http://whorulesamerica.net/power/domhoff_interview_2019.html (retrieved July 17, 2024)

Research on the Power Elite and Membership Network Analysis

An Interview with G. William Domhoff in 2019

By Francois Denord[1], Paul Lagneau-Ymonet[2], & Silvain Laurens[3]

This interview appeared in Researching Elites and Power: Theory, Methods, Analyses (2020), F. Denord, M. Palme, & B. Réau (Eds.).


In this interview, G. William Domhoff, author of the classic book Who Rules America?, recounts the genesis of Who Rules America? in 1967 and the development of power structure research since the 1960s. Who Rules America? has been frequently updated in both its theory and empirical information, including a seventh edition in 2014.

Let's start from the very beginning. Basically, your background, education...

GWD: So, I'm a middle American from the middle west. I'm from Ohio. And I grew up in a family that was not very political, if it was political at all. I never heard anything about politics. My father was not very religious, and neither really was my mother, but she took me and my sister, who is three years younger than I am, to a Methodist church. I grew up in that atmosphere, which made me part of the mainstream, but my parents, probably inadvertently, left enough room for doubt, which I now can see made it possible for me to become a scientific thinker. My dad worked as a manager of an office owned by a small loan company, and my mother was a homemaker, who was always focused and in motion. For me, it was a very apolitical time in the 1950s. I later learned, though, that I grew up in one of the most ethnically diverse and segregated areas in America, Cleveland and the many surrounding suburbs, when Cleveland was one of the six or seven largest cities in the country in the 1940s and 1950s, if you can imagine that today, now that Cleveland has shrunk to a minor and failing city.

When I later looked at the voting records of people in the suburb of Cleveland where I grew up, I learned the place was 80 or 90 percent Republican. I grew up in a world of white Republicans, although I didn't realize that at the time because I wasn't interested in politics. However, starting at age 18, I accidently ended up with a very wide experience over the next eight or nine years. I spent four years in North Carolina and three in Florida. I met a woman from California when I was traveling in Europe in the summer of 1960, and we were married in the summer of 1961. We moved from Florida to California in the spring of 1962, and I have loved that state and lived there ever since. But that is getting ahead of the story a little bit. It was a future I never could have predicted.

Anyhow, during my high school years, I was a very good student, in the sense of learning what I was supposed to learn and receiving excellent grades, but my passion was sports. I was successful as a member of the football, basketball, and baseball teams. As a result, I went to college to play baseball at Duke University in Durham, North Carolina, in the Fall of 1958, which seemed a million miles away from Ohio in that era. It turned out to be a different world, one that was fully segregated in every aspect. I soon learned I wasn't much of a baseball player, although I was a starting left fielder for two of the three years I was on the varsity team. At the same time, I was considered to be an oddball by most of my teammates because I was a good and serious student, so I began to learned what it feels like to be marginal for the first time in my life. I was also in a fraternity and worked for the college newspaper, which came out once or twice a week. And I did receive a great education both inside and outside the university. It included a two-semester class on the Bible, which was required of all first-year students because Duke was a strongly Methodist school at the time, although I did not know that fact when I decided to go there. The Bible course, Old Testament one semester, New Testament the other semester, was in effect a course in history and anthropology, and was fascinating. It had the opposite effect on me and many other students than I would suppose the founders of Duke had imagined when they insisted upon that requirement back in the 1920s.

Outside the classroom, I learned about the diversity of the Cleveland area in detail during the summer between my sophomore and junior years in college, due to an interview and observation study I did as a temporary summer poll taker for a large local newspaper back in my home area. The newspaper's leaders, who bragged on the masthead that the newspaper had the largest circulation of any newspaper in Ohio, had decided they wanted to know about the demographics and reading habits of its readers. I was chosen to do the interviews because I already had worked for them for a summer doing chores and errands in the newsroom, including carrying newly finished drafts of stories down the stairs to the floor where the newspaper was printed. I went to every possible neighbourhood in the Cleveland area over the next three months, knocking on doors, with a different page of the newspaper in my hands each week. I realized a few years later that this interviewing experience may have been my first introduction to sociology. Back then, though, I was basically headed for a career in journalism.

Another big part of my extra-curricular education during my college years, based on knowing elite Southern white students, turned out to have a lifelong impact. I came to realize soon thereafter that elite southern whites have a very complicated and contradictory sociological outlook on their position in the power structure. On the one hand, they grew up in, and still today grow up in, an arrogant, dominant, white-supremacist culture due to their total control of the black population in one way or another from the middle of the sixteenth century until the late 1960s. This control always involved the quick and brutal use of murderous force, whether during the centuries of slavery, or the century of total segregation. On the other hand, they see themselves as a conquered and humiliated group because they have been defeated twice by more powerful whites in the North, first in the Civil War in the 1860s, and then during the civil rights movement in the late 1950s and the 1960s, when the federal government sent in troops to enforce desegregation as the new law of the United States.

As a result, all Southern-born whites who stay in the South resent and dislike "Yankees." They developed a culture of resentment, which became even more open and strident as they moved into the Republican Party starting in 1968, and became the core of the racist and anti-immigrant white patriarchal nationalism that has been at the heart of that party since at least the 1980s. They exude the same white resentment the Southern rich displayed when they were the ultimate power brokers in the Democratic Party in the century after the Civil War. Back then, the Southern Democrats, who controlled Congress through bloc voting on race and labour issues, and on seniority. They made sure that any new legislation that helped lower-income whites in the North, such as the Social Security Act and the National Labor Relations Act, always included provisions that excluded African Americans from any new benefits and protections. This didn't bother most white Northern Democrats before 1965.

Like most social scientists of the era, I thought that some combination of increased inter-racial personal contacts, the country's highly individualistic values, and maybe even eventual class solidarity in the blue-collar class, would lead to a gradual decline in segregation. At first that seemed to be correct, but by the beginning of the twenty-first century I had concluded that the continuing segregation in all aspects of life meant that the United States still had strong remnants of the old caste system. I wrote in my most recent book, The Corporate Rich and the Power Elite in the Twentieth Century: How They Won, Why Liberals and Labor Lost (Domhoff 2020), that the United States is rather uniquely characterized by both a class system and a caste-like system, based on once having the largest, most powerful, and most lucrative slave system in history.

Immigrants always have faced discrimination when they first arrive in this country, but they left their home countries for a reason, and with hope, and they have the protection of their own language and culture until they eventually are accepted. The same thing is now happening for the more recent immigrants from Latin American and Asian countries as happened for white immigrants from Europe, and at the same pace, but the caste-like segregation of African Americans continues.

You were a psychology major in college, so could you tell us what psychology was all about back in those days? What were the main references, what was the kind of stuff you were interested in?

GWD: Well, psychology was a very divided field at the time. By far the dominant influence was "behaviourism", which was focused on studying how learning occurs, using animal models, namely rats and pigeons, with many of the results generalized to humans. I intensely disliked that kind of work, and didn't think it was worth much, so I avoided it as much as I could. I was interested in personality and motivation, and in social psychology. And that was a smaller part of psychology back then. And the whole field of what is now called cognitive psychology, which got started through the work of the Gestalt psychologists, who came to this country from Germany before World War II, and also the work of the Swiss psychologist, Jean Piaget, was not a very significant part of psychology at all. But despite the behaviouristic emphasis overall, I enjoyed several of my courses and I became a psychology major. And I had a couple of anthropology courses, and a couple of sociology courses, and eventually I decided to go to graduate school in psychology rather than become a journalist.

I earned an MA in the summer of 1959 at Kent State University in Ohio. I then went to the University of Miami in Florida and finished my PhD in the late summer of 1962. At this point the fact that I happened to be born in 1936 enters the picture as an important factor in my life, as one of those accidents of fate. It meant I was too young to serve in World War II or the Korean War, and too old for the Vietnam War by the time I had finished graduate school and that war came to involve a large number of American troops. Besides, I was married by then, and also had children, which in that era led to deferments from the military whatever your age. So, I never had to face the problem of going into the military and fighting in a war, and by the fall of 1962, at age 26, I had a faculty position, which virtually never happened for anyone who went to college after the mid-1960s, because everything changed due to the civil rights movement and the Vietnam War.

By comparison, these people only slightly younger than I had their lives interrupted by the Vietnam War, and they had to make tough choices. And due to the civil rights movement, many white college students outside the South, women and men, came to be activists and postpone finishing their college educations. Both the civil-rights and anti-war movements had a big influence on me, and soon thereafter, the feminist and gay-lesbian movements. I admired those activists, and I think that was one of the main reasons I moved into sociology. Furthermore, my experience with Southern whites gradually led me to fear that they would kill as many black people as they thought they had to in order to maintain white dominance, which was a chilling and sobering thought for me. I think my effort to understand that kind of mentality was another factor in why I eventually turned to the study of power within a sociological framework.

And how did you come to teach at California State College in Los Angeles? How long did you stay there?

GWD: Well, it was the best job I could find at the time in California, where my wife wanted to be, and I did, too, as I already said. I spent three enjoyable years at Cal State LA between 1962 and 1965, where I made many new friends on the faculty in political science and anthropology. They helped to open up a whole new world for me that very much fit with my interest in the civil rights movement, and then the anti-war movement and later social movements. As a result of my new interests, I was asked by students to speak at some of their rallies on campus, and in that way became very peripherally involved in the world of political activists. In addition, some of the students in my social psychology courses were activists.

And how did you encounter C. Wright Mills?

GWD: I began to read C. Wright Mills right around the time I was becoming interested in power because of the civil rights and anti-war movements. Mills died in 1962, shortly before I had begun to read his work, and I didn't know anything about his life or personality when I began my research. Reading The Power Elite had a big impact on me. I think it shows how the leaders of large institutions, including the owners and top-level managers in large corporations, are able to work together to maintain themselves as the dominant power group. It provides a starting point for figuring out how they shape the federal policies that have a big influence on the lives of everyday people in terms of their incomes, their job security, and their overall well-being. At the same time, I thought the book demonstrated very nicely that the members of the power elite are divided into two factions, as they still are today, the "sophisticated conservatives," whom I usually call the "moderate conservatives," and the old-fashioned, hidebound conservatives, whom I call the "ultraconservatives." And that division sometimes leads to policy conflicts among rival corporate leaders on a few issues. I think Mills showed that both cohesion and competition characterize the power elite.

However, I also want to emphasize that I also read every past criticism of Mills's The Power Elite that I could find. I wanted to have a better basis for making up my mind about all aspects of it when I started to do my own research. That gave me the basis for comparing rival hypotheses based on my own data. A few years later a political science friend of mine at Cal State LA, Hoyt Ballard, suggested that we put all the various criticisms of The Power Elite into an edited book, and write a commentary on both the critics and Mills. His idea led to a co-edited book in 1968, which I still think is very useful. It's called C. Wright Mills and the Power Elite. It also allowed us to critique all the different theorists of the day including Mills, and thereby establish our own perspective.

From the mid-1960s to this day, I have been involved in an inner dialogue in my thinking with Mills, as well as with the pluralists and Marxists he criticized, and who in turn were highly critical of his work. Based on my reading and research, I always disagreed with Mills's claim that the military was anywhere close to having equal standing in the power elite with the corporate rich, which is the very useful term Mills coined to stress that the upper class had been reorganized since at least the 1940s to include corporate executives as well as wealthy owners. And I came increasingly to disagree with his views on the "mass society" as I read more sociology. I think there are many power niches and potential bases for exercising power below the top. I mentioned those disagreements here and there in my work over the next several decades, but I spelled then out more fully in an invited retrospective review of The Power Elite written for Contemporary Sociology on the occasion of the book's fiftieth anniversary (Domhoff 2006), "Mills's The Power Elite 50 years later," which is most readily available on my website.

In my view, the problem is not lack of power bases below the top, but the disagreements among the people on whom these various power bases are built. Middle-level workers, for example, try to keep advantages over lower-level workers, and the established racial, ethnic, and religious groups attempt to hold on to the advantages they have over more recently arrived or previously subjugated racial, ethnic, and religious groups. In the specific case of the United States, there is a loosely knit liberal-labour coalition, as I've already noted in passing, and it is sometimes joined by progressive social movement activists. This loose-knit coalition was able to develop some degree of power, especially to hang on to unions from the mid-1930s through the mid-1970s, as part of the complex electoral coalition within the Democratic Party. Later it was able to stop the full cuts in Social Security that the corporate rich fought for in the 1980s, but those small victories were hardly enough to make the United States a "pluralist" society back then.

And then, as part of my further studies on power, I read E. Digby Baltzell's work (1958), in which I discovered that the upper class had an in-group telephone book, so to speak, where they list themselves. It was called the Social Register. It immediately hit me: that's something I could use as an indicator that a person is a member of the upper class or not. Building on listings in the Social Register, along with memberships in the exclusive social clubs listed in the Social Register, and attendance at the expensive private high schools listed in the Social Register, I realized I could then use a list of what I called "social indicators" to determine if there's a disproportionate number of these Social Register people that are in positions of power, which is to my mind a good indicator of power. Sure enough, my students and I then found that many upper-class people are on many different organizational boards, including corporations and foundations. All that was a revelation to me, coming as I did from the middle class, and having had no previous knowledge of exclusive clubs and expensive private schools. It was exciting, too, because I had been taught to be as empirical as possible, and to collect data, and I liked that approach.

At that point I also knew a little about social networks due to the British social anthropologists who studied family networks, and I also knew about the work of Jacob Moreno on sociometry (Moreno 1960). Then, when I read sociologist Floyd Hunter's 1957 book, Top Leader, USA, in about 1965 or 1966, I saw he used networks that were inspired by Moreno. Then I studied his list of leaders, which he had developed through his nationwide interviews with corporate leaders, by basically asking them "who are the big deals in this country?," an approach I will explain more about in a minute. When I studied the names on his list by looking at my social indicators, I found many of the same people that I claimed were members of the upper class. I was surprised that some of them were still important eight or nine years after his data gathering was completed. It gave me more confidence in my method and findings. Today we would say I had convergent evidence based on the use of two very different methods, Hunter's and mine.

By the 1960s, however, Hunter had little or no influence in sociology or political science. I learned he had been marginalized ,and even demonized, by the pluralists by the time I was working on these matters. They said that his interview method was merely learning about "reputations," so his work allegedly was of no use. When I coincidentally came to know Hunter personally in the 1970s, and I read his other work on power more carefully, I realized his work was even better than I initially thought. By the late 1970s his method had received strong vindication in a big interview study in which important American decision-makers were interviewed in a variety of issues, the best summary of which appeared in an article in the American Political Science Review in 1981, written by sociologists John Higley and Gwen Moore (1981), and entitled "Elite integration in the U. S. and Australia"

Since Mills had died in 1962 and Hunter had been marginalized by the 1960s, there was no ongoing tradition of power structure research at that time. The people doing power structure research were the activists working in social movements, who did useful pamphlets such as The Care and Feeding of Power Structures, which was used by civil-rights activists to figure out which corporations in the North were indirectly supporting segregation in the South. Soon activists also wrote Who Rules Columbia?, How Harvard Rules, and Go to School, Learn to Rule, the latter of which was about the ways rich elite Yale graduates involved themselves as leaders in the power structure. Perhaps Baltzell could be said to do power structure research, but his main concern in the 1960s was that the members of the upper class were not open enough to upwardly mobile Catholics and Jews, which would eventually weaken class power if that didn't change (Baltzell 1964). He didn't have a very deep analysis, in my opinion, although I did find his work on the social aspects of the upper class to be very important to my work at that time.

Anyhow, in the winter-spring semester of 1965, which turned out to be my last semester at Cal State LA, I went into my social psychology class and said something like "students are now saying they want to learn something new and more relevant, so we are going to study this power structure stuff; I tell you that at the outset so you can switch to a more typical psychology class if you want to." Most of them stayed and they did studies of foundations, or corporations, or banks, or whatever they wanted to study that related to the power structure. You can see their names in the endnotes in the first edition of Who Rules America?.

Was it at that time, when you moved from Los Angeles to Santa Cruz, that you switched from psychology to sociology?

GWD: Yes, although I did begin changing my interests while I was at Cal State LA, as I already mentioned. At that point, the University of California system, which has more money and a lighter teaching load than the state-college system, opened up a new campus in Santa Cruz in 1965, as well as new campuses in Irvine and San Diego. More generally, the University of California was hiring professors like crazy because it was expanding very rapidly in the face of the big "Baby Boomer" generation born shortly after World War II. To my good fortune, the Santa Cruz campus was run by people who were atypical; they were very liberal about education, and in general iconoclastic, and were breaking way from the standard university way of educating students at that time. For example, they didn't like the way in which big-time sports teams, like football and basketball, dominated the atmosphere on most university campuses. They didn't like the fraternity/sorority system either, and they didn't like how rigid and separated the academic departments were, and still are, on virtually all university campuses. They tried to break all that up.

The people doing the hiring at Santa Cruz were glad to have a young faculty member who was interested in both psychology and sociology, and who was working on an atypical topic, power, besides. They mostly succeeded in instituting their plans until about the year 2000, because the Santa Cruz campus did not have fraternities and sororities for decades, and still doesn't have any big-time athletic teams. The campus therefore has a very different atmosphere than most campuses. However, the departments slowly came back, and the campus has big, fairly rigid departments by now, in part because it went from a few thousand students to 18,000 students, and in part because a new, more conventional set of professors slowly replaced those of us who originally came to the campus.

Thanks to coming to Santa Cruz in the mid-1960s, I was able to become even more involved in studying power, and I had even more students who were interested in studying this topic, both in classes and as research assistants. They made it feasible for me to do the further work that made it possible to publish Who Rules America? in 1967, as well as doing a large amount of original research for my next book, a set of empirical essays entitled The Higher Circles (1970).

Your 1974 book on the Bohemian Grove (Domhoff 1974) deals with the cohesion question.

GWD: Ah yes, and glad you mentioned that book. The Bohemian Grove and Other Retreats: A Study of Ruling Class Cohesiveness, which is based on archival data, interviews, and a visit to the Bohemian Grove itself, was important to me as an answer to the pluralists of that long-ago time. They said, in effect, "Look, you are talking about these rich owners and top corporate executives as if they are cohesive and as if they get together and make policies, but you haven't shown us that." Contrary to their assertion, my study of the Bohemian Grove showed the rich and powerful were members of a handful of for-fun retreats and clubs, and thereby knew at least someone who knew someone else, so they were part of what is today called a small-world network. Furthermore, and this was the real reason I did so much empirical work on a seemingly frivolous side show like the Bohemian Grove, I had figured out through name-tracing that the corporate owners and executives in the network of retreats and social clubs were also were the trustees of foundations and think tanks, and also members of the elite policy-discussion groups where policy consensus is reached. I also could show by then, better than I had been able to in the previous ten years, that their policy statements help shape the overall policy dialogue, and that those statements are put in the hands of government decision-makers. Moreover, some of those decision-makers are corporate executives that have been appointed to government positions.

In addition, I drew upon the social psychology literature to show that social cohesion does in fact facilitate policy cohesion. Social cohesion causes people to like each other more, trust each other more, and listen to each other carefully in policy-discussion groups. Coming to know people better in a relaxed, high-status setting therefore facilitates policy cohesion, as social psychologist claim, and hats off to the great social psychologists who did that research decades and decades ago. In my mind, the book was also a replication of my earlier use of exclusive clubs and private schools as indicators of upper-class standing, since it used a very different starting point and ended up with the same findings as far as who ran the corporations and held top positions in government. I therefore used the last chapter of that book to bring together everything I had learned up to that point. Anyhow, I thought I had the pluralists dead to rights with my new evidence. And by that point the younger members of the upper class were losing interest in the Social Register, so the list of names grew shorter and it lost its usefulness as an upper-class indicator.

And it's not a minor point to add that I lucked into this study because I interviewed a very open and liberal person for my book on campaign finance, Fat Cats and Democrats (Domhoff 1972). He was a member of two elite clubs in San Francisco, and he was willing to give me the list of members for both of those clubs. That's what made the initial network analysis possible, and then to carry out my historical analysis, and then do my interviews. Then, too, the interviews gave me the opportunity to do observational analyses three different times at the Bohemian Club, which sponsors the Bohemian Grove retreat, asking my same seemingly dumb and innocent questions each time. I even had the chance to that that kind of thing once at the Bohemian Grove itself at one of their one-day events in June, an event that happens about a month before their annual two-week retreat to the Bohemian Grove that began in 1880 and still goes on to this day. Sometimes it's necessary to be very opportunistic in doing power structure research due to problems of access.

How do you define "power"?

GWD: Power is one of those words that's intuitively easy for most people to understand based on their own experience in groups and workplaces, but hard to define in a precise manner. We know it means "clout" or "juice" or "muscle," or being a "rainmaker," which is one of the in-group slang terms for power in the power elite. More academically and formally, I agree with those social scientists who define "power" as "the capacity of some persons to produce intended and foreseen effects on others"(Wrong 1995). This is a very general definition that allows for the many forms of power that can be changed from one to another, such as ideological power (which usually is based in religious organizations), economic power, military power, political power, and also intellectual power, by which I mean knowledge and expertise produced in universities and think tanks. That definition leaves open the question of whether "force" or "coercion" is always lurking somewhere in the background in the exercise of power, as many definitions wrongly imply, in my opinion. However, a formal definition does not explain how a concept is to be measured, or "operationalized," as we say in the trade. In the case of power, it is seldom possible to observe interactions that reveal power in operation even in a small group, let alone to see one "social class" producing "effects" on another. It is therefore necessary to develop what I call "indicators" of power.

Starting with the idea of power indicators, power can be thought of as an underlying trait or property of a social group or social class. So "power" is measured by indicators that bear a probabilistic relationship to it. This means that all the indicators do not necessarily appear each and every time power is manifesting itself. Research proceeds through a series of "if-then" statements: "if" a group or class is powerful, "then" it should be expected that certain indicators of this power will be present. Ideally, there should be more than one power indicator if possible, and they should be of different types so that any irrelevant components in them will cancel each other out.

What are the main power indicators you use?

GWD: There are three primary indicators of power, which can be summarized as (1) who benefits? (2) who governs? and (3) who wins?, all of which come from different types of research information. But it's also important to note that Hunter's indicator for power, a "reputation" for power is a good one, too. In fact, it usually overlaps well with the other three indicators, and sometimes uncovers a few leaders that aren't easily detected through public records. But the reputation indicator is more difficult to use because it requires access to people in the group being studied, or to people who follow policy-making closely, such as civil servants, social workers, or maybe best of all, the journalists who cover business or government. It also takes two or three rounds of interviewing. But as I said, it's been proven to be useful when it is feasible to employ the method, contrary to the pluralist critiques of it.

Returning to the three power indicators that are most frequently used, we know from anthropological and sociological studies that every society has experiences and material objects that are highly valued. So, if it is assumed that everyone in a given society would like to have as great a share of these experiences and material objects as they possibly can, then the distribution of values can be utilized as a power indicator. Those who benefit the most, by inference, are powerful. In American society, wealth and well-being are highly valued. People seek to own property, earn high incomes, to have interesting and safe jobs, and to live long and healthy lives. And there's much economic information that can be use of study these distributions. All of them show the same small upper class of owners and top managers at the top. Based on all the studies of the huge changes in the wealth and income distributions in the past 20 years, which everyone agrees are very solid studies, no one doubts this power indicator any longer. However, it's important to remember that the concentration of wealth and income in the top 0.5% was very high in the 1950s and 1960s, so not as much has changed as the social scientists who don't know the past literature may think. That assertion is demonstrated in detail in the second chapter of the first edition of Who Rules America?.

Then, too, power also can be inferred from studies of who occupies important institutional positions and takes part in important decision-making groups. That's the method I relied on the most in Who Rules American?. If a group or class is highly over-represented in relation to its proportion of the population, it can be inferred that the group is powerful. If, for example, a group makes up 1% of the population but has 30% of the seats in the main governing institutions, then it has 30 times more people in governing positions than would be expected by chance, and there is thus reason to believe that the group is a powerful one. Once again, that kind of information is readily available. Every study that has been done since the 1950s, including many by me and my students, has shown that the corporate rich have been appointed to these positions in disproportionate numbers ever since the 1940s. No one can any longer say, as the pluralists used to, that the Eisenhower Administration, which Mills focused on in The Power Elite, was somehow unusual.

In terms of the third power indicator, who wins in policy disputes, there have been many studies by now of policy issues over which the corporate rich, the liberal-labour coalition, and others groups disagree. In the United States at least, there always have been different policies suggested by opposing groups on such important issues as foreign policy, taxation, welfare, and the environment. This indicator, which focuses on who successfully initiates, modifies, or vetoes policy alternatives, comes closest to approximating the process of power that is contained in the formal definition. But it has to be stressed that it is no less an inference to say that who wins on issues is an indicator of "power" than to do so on the basis of value distributions and positional over-representation. The problem with the who-wins indicator is that decision-makers are rarely willing to let social scientists sit in on power discussions, so reconstructing a decision-making process is based on after-the-fact interviews, and if the researcher is lucky, through access to documents that were part of the final process. Even then, it may be difficult to gain access to decision-makers to interview them, some interviewees may exaggerate or play down their roles, and people's memories about who did what often become cloudy shortly after the event. Contrary to the pluralists, it is not the ideal power indicator.

With all those cautions kept in mind, the fact is that a large accumulation of decisional studies over the past 100+ years shows that the corporate rich win on the issues of most concern to them virtually every time. I summarized the new studies that make this point in later editions of Who Rules America? In The Corporate Rich and the Power Elite in the Twentieth Century, I showed corporate dominance for a whole range of big issues relating to unions, government social benefits such as old-age pensions and health insurance, and trade expansion overseas for the entire twentieth century. Since the who-wins indicator is the only one pluralists think is fully legitimate, I frankly don't see how anyone can claim to be a pluralist any more, any more than Marxism holds up as a theory in the face of all the evidence that does not support the key aspects of the theory. I think the 60+ years of research since Mills wrote The Power Elite, and the 50+ years that have passed since the first edition of Who Rules America? appeared, completely supports the common aspect of our views, which is that power is based on control of the major institutional structures of a society, and in the United States that means the big banks, corporations, and agri-businesses.

I fully recognize that most social scientists probably still don't agree with my conclusion, but I also know that fewer and fewer sociologists focus on big issues like societal power, and that the political scientists who study power still focus mostly on government, and entirely ignore what Mills and I have written. According to most political scientists, power was concentrated until the Progressive Era came along and changed some things, and then power calcified at the top again until the Great Depression disrupted the established order. After that, the political scientists continue, power supposedly didn't start to become highly concentrated again until the late 1970s, and then became highly concentrated in the past 20 years. So, even if some political scientists are familiar with the work Mills and I did, they figure we were wrong for the era we wrote about.

So, what is power structure research, exactly, and how can power structures be studied empirically?

GWD : In today's terms, it is first of all an attempt to construct the networks of people and institutions that run the show, although it must immediately be added that the overall theory concerns more than the top levels of society. Power relations run from top to bottom, like a pecking order, only more complicated. The social structure is not simply an organized elite or class at the top, with an unorganized mass below it, as I noted in stating my disagreements with Mills in my answer to a previous question. Nor, for that matter, is there a more or less unified working class below the top levels, as Marxists claim is at least potentially the case. As my critique implies, the power structure is more like a pecking order, and in that regard there is an old German folk saying from two or three centuries ago that captures what it's like to live in a power structure. "Life is like a chicken coop ladder," which I read in a book by a great folklore professor at the University of California, Berkeley, Alan Dundes, who uses that folk wisdom as the book's title (Dundes 1984). That is, all the bad stuff goes raining down the social ladder, and eventually really piles up on the people on the bottom. The only saving grace that social scientists can add is that there are multiple status hierarchies in human societies, which means that people can find enjoyment and solace outside the power hierarchy through being a great athlete, or a fine musician, or a popular entertainer.

From an empirical point of view, how do you study the power structure?

GWD : The empirical study of power begins with a search for connections among the people and organizations that are thought to constitute the powerful group or class. This procedure is called membership network analysis. It starts with a study of people and all the organizations to which they belong. Or conversely, you could say the study starts with a list of organizations that also includes all of the members of each organization. Either starting point leads to the same results. These results are usually presented in the form of a matrix. The people are listed from the top to bottom and the organizations are arrayed from left to right. The "cells" or boxes created by the intersection of a person and organization are filled with "relational" information such as "member," "director," "owner," or "financial donor." The attitudes a person has toward any given organization or person in the matrix also can be included, such as "supporter" or "opponent." Maybe think of attitudinal information as a "psychological" relation to an organization or person. The information used in filling the cells of the matrix is obtained in a variety of ways, which mostly means written documents, surveys, and interviews.

Large and complicated membership networks can be analysed using computer software based on sophisticated mathematical techniques, such as graph theory and matrix algebra. Once the membership networks have been established, there are many other types of links that might be analysed, such as money flows among organizations. After the most central organizations and people in the one or more cliques, factions, groups, or classes in the network are established, then it is necessary to study the "output" of those organizations and people, such as the policy and public relations statements published and disseminated by the organizations, which reveal the shared positions within that clique or the overall network. It's equally important to examine the speeches and interviews by the most central people in the network that are available in text form, which provide an excellent indication of what the main leaders are thinking, or at least saying. Here I am talking about a long-established methodology in the social sciences and communication studies. It's called "content analysis," and it can be done with quantitative analyses, including examining word networks. These methods are just as rigorous as those used in doing membership network analyses.

At that point, it's time to interview spokespersons for the organizations, or the top leaders to whom you can gain access. You then ask all your warm-up questions, the easy questions, so to speak, and then you ask increasingly difficult questions. Finally, you ask the biggest questions on your mind, which may have been refined for you by hearing their earlier answers. It's at that point that the interviewee may kick you out of the office. I call it the onion theory of interviewing, in which you start on the outside, and gradually peel back layers, until you reach the core. Here I think of the great interview study of upper-class women carried out by sociologist Susan Ostrander for her book on Women of the Upper Class (Ostrander 1984). Using the grounded-theory methodology that has proven to be highly useful, Ostrander started with the very general question of what the woman being interviewed did during her typical day. It turned out to be a very important opening question because she unexpectedly learned that the role of "volunteer" was very important for upper-class women, at least in the first 80 years of the twentieth century. Toward the end of the interview she asked if they considered themselves to be a member of the upper social class, which usually received an awkward or chilly answer, and often put an end to the interview.

In another good example of the importance of interviewing in power structure research, a sociologist who was studying campaign finance in the late 1980s, an indefatigable and relentless researcher named Dan Clawson, asked me what I'd infer from his complicated findings on corporate campaign-finance donations to members of Congress. He asked this question based on a few puzzling patterns he found through a very fine network analysis, using money flows as the network links. I told him I didn't know, but that I'd sure ask the executives who ran political action committees (PACs) for the corporations he and his co-author graduate students were studying. And based on those questions that emerged from the network analysis, the executives told them great stuff that solved many of the mysteries, which can be found in their 1992 book, Money Talks: Corporate PACS and Political Influence.

For me, it was my experience doing door-to-door poll interviews during that summer between my junior and senior years in college, along with my journalism experience along with reassurances from Hunter that I could gain more access than I figured was possible, that got me into interviewing. I talked to big donors to political campaigns, the experts that worked for think tanks and policy-discussion groups, and the corporate bigwigs who are leaders in policy-discussion groups. And I of course interviewed employees and members of the Bohemian Club. In the process, one of those employees practically handed me the opening paragraphs of my book when he told me that every encampment began with a hookey, cornball initiation ceremony called "The Cremation of Care," complete with high priests, druids, tree spirits, and the mock sacrifice of the effigy of Mr. Dull Care. But I had to smoke a joint with the nervous employee before he'd talked to me. As a person who never smoked much marijuana, I ended up stoned for most of the interview, but with no regrets, just embarrassment.

As far as I am concerned, the work by Ostrander, Clawson, me, and other power structure researchers, who had our heyday from the 1970s to the early 1990s, demonstrated that interviewing powerful people is possible and useful. It takes time, and it can be disconcerting to sit in big intimidating offices, and sometimes there are refusals, but it's worth the effort. By now, though, this kind of work appears to be outsourced to the excellent researchers at a wide range of non-profit public-interest organizations, as well as to investigative journalists, who together provide valuable grist for various social-science mills to grind on.

Wrapping up on the issue of how power structures are studied, if the networks that are constructed and analysed on the basis of documents and interviews are indeed the relevant networks, then they should score high on one or more of the four main power indicators that it was possible to use in any given study: who benefits, who governs, who wins, and who has a reputation for power.

Any final thoughts?

GWD: For a long time now, it seems to me that power structure researchers in fact have been finding the right networks, and the scores on the power indicators for the organizations and leaders in these networks are higher and higher every time there is a large-scale study. Our findings and theories have turned out to be more accurate, or at least less wrong, than any of the alternative theories that have been proposed since Mills, Hunter, and I started doing power structure research in the 1950s and 1960s.


Aron, R. (1960). Classe sociale, classe politique, classe dirigeante. Archives européennes de sociologie, 1(2), 260-282.

Digby Baltzell, E. (1964). The Protestant Establishment: Aristocracy and Caste in America. New York: Random House.

Digby Baltzell, E. (1958). Philadelphia Gentlemen. The making of a national upper class. Glencoe, Ill.: Free Press.

Domhoff, G. W. (1967. 1983, 1998, 2014). Who Rules America? New York: McGraw-Hill/

Domhoff, G. W. (1968). "The Power Elite and Its Critics." Pp. 251-78 in C. Wright Mills and the Power Elite, edited by G. W. Domhoff and H. B. Ballard. Boston: Beacon Press.

Domhoff, G. W. (1970). The Higher Circles. New York: Random House.

Domhoff, G. W. (1972). Fat Cats and Democrats: The Role of the Big Rich in the Party of the Common Man. Englewood Cliffs, NJ: Prentice-Hall.

Domhoff, G. W. (1974). The Bohemian Grove and Other Retreats: A Study in Ruling-Class Cohesiveness. New York: Harper & Row.

Domhoff, G. W. (2006). "Mills's the Power Elite 50 Years Later." Contemporary Sociology 35(6):547-50.

Domhoff, G. W. (2020). The Corporate Rich and the Power Elite in the Twentieth Century: How They Won, Why Liberals and Labor Lost. New York: Routledge.

Domhoff, G. W. and H. B. Ballard, eds. 1968. C. Wright Mills and the Power Elite. Boston: Beacon Press.

Dundes, Alan. 1984. Life Is Like a Chicken Coop Ladder: A Portrait of German Culture through Folklore. New York: Columbia University Press.

Higley, John and Gwen Moore. 1981. "Elite Integration in the U. S. And Australia." American Political Science Review 75:581-97.

Hunter, F. (1953). Community Power Structure. A Study of Decision Makers. Chapel Hill: The University of North Carolina Press.

Mills, C. W. (1956). The Power Elite. New York: Oxford University Press.

Moreno, J. (1960). The Sociometry Reader. Boston: Beacon House.

Ostrander, Susan A. (1984). Women of the Upper Class. Philadelphia: Temple University Press.

Wrong, D. (1995). Power: Its Forms, Bases, and Uses. 2nd ed. New Brunswick: Transaction Publishers.


[1] Centre National de la Recherche Scientifique, CESSP-CSE, EHESS, Paris

[2] Université Paris-Dauphine, IRISSO, Paris

[3] École des Hautes Etudes en Sciences Sociales, CESSP-CSE, Paris, France