerned with power, but power structure research is the interdisciplinary field that focuses exclusively on mapping power structures in a systematic way and developing a general theory of power. Unlike most fields of inquiry, it did not grow slowly or develop over time. It appeared all of a sudden and with much fanfare between the years 1953 and 1956 in widely read books by sociologists Floyd Hunter and C. Wright Mills.

Hunter, new to academic sociology after many years working for community agencies in various cities, arrived upon the scene unexpectedly with Community Power Structure (1953). The book gave the field its name and offered a new method for discovering power networks and their functioning through systematic interviews. (See "Atlanta: Hunter Was Right" for a detailed overview of the book.)

On the other hand, the publication of Mills' The Power Elite (1956), which focused on power at the national level, was not entirely unexpected. His earlier articles and books had made it clear that he intended to construct an overall portrait of American society which included the level he also called "the high and the mighty" and "the higher circles."

True enough, there had been scattered studies of both community power and national power before Hunter and Mills, but those studies did not present a systematic methodology or put forth a new theory. Power structure research was new because it had several methods and also a theoretical stance that stood between previous traditions in ways that will be explained in a minute by contrasting power structure research with the pluralist and Marxist approaches. Moreover, power structure research had a fairly dramatic impact as these things go in the social sciences. It caused an immediate stir. It elicited critics, and it inspired new researchers.

Hunter's claim that power in Atlanta was in the hands of a relatively few corporate leaders was deeply disturbing to many members of the social science community, and especially the political scientists, who had enjoyed a near monopoly on the study of policy formation and the workings of government up to that point. Convinced that most Americans had at least some say-so at one or another level of American government, even though well-organized and well-financed groups had more power than other groups, most political scientists quickly denied that power could be highly concentrated in any city.

They also doubted that power could be studied by a sociologist who had arrived at his basic findings by asking upper-middle-class professionals and the allegedly powerful themselves to tell him who, in their opinion, were the most powerful people in a city or community. They dubbed this sociometric interview method "the reputational method" and heaped scorn upon it. But soon many sociologists were using and refining the method in dozens of new community studies inspired by the Hunter example (Hawley & Svara, 1972, for useful abstracts of over 300 articles). A few of the skeptical political scientists even ventured into the cities near their universities to determine how specific decisions are "really" made, and emerged with the conviction that their "decisional method" of tracing the specific influences on a variety of decisions had yielded a more valid conclusion: power in America is multi-based, issue-specific, and shifting. In a word, it is "pluralistic," and that is the name usually given to these opponents of Hunter and Mills.

The best and most important of these decisional studies was carried out in the late 1950s in New Haven, Connecticut, a declining city notable only as the home of Yale University. The study is reported in political scientist Robert A. Dahl's Who Governs? (1961). This book won a foundation prize as the best political science book for the year. It quickly became one of the most frequently quoted books within the social science community. I have criticized Dahl's book in great detail using new empirical information I collected in the 1970s and 1990s, by means of both interviews and archival files that only recently became available.

C. Wright Mills

Hunter's book caused consternation within political science, but Mills' The Power Elite created waves in the larger academic community and even spilled onto the pages of popular magazines and political journals. Claiming that the great mass of Americans were dominated by a small triumvirate of corporate rich, political insiders, and military warlords, who came together at the top of their institutional pyramids as an interlocking and overlapping "power elite," Mills criticized both the "higher immorality" of this power elite and the emptiness of pluralist formulations about power in America.

But it would be wrong to overemphasize tone and style in assessing the impact of The Power Elite. Lurking beneath the angry metaphors was a new theory of power that challenged both pluralists and Marxists, a theory which took seriously the vast corporate, military, and administrative hierarchies of twentieth-century America. Moreover, this theory was buttressed by information on the social and economic connections of those in positions of authority in each of his three domains. This "positional method" of studying power, as it came to be known, was only one aspect of the problem as far as Mills was concerned (1956, p. 280-281), but even that aspect was no more acceptable to the pluralists than Hunter's reputational method. Pluralists emphasized even more than Mills that there is no direct relationship between political decision-making and the social origins and occupational careers of policy-makers. "Now it is a remarkable and indeed astounding fact," intoned Dahl, really speaking for all pluralists on this point, "that neither Professor Mills nor Professor Hunter has seriously attempted to examine an array of specific cases to test his major hypothesis" (1958, p. 31).

And so the battle was joined at the methodological level. The reputational and positional methods were ridiculed, and the decisional method was held up as the gold standard. In the end, subsequent studies of Atlanta, using both the reputational and decisional methods, showed that Hunter was right about that city (Jennings, 1964; Stone, 1989; Stone & Sanders, 1987). Furthermore, the two methods pointed to the same suspects in other places as well. In addition, a study of Dallas showed that the reputational, positional, and decisional methods all led to the same conclusion for that city (Thometz, 1963). Undaunted, the pluralists were not satisfied. They said that Southern cities, and especially Atlanta and Dallas, were unique. They kept referring back to the study of New Haven (Polsby, 1980; Wolfinger, 1960). For most social scientists, this demand for methodological purism ended in the 1970s, but some political scientists still insist that only decisional studies can tell us anything about power.

Although Mills' major attack was on pluralist theory, his theory annoyed Marxists almost as much as it did pluralists because he rejected their class-based perspective. For the Marxists of that era, there was a dominant ruling capitalist class, which was locked in constant struggle with a growing and restive working class. Mills dismissed their view in a single footnote as "a rather simple theory," claiming that the phrase "ruling class" was a "badly loaded" one. He said it assumed what needed to be studied and demonstrated, namely, that a particular economic class was able to turn its wealth and status into political power (Mills, 1956, p. 277). More generally, Mills did not think the idea of a "ruling class" allowed enough "autonomy to the political order and its agents, and it says nothing about the military as such" (1956, p. 277). By stating there were multiple organizational bases of power, Mills staked out the kind of "left Weberian" position later developed more fully by sociologist Michael Mann (1986) through his Four Networks theory of power. Mills also doubted that the working class was in inevitable opposition to a power elite or ruling class, or bound to prevail eventually. But it was not until a few years later that he expressed his doubts more dramatically by calling the Marxian faith in the eventual victory of the working class a "labor metaphysic" (Mills, 1962),

The pluralists dismissed Hunter and Mills as "elitists," a pejorative term at the time, implying that they favored such arrangements, and linking them with the elitist theorists in Europe, who were opposed to the spread of the vote and to anything smacking of socialism or the expansion of the welfare state. They had a negative and dismissive view of the great mass of people, and thought that elites are inevitable (Bottomore, 1964; Meisel, 1965). Contrary to these classical elitists, Hunter and Mills based their critique on a belief in the possibility of more egalitarian and participatory social structures. Although they said the current society was dominated by the powerful, they did not celebrate this fact, or see it an inevitable.

It is too often overlooked that the fundamental motivating force for their work was the promise of democracy. The were disappointed that their research led them to believe that there were powerful elites in the United States whose activities interfered with the full flowering of democracy. Power structure research actually derives its moral appeal and intellectual force from a concern with the fulfillment of egalitarian and democratic values.

This does not mean Hunter and Mills were necessarily right in everything they wrote. There are subsequent reasons to think they were too cryptic or wrong on some issues. For example, Hunter did not study the political process very closely, and theorists of all persuasions and subsequent research studies suggested that Mills overestimated the role of the military in the United States since World War II (Domhoff, 1968; Domhoff, 1996, chapter 6). Nor do their conflicts with pluralists and Marxists mean that these disagreements cannot be bridged by new formulations or new findings on some issues. Instead, the main point is that Hunter and Mills had their own starting point and methods, as I now explain more fully.

In closing this little historical vignette, it is important to note that the early power structure researchers did not try to create a general theory. They did not reach as high or as wide as the pluralists, elitists, or Marxists, and they focused on the United States in the 20th century. In an intellectual world that favors big theories, this left power structure research somewhat vulnerable. However, this theoretical shortfall has been remedied by Mann's four-network theory of power. His theory draws on all previous theories to suggest a new synthesis. This synthesis encompasses the major findings of power structure research in the United States and provides a good framework for future studies.

Power Structures Are Based in Organizations

Power structure research differs from rival views because Hunter and Mills assumed that societal power is rooted in large-scale organizations. Their leaders command great resources, have more information than those below them in the hierarchy, and can reward followers and punish critics. They can shape lower-level jobs so that the flexibility and information available to employees is limited. They have the time, money, and contacts to make alliances with the leaders of other organizations, which strengthens their positions inside and outside their organizations.

Rather than studying organizations per se, however, Hunter and Mills concentrated on the ways in which leaders of organizations come together to form a power structure in local communities and at the national level. They therefore differed from the pluralists, who say that power is based in individuals and groups who come together to form voluntary associations and interest groups, and then make their wishes known to responsive government officials through (1) public opinion, (2) lobbying, and (3) their votes in elections Power structure researchers saw these claims as more of a hope than a reality.

Hunter and Mills also differed from the Marxists, who begin with social classes based in the economic system as the basis of power. Unlike the Marxists, they saw the economy as only one possible basis for power. There can be antagonistic social classes, as is the case under capitalism, but that was not their starting point. For them, and in the four-network theory put forward by Mann (1986), class-based power structures are not the only kind that are possible. As Mills (1962, p. 126) said in his critique of Marxism:

I do not mean to replace "economic determinism" by "political determinism" or "military determinism," but only to suggest that the causal weight of each of these types is not subject to any historically universal rule. It must be historically determined in the case of any given society.

Although Hunter and Mills started with organizations as the basis of power, they were not "organizational theorists" in the neutral and benign sense that the label is meant in most areas of the social sciences, including sociology. They had great differences with other social scientists concerning their basic assumptions about the nature of organizations. For them, organizations are instruments of rule and domination within a system of organized irresponsibility and self-serving rationalizations. They are the means of power for an elite who are defined by their positions in the command posts of the corporate, military, and government bureaucratic hierarchies that had grown large and interdependent in the years since the Great Depression and World War II. The old saying, "who says organization says oligarchy" is not true in each and every instance, as critics of the statement are quick to point out, but it is certainly a good starting point given the rarity of democracy and full citizen participation in the history of civilizations over the past 5,000 years.

Organizational theorists, unlike power structure researchers, tend to think of organizations as ponderous, but marvelous systems of coordination whose power is used to attain objectives that benefit society as a whole. They recognize that there can be problems within organizations, but on balance they extol the virtues of organizations. They are not very concerned with power as domination, but with improving morale and communication within specific organizations, increasing the rationality of decision making, and determining the factors that lead to greater cooperation and efficiency within organizations.

In fact, the original interest in organizational theory came in large part from those who worked with and for big corporations, and their research was generously funded by the major foundations of the 1930s and 1940s. Make no mistake, the original organizational theorists were out to solve the problems faced by the corporate rich. One of their specialties was "industrial relations," a benign term for trying to avoid and/or subvert unions. More recently, say in the past 25 years, some organizational theorists within sociology, political science, and the policy sciences have shown greater interest in issues of power inside and outside organizations. But most of them are still far from a critical stance toward the organizationally based power structure called American society.

For one important recent exception, see Charles Perrow, Organizing America: Wealth, Power, and The Origin of Corporate Capitalism (2002). From his vantage point as one of the leading organizational theorists since the early 1970s, the big private organizations that rose to power in the last part of the 19th century, and have been there ever since, have the following effects:

  • They create wage dependence. Over 90% of Americans work for someone else, and over 50% work for organizations (profit and nonprofit) with over 500 employees. When it comes to the for-profit organizations, where there were 5.5 million corporations, 2.0 million partnerships, and an estimated 17.7 million non-farm proprietorships in 2000, 44% of all private-sector employees worked for the 8,300 companies with 1,000 or more employees.
  • They centralize profits. Just 500 companies, the heart of the corporate community, earned 57% of all profits made in the United States in the year 2000, while employing 16.3 percent of the private-sector workforce (White, 2002).
  • They are big enough to shape their "environments," i.e. communities and governments.
  • They lead to a concentration of wealth and power.

A power structure researcher could not ask for much more than what is on this list. The common assumptions and interests shared by power structure researchers and organizational theorists like Perrow, four in all, are discussed toward the end of this document.

Further Detail on Power Structure Research

So, it is clear that power structure research is neither pluralist, Marxist, or mere organizational theory as it is conventionally understood. Power structure research takes a critical stance toward the way unequal societies are organized, and it assumes the potential of great equality, participation, and democracy. But what exactly do we mean by power structure research?

Power Structures And The Power Elite

For purposes of this Web site, and for most current power structure researchers, a "power structure" is a network of organizations and roles within a city or society that is responsible for maintaining the general social structure and shaping new policy initiatives. A "power elite," on the other hand, is the set of people who fill the roles within the power structure. Because the social order maintained by the power structure is a stratified one, with great inequalities of wealth and income, a power structure is also a system of organized domination, and the power elite often will use intimidation and coercion on their critics and opponents if necessary (see Hunter, 1953, pp. 193-194, 242-243).

The emphasis within power structure research is not on the day-to-day routines of the lower levels of the power system, which tends to be the case in mainstream studies of organizations. That level is accepted as a given by power structure researchers, and taken for granted by most members of a society, who go through their routines as long as the organizational structure meshes together. In power structure research, the focus is on the introduction of new policies and the defense of the social order against challenges from within or without. "Broadly speaking," wrote Hunter, "the maintenance of this order falls to the lot of almost every person in the community, but the establishment of changes in the old order falls to the lot of relatively few" (1953, p. 9). Put another way, power structures become most visible when some relatively powerless group or courageous activists try to push up against them. Accidents and scandals that can't be controlled also tend to expose the contours of power structures, such as nuclear accidents, oil spills, Enron-type scandals, purloined documents, and leaked memos (Molotch, 2004; Molotch & Lester, 2004).

However, the powerful cannot control the whole social system. The concept of a power elite does not imply that all aspects of a social system are controlled by the very few, or that large-scale changes in a social system can be created or stopped even by the most organized and self-conscious of power elites. For Mills (1956, pp. 20-22), the power elite sitting atop the major institutional hierarchies of modern America merely make those decisions of national consequence that are made; they do not control history.

Analyzing Power Structures

There are two general aspects of any study that attempts to delineate a power structure and understand its workings. The first aspect is a "network analysis." A network analysis provides an empirical picture of (1) the interconnections among the key organizations and people in the power structure and (2) the operations of that network on policy issues. Such studies can reveal power structures to be more or less tightly knit, more or less focused in a few organizations or individuals, and more or less linked to a single social class. The second aspect of a power structure study is a "content analysis" of what is said and done within the power network. Such studies can provide information on the substance of new policies and an understanding of the strategies and values that underlie the policy planning of the power elite.

As indicated earlier, there are three methods by which the networks of the power structure can be constructed: positional, reputational, and decisional. They lead to three indications of power:

Is there a group or class that is over-represented in positions of authority, as compared to its percentage of the population, which implies the group is powerful? This is the who sits? indicator.

Who receives the most nominations from those interviewed in reputational studies? This is the who stands out? indicator.

Who has the final say-so, in terms of proposing or vetoing policy changes, on a variety of issues? This is the who wins? indicator.

A positional analysis uses printed public information available in libraries and other archives , including the internet, to establish the leadership interlocks among profit, nonprofit, and governmental agencies as well as to trace the flows of money, information, and other resources among the same organizations. Such studies often begin with the names and interconnections of thousands of people and a wide range of organizations. Then information on their social, educational, and occupational backgrounds is obtained to see if there are any patterns of over-representation for any social group, universities, or occupational settings.

This information can be analyzed in great detail with the help of computer programs based in the mathematics of graph theory, matrix algebra, and Boolean algebra. For a complete presentation of informational sources and methods relevant to network analysis, see the invaluable web site, Who Rules? An Internet Guide to Power Structure Research, by Val Burris of the University of Oregon.

Mills' 1956 classic on the American power elite is rooted in a positional analysis, and most subsequent research on power at the national level in the United States and other countries relies on this method.

A reputational study, in contrast to the positional method, is based on the personal opinions of a wide range of people who are interviewed about who they think to be the key people and organizations in the power structure. As already mentioned, this method was developed and utilized by Hunter, who later characterized it by saying it was equivalent to arriving in a previously unknown tribal society and asking the natives to take you to their leaders. Although such a study can begin with a random survey of people within a city or nation, in practice the method relies on the observations and insights of expert advisers and members of the power structure itself. Like the positional method, it can be made quite exact and mathematical if members of the power structure are systematically interviewed with questions that assign ranks or weightings to people and institutions and their involvement in a range of policy issues. With the exception of work by Hunter (1959), Allan Barton (1971; 1985), and Gwen Moore (1979) on the United States, and by John Higley (1985; 1976; 1981), who has done important reputational studies in the United States, Australia, and Norway, this method has found its greatest application at the local level, where it is easier and cheaper to use.

A decisional mapping of a power structure is based on case studies of the people, organizations, and pressure groups that become involved in trying to influence the outcomes of specific policy initiatives in a range of "issue-areas" ranging from local planning policies to federal taxation policies to foreign policies.. Although pluralists claim this method was overlooked by the original power structure researchers, Hunter (1953, p. 214) in fact asked all his informants, "What are two major issues or projects before the community today?" He found out that local leaders expressed their greatest concern with land acquisition and the development of new buildings. This is exactly what was later predicted by Harvey Molotch's (1976; 1979; 1998) "city as a growth machine" theory, which stresses that land-based growth coalitions dominate city governments with the intent of increasing the value of land and buildings.

After the vigorous debates in the 1950s and 1960s about the relative merits of the three methods of identifying power structures, there has been greater recognition of their similarities. In effect, all three methods generate a list of people and their connections to each other and to organizations and issues. And, as noted earlier, the methods usually provide about the same picture even though each has its unique contributions to make in a full-scale study that can utilize all three. It now seems that the best approach is to use as many of the methods as feasible in the circumstances of any specific study.

The content analyses that compose the second general aspect of a power structure study are conducted on the memos, minutes, letters, speeches, position papers, and legislative proposals that are produced by central organizations and well-placed leaders within the network of elite institutions and people. These content analyses can entail no more than the researcher's general impressions based upon an application of his or her subjective categories to the material under inspection. This is in effect what journalists, commentators, and policy analysts do when they have to provide instant accounts of new policy directions.

In the past, formal content analyses relied on carefully constructed analytical categories based on theoretical concerns. Now content analysis also can be carried out with software that does not require prior analytic categories. Such software figures out what words or phrases appear most frequently, organizes them into clusters, and determines which clusters go with other clusters. It is a powerful way to detect new policy directions and changes in rationales and ideologies.

The Five Substantive Areas of Power Structure Research

The network and content analyses carried out by anyone who studies power empirically, whatever their theoretical orientation, have been focused on five very different aspects of the social structure, but always with a concern for if and how these aspects come together to form the power structure. These five areas of interest are:

  1. The social upper class
  2. Corporations
  3. Non-profit organizations such as foundations, think tanks, and policy-discussion groups
  4. Political parties and elections
  5. The "state" or government

Both Hunter and Mills studied the social upper class as one dimension of the power structure. They did so by focusing on the neighborhoods, schools, private clubs, and other social organizations that constitute the institutional infrastructure for the interacting and intermarrying families that make up this class. Hunter found that private clubs in Atlanta are sometimes the setting for policy discussions among members of the power structure, and Mills stressed that members of the upper class are important participants in corporate and governmental bureaucracies. However, neither theorist attached primary importance to the upper class in his conclusions about the American power structure.

Unlike Hunter and Mills, class-dominance theorists of both Marxist and non-Marxist persuasions (I, for one, am a class-dominance theorist of a non-Marxist persuasion, but there are others) place a greater emphasis on the upper class as a basic part of the power structure. For Marxists, who begin with the assumption that power structures are rooted in the ownership and control of income-producing property, studies of the upper class can be relevant and useful because they reveal the consciousness and policy preferences of the capitalist class in its conflict with the working class over the production and disposition of the economic surplus. For non-Marxists, ownership of profitable private property is one important basis among several for upper-class domination. More generally, studies of the upper class, whether Marxist or non-Marxist, have demonstrated its socialization practices, its social cohesiveness, its continuity, and its class consciousness through historical and interview studies of men, women, and children. Some of the most interesting and revealing of such studies have been done on women of the upper class (Daniels, 1988; Kendall, 2002; Ostrander, 1984).

Mills and Hunter also focused on corporations in their power structure studies, putting greater emphasis on their centrality in the power equation than on the social upper class. Although some researchers since have emphasized the intertwined nature of the upper class and the corporate community, the fact is that corporations have been regarded as major actors in their own right by all researchers interested in power.

The focus on nonprofit organizations as important parts of the power structure is perhaps the most controversial claim of power structure research. Most mainstream theorists talk about these organizations as charitable or educational institutions with no power function, or as the "non-profit sector," or even as a "third force." Power structure researchers, however, believe that most, if not all, of these organizations have a class or corporation bias in their values and policies due to their histories, current funding, or present leadership. Some of them come together as a policy-planning network that is linked to corporations on the one side and the government on the other (Darknell, 1975; Domhoff, 1990; Domhoff, 2005; Weinstein, 1968). Many others are social service organizations for middle-class and low-income people, financed by the corporate rich through foundations, the United Way, and other charities (Shaw, 1999. Chapter 5).

(For a recent summary and update on the role of nonprofits, including several new case studies, see a 2009 article by Bill Domhoff that appeared in American Behavioral Scientist.)

Neither Hunter nor Mills paid very much attention to political parties and elections, which critics believe to be one of the main weaknesses of the field of study they created. Mills explicitly relegated legislative battles to the "middle levels of power." Hunter's brief account of relations between the Atlanta power structure and the Georgia state government left little room for independent action on the part of elected officials. Later research suggests they were wrong to ignore these issues, or at least it is so argued on this web site. Today, power structure researchers emphasize the critical role of large campaign contributions as one crucial way by which members of the power elite have been able to have great influence in political parties and elections (Burris, 1987; Burris, 1991; Clawson, Neustadtl, & Weller, 1998; Webber, 2000). Most of the money comes from wealthy individuals, but some of it comes from corporate-financed PACs (Political Action Committees)

The fifth and final concern of power structure research has been the role of "the state" (ie. the "government," to use the term preferred in everyday English). Due to its open-ended and empirical nature, power structure research does not begin with the theoretical assumption that the state is always part of the power structure. Instead, it begins with the idea that the key policy-initiating and decision-making groups and organizations in a society must be identified. Those organizations may or may not include the state, which is an organizational base because it has the important function of regulating activity in the area of which it is a part, not because it is always an instrument of domination. There have been eras where the state was not central to the power structure, such as the Middle Ages in Europe (Mann, 1986).

However, for most countries in modern times, power structure research quickly moves in actual practice to a consideration of how the state fits into the power system. This is necessary because it cannot be assumed that the state always serves the interests of a dominant social class. Indeed, power structure research derives its basic rationale from the assumption that the state in large-scale societies is at least potentially autonomous from social classes and private organizations, as Mills first emphasized. He thereby anticipated the concerns of a rival theory group, the "state autonomy theorists," who have mischaracterized power structure research in several ways (see Mills, 1956, pp. 170 and 277n).

In fact, power structure researchers have devoted almost as much attention to the study of government as they have to corporations. Most of all, they have studied the social backgrounds and corporation connections of appointed and elected officials, and they have found that members of the upper class and corporate communities are greatly over-represented in government positions in comparison to their proportion of the general population (Burch, 1981; Mintz, 1975; Salzman & Domhoff, 1980). They also have traced policy initiatives from corporations and policy groups into government, showing that certain issue-areas, and especially foreign policy, are heavily influenced, if not completely dominated, by these private organizations.

Current Commonalities With Organizational Theory

As noted earlier, power structure research and organizational theory both start with an emphasis on organizations. This makes them different from pluralists and Marxists, but they originally held very different views of organizations and had very different reasons for studying them. However, some of these differences have declined over the decades. Today, power structure and organizational researchers share several assumptions and interests in common.

First, they share an interest in identifying key positions and the social backgrounds of the people occupying those positions. They share this interest because "socioeconomic background is a means to predict what the internal values of managers will be" (DiTomaso, 1980, p. 258). Such predictability is extremely important, they agree, when it comes to those people who hold positions where there is a wide latitude of choice and action, namely, positions involving (1) the allocation of resources, (2) great uncertainty about the choices that will be faced, and (3) contacts outside the organization. At the same time, both organizational theory and power structure research are concerned with the ways in which roles are "institutionalized" within a set of circumscribed rules and practices so that discretion and social background will be of less importance.

Second, both organizational and power structure researchers are concerned with the inter-organizational environmental within which organizations must seek out the resources they need in order to reach their goals. Power structure research conceptualizes this inter-organizational environment as the power structure itself. It emphasizes that the necessary resources -- whether money, expertise, or raw materials -- are properties of the power structure as a whole as well as assets and needs of individual organizations within it.

Third, organizational and power structure researchers agree that the American government is a critical element in the environment of any large private organization in the United States in that it secures order and confers legitimacy on the existence of such organizations. Thus, power structure and organizational researchers both see governmental decisions as an area of great importance and potential uncertainty for private organizations

At the same time, they agree that those with roles within the government also have a need to reduce "uncertainty" and to stabilize their own environments. Thus, private corporations and governments at all levels of American society try to enter into enduring coalitions that become, for power structure researchers, a "power elite" at the national level and a "growth coalition" at the city level. However, power structure researchers continue to disagree with organizationally oriented theorists in that they are not just interested in the fact that governments enter into coalitions. They want to know if the private interests, meaning the corporations for the United States, have the upper hand in the coalition. They want to know the substance of the matter, not just the institutional form.

Fourth, the common concerns of organizational and power structure researchers reveal themselves most directly in the least studied area in both fields: the control of private organizations, profit and non-profit alike, by boards of directors (which include top-level executives). For class-dominance theorists, the boards are the location that brings together the class-wide concerns of wealthy directors with the organizational concerns of the day-to-day executives.

Far too little research has been done on the way in which this top leadership group relates to the middle and lower levels of an organization on the one hand and to other organizations on the other hand. That's partly because this is a very difficult area to study: it is not easy to gain access. Then, too, organizational theorists often assume that the boards are of little consequence anyhow. They sometimes claim that boards are controlled by the mangers within the organizations. This view is an echo of the old "managerial theory," in which the owners of corporations are allegedly replaced as power figures by the top executives.

So far, most research involving boards of directors has concerned the social backgrounds of directors and the linkages created among boards by shared (interlocking) directors. Work of a historical and interview nature on the actual functioning of these boards would draw the two fields closer together. For now, the best substantive work supporting the class-dominance theory on the role of boards of directors has been done on non-profit boards, where access has been somewhat more available (Ostrander, 1987).

The Return of a Revised Elite Theory

Traditional elite theory, created by Gaetano Mosca (1858-1941), Vilfredo Pareto (1848-1923), and Robert Michels (1875-1936) in the late 19th and early 20th centuries, had no influence on the work of Mills and Hunter. Mosca and Pareto were first and foremost anti-Marxists and anti-egalitarians with a jaundiced view of human nature. They were cynical about politics and thought that elites are inevitable. Michels, who started out as a strong socialist and was a member of the large German Social Democratic Party in the years before World War I, became a conservative and an Italian nationalist (by then he was a professor in Italy). He came to the conclusion that even parties advocating socialism are controlled by the leaders at the top. He is famous for the phrase "the iron law of oligarchy" to describe the undemocratic tendencies in any organization (Mitzman, 1973, Chapter 27).

In addition, classical elitist theory put as much emphasis on psychology and strong men as it did on sociology. In Pareto's extreme formulation, men willing to use force ("lions") come to leadership and create strong organizations, to be followed by cunning caretakers ("foxes"), who are cautious and mostly try to keep a good thing going. They are soon superseded by new lions, and so the cycle goes, a "circulation of elites." As for the general citizenry, they have little or no impact, and don't deserve any. Pareto also had a very low opinion of women. As one expert on Pareto summarized his theory, "elites merely circulate between lions and foxes while feeding on the sheep or masses" (Adams, 2005, p. 547).

However, traditional elite theory also had a sociological dimension, an emphasis on elites as the leaders within organizations. It is this aspect of the theory, stripped of the psychological and anti-feminist aspects, that has been revived and revised as a basis for research on elites in modern countries. For example, in their article entitled "Invitation to Elite Theory: The Basic Contentions Reconsidered," sociologists Michael G. Burton and John Higley (1987a) recast the theory in terms of three basic assumptions:

  • Elites are inevitable in large-scale, bureaucratized societies. Elites are defined as the people who come to leadership in the major organizations in such societies.
  • Variations in elite structure and functioning are very important in determining the general nature of a society, such as its openness to compromises and its use of elections to settle policy disputes. According to elite theorists, voting in Western societies began when rival elites in European countries, exhausted by their brutal and unending conflicts, compromised their major differences in "pacts" or "settlements" (Burton & Higley, 1987b; Higley, Hoffmann-Lange, Kadushin, & Moore, 1991).
  • The relationship between elites and non-elites is one of interdependence, which puts limits on what one side can do to the other. It follows that elites are not omnipotent, and that there is no inherent opposition between elites and non-elites.

Although still diametrically opposed to Marxism, this revised elite theory is clearly a sociological one that has some commonalities with the other alternative theories. For example, when pluralists talk of "polyarchy," stressing the "dispersed inequalities" in American society, the author of an excellent elite analysis of France notes the similarities between the two theories (Dogan, 2003).

It is also important to mention that elite theory no longer contains a political agenda, except in the sense that it denies that non-hierarchical societies are possible:

To the extent that it rules out all social and political utopias, elite theory can perhaps be termed 'conservative,' though we would prefer the term 'realist.' Beyond this, however, its contentions are probably compatible with just about any major ideological position, including democratic socialism. Thus, criticizing and attacking a particular elite's functioning is fully consistent with the basic contentions of elite theory. (Burton & Higley, 1987a, p. 237.)

The Decline of Power Structure Research — And Why Did It Happen?

After a fast start in the 1950s, power structure research came of age in the 1960s in the context of the general social ferment of that era. Activist student organizations made power structure studies that were often useful in challenging corporations or their own universities (some titles: "Who Rules Columbia?", "How Harvard Rules," and "Go To School, Learn to Rule," a study of Yale and its power both nationally and in New Haven). Power structure research also attracted many new social science researchers, including the author of this web site. It adopted "membership network analysis" as its methodology and seemed to be on its way to respectability (Breiger, 1974).

Now-classic studies were done using early software to study corporation networks (Mintz & Schwartz, 1981; Mintz & Schwartz, 1983; Mizruchi, 1982; Sonquist & Koenig, 1975). There were also useful and workmanlike studies of club networks (Domhoff, 1975) and of the intersections of corporations, foundations, and policy-discussion groups (Bonacich & Domhoff, 1981; Salzman & Domhoff, 1983).

But power structure research began to wane in the 1980s. It was eclipsed by:

  • State autonomy theorists, who do no original research, relying instead on secondary sources based on original research by historians (Orloff, 1993; Skocpol, 1980; Skocpol, 1992). State autonomy theory is critiqued in detail on this site.
  • There also was a rise in interest in the new institutionalism, a rather bland theory which focuses on norms and practices within institutions, and on how various organizations develop "institutional fields" and relate to each other (Powell & DiMaggio, 1991). There's little or no concern with power or domination. As noted by Perrow (2002, p. 19), "this theory emphasizes routines, imitation, unreflective responses, custom, and normative practices, and convergence of organizational forms; it deemphasizes power and conflict." For my money, it is a return to conventional wisdom. When the state autonomy theorists had trouble locating empirical studies of the present-day United States that fit their theory, they took to calling themselves "historical institutionalists."
  • Meanwhile, those who practice in the burgeoning field of network analysis ignore the early corporate interlock studies and continue to use hypothetical or small-group data for the most part while they hone their methodologies. It is as if these rigorous people are embarrassed by part of their early history.

No one knows why this decline happened. It would make an interesting study in the history and sociology of the social sciences to try to figure it out. Here are a few of the possibilities, which are not mutually exclusive:

  • As the large-scale and highly visible civil rights and anti-war movements declined, interest in power structures declined; there was a "right turn" in the society at large.
  • As the concerns of women, people of color, and gays and lesbians became more salient and pressing, social movements and issues of identity attracted more attention from new researchers, who focused on the middle-level settings where these previously excluded groups were trying to make breakthroughs.
  • The arguments among the rival theorists (pluralists, non-Marxist class-dominance theorists, and various types of Marxists) became so contentious and politicized that new sociologists avoided any involvement and went into more cooperative fields.
  • None of the power structure researchers was at the elite private universities that train the graduate students who go on to shape the discipline, and few were at high-status public universities with large graduate programs. As sophisticated network studies show, good jobs go to those who study with professors at the major universities (Burris, 2004, for new findings and a summary of other studies). These professors in effect set the agenda for the discipline. Basically, that means professors at Harvard, Yale, Princeton, Chicago, Columbia, Stanford, Michigan, Wisconsin, and the University of California at Berkeley.

Although there are now fewer full-time practitioners of power structure research, claims about the structure and distribution of power continue to be made by researchers in other fields, not to mention by investigative journalists, pundits, and ideologues of all persuasions. This site can serve to provide perspective on their claims, and I will also add critiques of some of these people from time to time.

For those who want to see how these general points can be turned into actual studies at the local and national levels, "How To Do Power Structure Research" in this site.


References

Adams, B. (2005). Vilfredo Pareto. In G. Ritzer (Ed.), Encyclopedia of social theory (Vol. 2, pp. 544-547). Thousand Oaks, CA: Sage Publications.

Barton, A. H. (1971). Empirical Methods and Elite Theories. Rivista di Sociologia, 22(Jan-Aug), 57-82.

Barton, A. H. (1985). Background, Attitudes, and Activities of American Elites. Research in Politics and Society, 1, 173-218.

Bonacich, P., & Domhoff, G. W. (1981). Latent classes and group membership. Social Networks, 3, 175-196.

Bottomore, T. B. (1964). Elites and society. New York: Basic Books.

Breiger, R. L. (1974). The duality of persons and groups. Social Forces, 53, 181-190.

Burch, P. H. (1981). Elies in American history: The Civil War to the New Deal (Vol. 2). New York: Holmes & Meier.

Burris, V. (1987). The Political Partisanship of American Business: A Study of Corporate Political Action Committees. American Sociological Review, 52(6), 732-744.

Burris, V. (1991). Director Interlocks and the Political Behavior of Corporations and Corporate Elites. Social Science Quarterly, 72(3), 537-551.

Burris, V. (2004). The academic caste system: Presitge hierarchies in PhD exchange networks. American Sociological Review, 69, 239-264.

Burton, M., & Higley, J. (1987a). Invitation to elite theory: The basic contentions reconsidered. In G. W. Domhoff & T. Dye (Eds.), Power Elites and Organizations (pp. 219-238). Beverly Hills: Sage.

Burton, M. G., & Higley, J. (1987b). Elite settlements. American Sociological Review, 52, 295-307.

Clawson, D., Neustadtl, A., & Weller, M. (1998). Dollars and votes: How business campaign contributions subvert democracy. Philadelphia: Temple University Press.

Dahl, R. A. (1958). A Critique of the Ruling Elite Model. American Political Science Review, 52, 463-469.

Dahl, R. A. (1961). Who governs? Democracy and power in an American city. . New Haven,: Yale University Press.

Daniels, A. K. (1988). Invisible careers: Women civic leaders from the volunteer world. Chicago: University of Chicago Press.

Darknell, F. (1975). The Carnegie Council For Policy Studies In Higher Education: A new policy group for the ruling class. The Insurgent Sociologist, 5(3), 106-114.

DiTomaso, N. (1980). Organizational analysis and power structure research. In G. W. Domhoff (Ed.), Power Structure Research (pp. 255-268). Beverly Hills: Sage.

Dogan, M. (2003). Is there a ruling class in France? Comparative Sociology, 2, 17-89.

Domhoff, G. W. (1968). The Power Elite and its critics. In G. W. Domhoff & H. B. Ballard (Eds.), C. Wright Mills and The Power Elite (pp. 251-278). Boston: Beacon Press.

Domhoff, G. W. (1975). Social clubs, policy-planning groups, and corporations: A network study of ruling-class cohesiveness. The Insurgent Sociologist, 5, 173-184.

Domhoff, G. W. (1990). The power elite and the state: How policy Is made in America. Hawthorne, NY: Aldine de Gruyter.

Domhoff, G. W. (1996). State Autonomy or Class Dominance? Case Studies on Policy Making in America. Hawthorne, NY: Aldine de Gruyter.

Domhoff, G. W. (2005). Who Rules America? (Fifth ed.). New York: McGraw-Hill.

Hawley, W., & Svara, J. (1972). The study of community power: A bibliographic review. Santa Barbara: American Bibliographic Center-Clio Press.

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., Hoffmann-Lange, U., Kadushin, C., & Moore, G. (1991). Elite integration in stable democracies: A reconsideration. European Sociological Review, 7(1), 35-53.

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.

Jennings, M. K. (1964). Community influentials: The elites of Atlanta. New York: Free Press.

Kendall, D. (2002). The power of good deeds: Privileged women and the social reproduction of class. Lanham, MD: Rowman and Littlefield.

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.

Meisel, J. H. (1965). Introduction: Pareto and Mosca. In J. H. Meisel (Ed.), Pareto and Mosca (pp. 1-44). Englewood Cliffs, NJ: Prentice-Hall.

Mills, C. W. (1956). The power elite. New York: Oxford University Press.

Mills, C. W. (1962). The Marxists. New York: Dell.

Mintz, B. (1975). The president's cabinet, 1897-1972: A contribution to the power structure debate. Insurgent Sociologist, 5, 131-148.

Mintz, B., & Schwartz, M. (1981). The structure of intercorporate unity in American business. Social Problems, 29(2), 87-103.

Mintz, B., & Schwartz, M. (1983). Financial interest groups and interlocking directorates. Social Science History, 7, 183-204.

Mitzman, A. (1973). Society and estrangement: Three sociologists of Imperial Germany. New York: Alfred A. Knopf.

Mizruchi, M. S. (1982). The American corporate network, 1904-1974. Beverly Hills: Sage Publications.

Molotch, H. (1976). The city as a growth machine. American Journal of Sociology, 82, 309-330.

Molotch, H. (1979). Capital and neighborhood in the United States: Some conceptual links. Urban Affairs Quarterly, 14, 289-312.

Molotch, H. (1998). Urban America: Crushed in the growth machine. In C. Lo & M. Schwartz (Eds.), Social Policy and the Conservative Agenda (pp. 53-71). Malden, MA: Blackwell.

Molotch, H. (2004). Spilling out (again). In R. Levine (Ed.), Enriching the sociological imagination: How radical sociology changed the discipline (pp. 87-90). Boston: Brill.

Molotch, H., & Lester, M. (2004). Accidents, scandals, and routines: Resources for insurgent methodology. In R. Levine (Ed.), Enriching the sociological imagination: How radical sociology changed the discipline (pp. 91-104). Boston: Brill.

Moore, G. (1979). The Structure of a National Elite Network. American Sociological Review, 44, 673-692.

Orloff, A. S. (1993). The politics of pensions: A comparative analysis of Britain, Canada, and the United States, 1880-1940. Madison: University of Wisconsin Press.

Ostrander, S. (1987). Elite domination in private social agencies: How it happens and how it is challenged. In G. W. Domhoff & T. Dye (Eds.), Power Elites and Organizations (pp. 85-102). Beverly Hills: Sage.

Ostrander, S. A. (1984). Women of the upper class. Philadelphia: Temple University Press.

Perrow, C. (2002). Organizing America: Wealth, power, and the origin of corporate capitalism. Princeton: Princeton University Press.

Polsby, N. (1980). Community power and political theory (2nd ed.). New Haven, CT: Yale University Press.

Powell, W., & DiMaggio, P. (1991). The new institutionalism in organizational analysis. Chicago: University of Chicago Press.

Salzman, H., & Domhoff, G. W. (1980). The corporate community and government: Do they interlock? In G. W. Domhoff (Ed.), Power Structure Research (pp. 227-254). Beverly Hills: Sage.

Salzman, H., & Domhoff, G. W. (1983). Nonprofit organizations and the corporate community. Social Science History, 7, 205-216.

Shaw, R. (1999). Reclaiming America: Nike, clean air, and the new national activism. Berkeley: University of California Press.

Skocpol, T. (1980). Political responses to capitalist crisis: Neo-Marxist theories of the state and the case of the New Deal. Politics and Society, 10, 155-202.

Skocpol, T. (1992). Protecting soldiers and mothers: The political origins of social policy in the United States. Cambridge, MA.: Harvard University Pressf.

Sonquist, J., & Koenig, T. (1975). Interlocking directorates in the top U. S. corporations. Insurgent Sociologist, 5, 196-229.

Stone, C. (1989). Regime politics. Lawrence: University of Kansas Press.

Stone, C., & Sanders, H. (1987). The politics of urban redevelopment. Lawrence: University Press of Kansas.

Thometz, C. (1963). The decision makers: The power structure of Dallas. Dallas: Southern Methodist University Press.

Webber, M. (2000). New Deal fat cats: Business, labor, and campaign finance in the 1936 presidential election. New York: Fordham University Press.

Weinstein, J. (1968). The corporate ideal in the liberal state, 1900-1918. Boston: Beacon Press.

White, L. J. (2002). Trends in aggregate concentration in the United States. Journal of Economic Perspectives, 16, 137-160.

Wolfinger, R. (1960). Reputation and reality in the study of community power. American Sociological Review, 25, 636-644.