Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS (politics.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 28 April 2017

Social and Political Power

Summary and Keywords

Power is a complex topic that is viewed in entirely different ways by different writers. Power can be seen as a property of agents, with some agents having more power than others. It can be seen as a property of social systems, where structures hold power. It can also be seen in terms of specific actions by people to coerce or dominate, or it can be regarded as a subliminal force that leads people to think and behave in one way rather than another. It can be analyzed descriptively to try to explain how it is distributed, and critically to argue for changing structures to provide a more egalitarian and fairer distribution.

Power studies flourished in the great community power studies of the 1950s and 1960s. Some of these works suggested that democratic nations were controlled by powerful elites who ruled in their own interests; some that power was more widely distributed and elites could not simply rule for themselves; others that in capitalist societies, despite some counterexamples, elites generally ruled in favor of developers and capitalists. Later studies examined how people’s interests are defined in terms of the structural positions in which they find themselves, and how the very ways in which we think and express ourselves affect our individual powers.

Keywords: Agency, collective action, domination, hegemony, interests, luck, power, structure

Introduction

Political or social power—or simply “power”—has long been a key concept in political science and international relations. Both theoretical and empirical work flowered in the 1960s and 1970s, but then tailed off for 20 years or so. More recently, interest has re-emerged, with a host of new books and articles and a research journal dedicated to the study of social and political power (http://www.tandfonline.com/loi/rpow21#.VrpiefFH8n8).

Power is such a massive topic that not everything written about it can be covered. The first part will concentrate upon the more empirically minded approaches that attempt to suggest ways in which we can identify power in society. When the early attempts to measure power tailed off due to methodological and theoretical critique, power studies tended to become more esoteric with debate tending to verbal rather than substantial dispute. Examining the early work on power and investigate why that empirical interest declined, before considering new departures. These departures can be seen in the influential work that suggested there were multiple dimensions of power. However, we can see that understanding collective action problems, misperception and unintended consequences of rational action can lead to outcomes that are otherwise attributed to some mysterious aspect of power. A major divide is between those who defined power in terms of agency and those who see it structurally but we see the two are not necessarily in conflict. Some make distinctions between political and social power, but there is no reason why the underlying concept of power should differ across the domains to which it is applied, and there is no need to get involved in verbal dispute about what is political and what is social.

The Community Power Debate

Power was a mainstream and much-discussed topic in political science from the 1950s through the 1970s. Political sociologists, such as C. Wright Mills (1956), saw elites controlling the major elements of the state within the military, within industry, and at the highest levels of politics in Washington. Other researchers studied local communities in towns and cities, and again saw the controlling powers within local industry and politics as largely determined by social and educational backgrounds (Hunter, 1953). The major method for identifying the power elite was reputation. People in positions of authority were interviewed and asked to name powerful figures; these figures were then interviewed and again asked to identify important and powerful people. By such triangulation the power elite was identified.

Critics contended that there is a difference between such a reputation and actual power (Wolfinger, 1960). They also argued that there is a difference between an elite governing in its own interests and an elite governing under the constraints of the democratic scrutiny of elections and pressure-group activity. These “behavioralists” or “decisionists” chose a different methodology. In the famous New Haven community power study directed by Robert Dahl (1961a, 1986), certain key issues were identified from newspaper reports; the researchers then followed the course of decisions made to see who influenced the final outcome. Dahl and his team concluded that power was more dispersed than the elitists claimed. The pluralist conclusion was that different elites tended to dominate in different issue areas. An elite may exist, but it did not simply rule in its own interests; rather, it was subject to influence from others.

Dahl identified two sorts of people: those engaged in politics (homo politicus) and those not so engaged (homo civicus). He argued that the potential power of the homo civicus kept the homo politicus in check. The former were not simply politicians and elites, but also included people who worked in organizations exerting pressure and those who voted. Dahl’s simple and important definition of power can be paraphrased as “[a] has power over B to the extent that A can get B to do something B would otherwise not do.” Accordingly, if a member of the public has even a marginal effect on some decision made by a politician, then that member of the public has exercised some power. That effect can also be potential: that is, B acts in anticipation of A’s actions.

Much has been made of the different methodologies; but actually the elitists and the pluralists all consulted archives, read newspapers, and conducted interviews using formal and informal contacts (Ricci, 1971, pp. 128–129). The difference, to a large extent, was that the elitists were more interested in identifying powerful people, whereas the pluralists focused on the breadth of influence, no matter how marginal.

The pluralists were heavily criticized on a number of grounds. The issues they identified were all the ones discussed in the press, but they ignored certain factors within New Haven. For example, they never discussed the tax-exempt status of Yale University (where they were based). Race relations did not figure in their work, though clearly tensions existed as race riots occurred soon after Dahl’s book was published (Ricci, 1971, p. 158). Later analyses of the evidence, carefully kept for replication purposes, offered a different interpretation, noting that New Haven’s politicians and local businesspeople tended to come from the same schools and universities—suggesting that New Haven did indeed have its own power elite (Domhoff, 1978/2010). William Domhoff has continued in the Mills tradition, tracking individuals and their networks of power through successive editions of books on the United States’ elite (Domhoff, 2013a, 2013b).

Community studies inspired by neo-Marxist analysis of capital relations (O’Conner, 1973; Castells, 1977, 1978) produced a set of power arguments suggesting that capital drives local growth. In the growth machine model, cities and communities compete with each other to attract developers through tax incentives and zoning decisions. Growth is seen as good for the community and electorally popular, though it often benefits the developers and local capitalists more than the local citizenry (Logan & Molotch, 1984, 1987; Domhoff, 1986). With different institutional forms, such growth coalitions might vary in shape, even where investment is state-led (Bassett & Harloe, 1990; Harding, 1991, 1994; Bassett, 1996), and are not always successful (DeLeon, 1992; Clark & Goetz, 1994; Orr & Stoker, 1994).

Clarence Stone (1989)—inspired by Floyd Hunter and studying the same city: Atlanta, Georgia—extended such accounts, creating “regime theory” to explain why mayors with very different backgrounds and electoral bases end up pursuing similar growth strategies. Regime theory in Stone’s hands is a nonformal model of coalitions that form around politicians to buy electoral success while defending development. Later work suggests that different types of regimes might exist with different electoral and coalitional bases (Elkin, 1987; Savitch & Thomas, 1991; Fainstein & Fainstein, 1986; Dowding et al., 1999), though the extensions tend to lose sight of the explanatory basis of regime theory (Dowding, 2001; Mossberger & Stoker, 2001)—that is, the coalitions forming around regimes do so to maintain political success. Regime theory is essentially a coalition theory, and its explanatory function works by suggesting how politicians gain electorally by giving rewards or opportunities to those outside their electoral coalition; it is not simply a label applied to specific policies pursued by local politicians. It is not an account of power at the local level if it is only about electoral coalitions for temporary gain.

While there have been more recent community power studies, in general the approach ran out of steam, in part because political science looked elsewhere, but also, as described later in this article, because those interested in power moved in directions that made empirical work more problematic. Empirical analysis could not go much further, apart from applying the same arguments and interpretations—occasionally new interpretations—to different cases. We see this, for example, in Flyvbjerg (1998), whose detailed study of development in Aalborg juxtaposes the realpolitik of power against utilitarian rationality. What is different in his study are the theories marshaled to comment on the empirics. They are not really explanatory but normative, juxtaposing power and rationality. Flyvbjerg (2001) develops this idea, asserting that claiming social science is explanatory in the manner of science is insincere; rather, it needs to follow the wisdom of phronesis, requiring the ability to advise on practical virtue. That is, social science should be about how to change the world and not just about describing and explaining it. Community power studies dead-ended because they lost sight of any way of empirically distinguishing between theoretical accounts of power.

The fundamental disagreement between those who thought power was bundled into the hands of a few people with similar backgrounds, clubs, and networks and those who thought power was more widely dispersed was theoretical. Radical critics even went so far as to say that power could not be studied empirically because it was so diffuse. Steven Lukes’s (1974) book, whose brevity belies its influence, played a significant part in that claim, though Lukes himself was adamant that power could be empirically analyzed.

Four Dimensions of Power

The analytic criticisms of the decisionists are perhaps more important in an account of what power is than the different approaches to empirical research. Bachrach and Baratz (1962, 1970) pointed out that there are nondecisions that the methods of the pluralists ignore. In rejecting Dahl’s account of power, they started to look at the subliminal effects of institutions and processes. They suggest that force and harassment, the threat of sanctions, prevailing norms, as well as the reinforcing of those norms or new norms to squash incipient demands and conflict would not be found by the decision methods. This began the “radical” critique of community power studies that, with a few exceptions, ended their popularity for a generation.

Lukes (1974) introduced the concept of the three dimensions of power. He saw decisionist analysis as the first dimension of power; Bachrach and Baratz’s nondecisions as the second; and the fact that society is structured in such a way that people often do not know their own interests as the third. The point of this analysis is that while it might appear that people are not subject to power in the first and second dimensions, they are still subject to power in the third dimension. Lukes argued that each dimension came with its own methodological and normative assumptions. Each reveals something about the power structure, but we need all three to truly unveil the nature of power in society. His account of the third dimension proved the most controversial. Many objected to the claim that people did not always know their own interests. It is not so much that it could never be true; clearly, lacking or having biased, incomplete, or simply incorrect information, people can be mistaken about what is good for them. Rather, how and the extent to which people can be mistaken about their own interests is contested. Lukes suggested that individuals could be systematically mistaken about their own best interests.

The fourth dimension of power, first mentioned almost ironically in seminars and conferences, is criticized in Lukes’s two added essays that more than tripled the length of his book in its second edition (Lukes, 2005). It is seen in Foucault’s conceptualization of power as a prevailing network of relations and discourses that exists in our entire awareness and from which there can be no escape for the dominant and the subordinate alike (Hardy & Leiba-O’Sullivan, 1998).

How far this fourth dimension constitutes or is already part of the third might be questioned. The problem with such an extensive view of a fourth dimension is that, if all relationships are power ones, and our awareness and interests are fully determined, then power loses any explanatory and critical edge. This Foucaultian trap needs to be avoided (Dowding, 2006; Clegg et al., 2006, pp. 217–220). Lukes himself argues that the third dimension entails attributable agency, whereas Foucault’s notion does not (Hayward & Lukes, 2008). Indeed, most of those who follow Foucault critique the fully deterministic interpretation, believing that there must be some room for us to reflect back upon subjugation, even if our habitual thought processes are complicit. The important idea in the third/fourth dimension is that the unthinking manner in which people often engage in various social relationships, customs, and habits can lead people to act in ways that might be part of their own subordination. Feminists have used such Foucaultian ideas in powerful ways (Allen, 1996, 1999).

So, for example, Bartky (1990) critiques and extends Foucault’s (1977) discussion of how discipline and punishment make people docile into the realm of women using disciplinary practices of dieting, shaving, wearing make-up, and engaging in socially suitable modest behavior. The third or fourth dimension is implicated because these practices are (at least partially) self-inflicted and self-regulated. Of course, such analysis should not imply that the other dimensions of power do not also operate in more obvious ways of coercing and harassing women. Foucault also extends this account to suggest that the very means by which people regulate their lives also create them. Subjects are made by their subjection. Judith Butler (1993, 1997) in her influential work offers this as a paradox: that the power of the system is not simply something to be opposed but is also what creates us as we are.

We can see that the account of power offered by these authors is highly contested. Some want to analyze power relations in society in order to explain the way the world is; others see the role of the analyst more in terms of demonstrating domination and exploitation: they are less interested in empirical analysis than in critique. Thus, the three or four dimensions of power reveal a number of fault lines in the conceptualization and analysis of power, which include:

  • Normative versus positive: Is power essentially a normative concept? Is it an essentially contested concept?

  • Theoretical versus normative: To what extent is power an empirical or entirely theoretical concept? Can social or political power be measured?

  • Agency versus structure: Is power a property of agents or systems, or both? What is the relationship between them?

  • Dispositional versus exercise: Is power best seen in dispositional terms, or should we concentrate on power as it is exercised?

  • Power over versus power: Should power analysis only involve conflict? Is power zero sum? Should we see power as “power over,” “power to,” and/or “power with”?

  • Is power an overriding concept, or is it one of a set? What is the relationship of power to other concepts, such as domination, coercion, hegemony, ideology, freedom, or free will?

  • Static versus dynamic analyses: Can we have a dynamic analysis of power?

Since these issues partly morph into one another, the first four are examined under the heading of methodological issues, and the remaining three are considered in the section entitled Conceptual Analysis and Collective Action and Power. As will be revealed, understanding the dynamics of collective action can serve to bring together what appear to be very diverse accounts of power.

Methodological Issues

Normative versus Positive

Positive or empirical analysis is usually concerned with describing and explaining the world. Normative analysis is more concerned with critique and progressive ideas about how to change the world. Some writers are explicitly normative and critical in their writings on power, whereas others are more determinatively empirical. In the face of plural values—people have different views about how the world is—and if power is largely normative, it is often argued that concepts like social power are essentially contested. This claim is separate from any radically relativist claim that there is no social reality, since if there is no reality to which we all refer, there can be no argument about the concepts referring to it. Rather, the claim is that there is some social reality, but how we describe it affects our attitude toward it. If we can authoritatively define power in such a way that all are equally subject to its vagaries, then we might claim that there is no one who need accept blame for any injustice.

One response to this claim is the subscript strategy (Chalmers, 2011), where different uses of terms can be subscripted by different stipulative definitions. One can then avoid the most obvious problems with different interpretations of concepts. Subscripting different usages makes debate clumsy, but both the empirical situation and the normative understanding of each person can be represented; the subscript strategy can help reveal the underlying conflict of value disagreement. Here value pluralism is maintained, but so is conceptual agreement, since all can agree on the subscripted versions of the term. Dowding follows this strategy for the term “interests” (Dowding, 1991, ch. 8) and for “power” (Dowding, 2003). One could see Lukes’s three dimensions in terms of the subscript strategy, along with an argument that some subscripted versions of power are superior to others.

Theoretical versus Normative

Can we measure social or political power? If power is some diffuse process that exists within social relations, it might be unmeasurable. If it is seen as a property of agents, then it is measurable, at least by their relative resources. So the view one takes on the agent–structure fault line might determine one’s attitude to the measurement issue.

One of the issues often ignored in discussions of Dahl’s account of power is that, as an “operationalist,” he distinguishes between “pure definitions” and “operational definitions”—the latter being those for which we can collect data (Dahl, 1968). As empirical political scientists, we need our basic concepts to be measurable. However, the danger of operationalism is that we measure one thing and then claim that we have measured something else. Baldwin (2015) defends Dahl’s New Haven study on the grounds that, at best, it only looks at measurable or operational aspects of influence and not power. However, the conclusions that Dahl (and more particularly Wolfinger) draw seem to amount to a rather narrow account of the nature of power relations in New Haven (Lukes, 2015). Critics could justifiably claim that some of the most important aspects of power (or indeed influence) have been missed, especially given that it is a critique of elitism.

Morriss’s (2002, esp. pp. 14–19) account of the vehicle fallacy can be seen as a critique of operationalism. Essentially, he claims that judging the power of an agent by her resources mistakes one thing—the grounds of her power—for another—her actual power. The resources might be the vehicle of power but are not power itself. The truth in that argument is that the power underlain by the resources of agents is always relative to the resources of other agents and how each uses those resources. Resources might be used wisely or poorly. In conflict situations, the powerful should generally defeat the less powerful, but we know from sporting contests as well as military battles and social conflicts that luck or bad tactics can cause the weaker side to prevail. However, over many conflicts, the stronger side should prevail in correlation with its greater overall resources. The vehicle fallacy is itself fallacious; it only shows that we should not confuse comparative statics of sets of actors with the dynamic analysis of particular power conflicts (Dowding, 2008). On this argument, power can be measured, if only indirectly, by the resources that agents command.

For example, if we wanted to compare the power of two political actors—say the power of presidents in two different political systems—we might look at the institutional resources they have, as presidents, in each system. We might say that the president in country X has more power than the president in country Y, since the former has the power to dismiss the prime minister, choose the cabinet, dissolve parliament, and unilaterally give herself emergency powers, whereas the latter lacks all of those institutional resources. So, generally speaking, presidents in country X have more power than presidents in country Y. It might be the case, however, that a particular president in X is weak: he is close to facing an election that he is bound to lose, and he has little support in parliament and the country. Meanwhile, the president of Y enjoys complete parliamentary and popular support. Using the resources that are available allows comparative statics power measurement, but resources are multifaceted.

The most explicit general account of measuring agential power in this comparative static manner is Dowding (1991). Following Harsanyi (1962a, 1962b; reprinted 1976), who formalizes Dahl’s (1957) simple definition, Dowding breaks these resources down into four major categories: information, legitimate authority, unconditional incentives, and conditional incentives. However, the categories themselves are theoretically general; more important for actual comparison are the resources for which we can actually gain data. Like much of the literature, Dowding’s comparative static approach suffers from the dynamism problem. It is all very well collecting data on resources, looking to make general comparisons across political or systems or across agents within such systems, but in the process of actual conflict, those general comparisons might be misleading. In other words, we can compare the relative resources of agents across communities to calculate power in terms of resource holding, but how those resources are used in specific power struggles is another factor.

There are other aspects of power that are more diffuse. To cover them, Dowding (1991) adds a fifth resource, the notion of reputation. Sometimes agents can be more powerful than their resources apparently justify because they have a reputation for being powerful. Moreover, agents can deliberately behave in ways that might seem irrational in order to gain a reputation that gives them more power in conflict situations. If I can make it appear that I do not care what costs I will have to bear to gain some reward, my opponents might decide to avoid conflict with me, since they do not want to bear the costs I will inflict on them, even as they defeat me. Dowding (2016, ch. 9) extends this analysis, suggesting that perceptions of power become actual power as the perceptions become part of the belief set of each person in a power game.

Agency versus Structure

Measures of power in terms of the resources that actors can command is obviously an agential account of power. It will not satisfy those who think of power as being in some way a property of systems. I suggest that this is a mistake. Examining the resources of agents within a system enables us to grasp at least one element of systemic power. It might not encompass all the components of what theorists think of as systemic or structural power, but perhaps we can capture those components in other ways.

The structure of a society is the relationship that agents bear to one another. Take the same people: give them different resources, place them in a different relationship to each other, and one has two differently structured societies. The same individuals could exist in an advanced capitalist society or in a centrally planned socialist society and their relationships would be very different. Or three societies with completely different conventions, norms, and moral beliefs could exist and the same people could lead very different lives. So the same men and women could reside together in a society dominated politically, economically, and socially by men; in one where women dominate politically, economically, and socially; or in one where any system of domination is not based on gender lines. The structure is composed of the relationships. At least some—perhaps not all—of those relationships can be captured by the resources the agents command.

Seeing power as systemic or structural, however, can mean two different things. It can simply mean that the description of the power structure reveals systematically different sets of power holdings across groups of people. However, it can also mean that structures themselves hold and wield power (which is one reading of Foucault), making power very different from its agential form. When discussing the measurement of power in terms of agential resources, we saw that those resources can be wielded adroitly or poorly. Agents can also choose whether or not to deploy the resources they hold. Structures can do neither; they are not acting agents. Morriss (2002) has an extended account of why “power” must be dispositional in nature and demonstrates the exercise fallacy committed by analyses that ignore that aspect. Dowding (1991) likewise suggests that the term “power” conceptually is dispositional and has a necessary relationship to causation. Structures are usually seen as the background conditions of causal forces, particularly where causation is seen in terms of intervention (Woodward, 2003). This suggests that power is best ascribed to agents and not structures, since structures cannot intervene. Second, Woodward (2003) argues that the term “power” applied to structures in this manner is redundant, since any description of the relationship between agents can be given without using the term “power.”

In theoretical applications and in comparative static analysis that attempts to examine causal mechanisms or describe structures across systems, “agents” are defined by their functional roles. These functional roles are already structured. In other words, agents are “firms,” “prime ministers,” “finance capitalists,” “voters,” “revolutionaries,” “sets of people who share a common interest,” and so on. A type is the set of its token examples. When we tell an actual history or examine a particular case study, we tell the story in terms of tokens. When we examine the world in more general terms, we are examining types. These are very different forms of explanation (Dowding, 2017, pp. 50–55). In most general analyses of the power structure, agents are types and not tokens—they are already theorized. Dowding (1991) argues that this feature of agential accounts demonstrates that there is no major divide between so-called methodological individualism and holism or structuralism. Although his analysis is agent-based, the agents are already structured; structures are built into the analysis. List and Spiekermann (2013) run a sophisticated version of this argument, using a historical case to illustrate that, even when we try to explain a token case, the structural variables are the key explanatory ones.

Dowding’s aim with regard to the community power debate was to separate two different issues: the epistemological stance of a particular theory and the problem of distribution of power in society. Agential theory is associated with pluralism and broadly just distributions; structural accounts are associated with elite theory, domination, and broadly unjust distributions. Dowding tries to describe a view of power that is agential, yet illuminates structural features of society, and so connects up macro-analysis of power distribution with individual accounts of freedom and responsibility. Systemic or structural power, used in many analyses (Stone, 1980, 1993; Winner, 1980)—notably stratification theory (Hunter, 1953; Stone, 1988)—can be handled in this way. The missing element is how structure is implicated in the essential fabric of agents—that is, how each of us is molded by the environment around us, which helps to determine both how we see ourselves as subjects and how we view our interests. It helps create our moral viewpoint. We saw this type of analysis in the discussion of Foucault and how subjects can be implicated in their very subjection. But this aspect of structuration goes beyond mere issues about power in society right into the heart of issues of free will and agency. If we see the relationship between people as the structure, how those relationships penetrate into the very way in which people react to each other can be called the deep structure.

One way of thinking about this issue is the following. Think of a structure in terms of the analogy with the relationships of inanimate objects. A redbrick house of a given form will be different from a house made of breezeblocks of the same shape—they might have different thermal properties, one might better withstand earth tremors, and so on. This is also true of two houses made of the same material but using a different design. Typically, the individual bricks or blocks will not in themselves alter, even if the structure responds differently to external conditions. However, human agents might respond differently, given the way society is structured. In an egalitarian society, some people might behave more responsibly than others; within hierarchical systems, some might be more likely to abuse their positions. We expect structures to affect human agents to a greater degree than material objects such as bricks.

How we act is determined by our history and its context; hence, social relations to some extent construct human agents. Or, perhaps better, we construct ourselves through our social relations. It is claimed that the authoritarian personality is established by early upbringing (Adorno et al., 1950; Altemeyer, 1996; see Martin, 2001 for a good review and critique). Family relationships are thus important, but so are other social relationships; and the organization of society can affect many aspects of agents.

This is not only so for biological agents. The predictions we make for agents defined by their roles will be strongly determined by the specific incentives provided by institutional structure. We expect governments in Australia, which have to face the electorate every three years, to demonstrate greater myopia in their policy prescriptions than UK governments, where the electorate goes to the polls only every five years. We should expect finance capital to be more risk averse if the rewards for taking risks are lower or the penalties for failure are higher. Crime rates should be determined not only by the rewards of crime and the penalties if caught, but also by the probabilities of being caught. Furthermore, what are considered criminal or acceptable acts will depend on the general behavior of the population, and so on.

Much of political science is concerned with structure in the former sense. It takes the agent at face value and considers how formal institutions affect behavior. That constitutes the essence of the behavioral revolution (Dahl, 1961b) and also, strangely enough, the nature of “new institutionalism” (Dowding, 1994). This makes sense, since behavioralist inferences are made in terms of comparing across institutions; while institutionalists argue that institutions affect behavior. Critics of mainstream political science believe that it ignores deep structure. And it does. But the problem is how to empirically examine deep structure, since it involves counterfactual inferences about the interplay not simply of institutions and behavior, but of institutions and dispositions of behavior. But we normally infer dispositions from actual behavior. Inferring dispositions from counterfactual behavior is thus doubly problematic.

Furthermore, those interested in deep structure are generally normatively committed to a critique of current arrangements. It is not simply that agents are affected by their environment—all agree with that—but that the environment is biased, or people are dominated or subjugated by that environment. This normative critique is also problematic, of course. Since it is impossible to live outside of any environment, any social structure, every way of behaving will be determined (or, more weakly, we might say is suggested) by the structure. So it is not that we are comparing some structure-free set of actions with structured ones, but that we are comparing different sets of structures. So here the critics must, explicitly or implicitly, assume that some structures are superior, and these are ones that are domination-free. Generally, these are structures that are egalitarian. But of course such egalitarianism is a normative commitment eschewed by libertarians. In this case, any account of power as domination within structural forms is highly moralized and usually (though not without important exceptions) nonempirical.

An important aspect of deep structural accounts that has received greater emphasis in recent years is discourse and how that affects the manner in which we view the world. To examine the full literature on language and power would take this entry too far afield, but some studies—the immensely influential work of James Scott, for example—locate narratives strongly within the type of agential power relationships. Scott (1990) tries to theorize how communication operates across divisions of economic power. He rejects the idea that subordinate groups accept the dominant ideology; they might, rather, acquiesce in the public space while undermining the dominant modes within their own communities using a “hidden transcript” that critiques power behind the back of the dominant groups (Scott, 1990, p. xii). Scott further suggests that we can see many activities such as petty crime in a political light, as disguised attempts to resist (Scott, 1985). While Scott’s work effectively counters the Marxist idea that the oppressed fall in with their oppression, it ignores evidence that oppressors, too, act out in the public sphere (Irvine, 1990; Gal, 1995, p. 413). It also ignores the fact that countercultures often resist their own oppression by oppressing others (Gal, 1995, p. 420).

Dispositional versus Exercise

We have already seen that most accounts of power are dispositional ones. Indeed, as Morriss (2002) importantly demonstrates, to analyze power in terms of its exercise is fallacious. The exercise fallacy is the claim that having the power to do something is nothing other than doing it. However, it is clear that having powers, just like holding rights, does not mean that we have to exercise them. I have the power to keep a secret, but I do not have to keep it. Sharing information does not show that I had to share information. Morriss points out that decisionists such as Dahl and Polsby at times made such exercise claims, and those who use the concept of structural power also often commit the fallacy.

So power is a dispositional concept. Nevertheless, while it might be defined in dispositional terms, it can still be argued that important analyses should concentrate on its exercise. By analogy, anyone might be capable of acting corruptly, but what we are socially interested in is actual corruption. So while different resources might give different powers, what we really want to know is who exercises those resources and for what ends. There is obvious truth in such claims. However, we need to keep in mind different purposes to which power analysis can be put. Do we really want to deny the worth of comparative static analysis of the institutional powers of different political and social actors? Does the claim deny the interest in thinking about the comparative powers of different chief executives in different political systems, or comparing the powers held by, for example, the European Union and those held by its constituent states? So looking at how power is exercised does not require that we forgo the comparative statics enterprise.

We can distinguish here between social science that looks toward discovering causal mechanisms and empirical generalizations and social science that looks toward narrating and explaining specific token outcomes—between causal analysis and historical analysis. When it comes to historical analysis, we are interested in how power is exercised and, furthermore, we are interested in it so that we can understand how resources can be exercised. In other words, examining the exercise of power teaches us about the dispositional aspects of power. (I return to these issues below when I consider the issues surrounding comparative static analysis of power and dynamic analysis of power games.)

Conceptual Analysis

This section looks at a series of issues that demarcate specific accounts of power. Many of these lines of dispute are somewhat trivial, if not merely verbal, disputes of little interest to serious social science. Many of the issues can be elucidated by considering collective action and coordination problems. The failure to fully accommodate the issues of collective action and coordination is apparent in writers on both sides of the dividing line.

Power over versus Power to

Hanna Pitkin (1972, pp. 276–277) first introduced the distinction between “power to” and “power over.” “Power over” refers to one agent’s power over another agent: that is, to one agent carrying out her intentions despite the intentions of another, usually by using the second agent as a means to her end. “Power to” is not necessarily directed at others at all, but only to an agent’s ability to bring about some outcome. For some, only the “power over” notion truly carries the meaning of social or political power. For others, it is a subset, albeit perhaps the most important aspect, of “power to.” Having power over others is one means by which agents can have the power to bring about outcomes.

Structural accounts of power usually use a form of “power over,” while agential accounts might utilize either “power to” or “power over.” Structural accounts use power over as they are normative, and since power to is always a dispositional concept, using power to in a structural sense obviously commits the exercise fallacy. People behave in certain ways because others, or “the system,” have power over them. Power over is, generally speaking, a conflictual concept, as it implies a clash of interests. This is not always the case, however: after all, it is fairly obvious that parents have power over their children and use it against their children’s immediate desires, though not necessarily against their interests. Sports trainers might have power over their trainees; but the athletes might be happy to be forced into rigorous training routines. Where interests conflict, power is sometimes described as zero sum; this is misleading since, strictly speaking, zero-sum games are concerned only with ordinal preferences. Clashes of interest might occur when social or political power is of most fundamental normative interest, but we should not let our normative concerns determine our conceptual analysis.

The distinction between “power to” and “power over” is most obviously moralized in feminist literature, where “power over” refers to the illegitimate domination exercised by men over women and “power to” is used to refer to the legitimate use of power that women can develop to face down male oppression (Pansardi, 2011). It follows from Hannah Arendt’s (1958/1998) distinction between power and violence, where any power of one person over another is seen as a form of violence (see also Rodriguez, 2005). In this literature, power over is not necessarily intentional because it can operate in routine ways. However, when agents routinely act in ways that dominate others, even when they do not intend to do so, they are still using their power. Some might say that in those cases agents are not in control of their power, but that would be too hasty a conclusion. After all, I might not have intended the outcome of some thoughtless act, but had I given it consideration I could have avoided that outcome if I tried.

In this analysis, “power to” can then be extended to “power with,” as women can work together to overcome oppression, and power within, as they can see through oppressive forces (Allen, 1999). While these approaches try to conjoin aspects of agency and structure in the sense that they bring together the idea of people working together, which itself involves accounts of how people overcome collective action problems, they tend to elide distinctions between intentional and unintentional action, and risk missing important ways in which oppression can be overcome. They also create a technical language removed from that of everyday speech that can lead to misunderstandings for those outside the specific literature. Moralized conceptions of any social concept always face the danger that they only make sense within a theory and rely on the theory being right in all its details for the concepts to do the job asked of them.

For some, the distinction between the intentional and the unintentional use of power is important. Not only does it enable finer distinctions in terms of blame for outcomes, it can also enable us to work to overcome problems without having to try to assign blame. For example, we might examine the incidence of medical staff giving out the wrong medicine. Perhaps in any given case the staff member did not read the packet carefully enough. However, if different medicines are packaged in similar ways, such errors are easier to make. We could point out how easy it is to make such mistakes, telling staff to take more care, and then blaming those who are not careful enough. Or we can reduce the incidence of such errors by changing the packaging, without worrying about how much blame should be attached to agents in any instance; the second is a structural change, the first relies on agency. However, we do not want to ignore responsibility altogether. It matters in any given case whether the incorrect medicine was given by accident or deliberately, and if by accident how negligent the agent was. We might also want to interrogate how much blame to assign when agents are unaware of the consequences of their action. How aware should they have been of those consequences?

Relationship to Other Concepts

A host of concepts are related to power: authority, coercion, control, domination, freedom or liberty, force, hegemony, influence, manipulation, and so on. Some writers try to demarcate these terms, using ordinary language intuitions, together with analytic deductions, to create multiple interrelated and overlapping, but unique, terms to cover all sorts of contingencies (Oppenheim, 1981; Dahl & Stinebrickner, 2003). Some terms are clearly separate from the most general understanding of the term “power.” “Authority,” for example, is usually defined in terms of the legitimate use of power. When a person recognizes the authority of another, he recognizes that, within certain specified legal and moral constraints, he is bound to carry out the authority’s commands. The mainstream analytic understanding of authority in these terms is that an authority gives content-independent reason for an agent to act (Raz, 1986). That gives the justification of authority, but in the context of structural power, we can question how justified we are in routinely acting for content-independent reasons. Foucault suggests that what is regarded as authoritative is context- and time-specific and often merely hides other systemic power (see Blencoe et al., 2013).

Some of the other terms, such as coercion, are used in certain contexts as synonyms for power, though, given an expansive definition of power, they are subsets of “power” broadly understood. In some formal work, “freedom” and “power to” become identical. For example, Matthew Braham (2006) defines the specific freedom of an individual as being a member of a winning coalition that allows (does not stop) some action. This is closely related to the idea in noncooperative game theory of the power any agent brings to a winning coalition that ensures it remains winning. Domination is a normatively important aspect of power; republicans relate domination to measures of freedom (Pettit, 1997) and then relate these to basic notions of power (Pettit, 2008).

I will not offer any account of the relationship of power to other concepts here. For one thing, I am not convinced that any such general account has any utility. If we need to demarcate manipulation from coercion for a specific normative reason in a particular argument, then a careful demarcation of the two concepts needs to be carried out for that purpose. However, we need not be too concerned about potential differences in many applications. Often careful distinctions are not required. Social or political power is best seen as a generic concept, and forcing the meaning into subsets, such as domination or coercion, is usually carried out for rhetorical force to lead the reader into specific normative conclusions. Scientific discourse should cleanse itself of such sleight of hand and use concepts that are as primitive and nonnormative as possible (Dowding, 2017, ch. 8). By keeping our analysis of power as general as possible, we can make all the fine distinctions in terms of sub- and related categories that we need as we go about our analysis.

Having said that, we should cover a couple of popular accounts of power that have not been mentioned thus far. In international relations, Joseph Nye’s account of “soft power” has attracted massive attention (Nye, 1990a, 1990b, 2004). The concept is not an analytic one, as evidenced by its arrival in journals for policymakers. Nye introduced it to persuade U.S foreign policymakers that they needed to communicate, persuade, and encourage other states rather than try to bully them into accepting their vision. Its acceptance was manifested by references by the U.S. president and secretary of state in major speeches at the end of the first decade of the 21st century (see Gallarotti, 2015, n. 1). It is called “soft” to distinguish it rhetorically from “hard” power in the sense of military might, coercion, and material payments.

In many ways, soft power is simply a new name for Gramsci’s (1971) notion of hegemony (Nye, 2003; Dowding, 2012). A new term was perhaps require, because hegemony came to be associated with negative normative connotations—though in fact that is not how Gramsci uses the term when he argues that revolution involves replacing one hegemony with another, superior, one. “Soft power” covers two rather different processes. The first, and Nye’s original, is persuading people through information that the interests of both parties are similar. The second is changing the structural conditions of another country, for example, through trade. These two processes are rather different, and we can see a more analytic version in Harsanyi’s game-theoretic work. Recently, attempts have been made to make the term more analytic, alongside introducing the notion of “smart power” (Nye, 2004, 2011; Gallarotti, 2011; Rothman, 2011). Smart power is simply the use of both hard and soft power to attain one’s ends (Chong, 2015; Gallarotti, 2015). Nevertheless, the suspicion remains that these are simply labels or rhetorical flourishes that are tacked on to processes that have been examined more carefully in the analytic and discourse traditions of power study.

Collective Action and Power

Various aspects of power and controversies in the analysis of power can be visualized more clearly through considering collective action and coordination, both the metaphysics of a coordinated group and how we handle its resources and power and the problems associated with engaging in collective action. In this way, we can see more clearly the metaphysical necessities and contingencies of the power structure and the epistemological aspects of how we measure power in both comparative static and dynamic contexts.

The Dimensions of Power Again

Dahl (1961a) and Polsby (1980) explore the idea that some people do not engage in politics, and they assume that is because they do not wish to. The pluralist assumption of group activity suggests, at least implicitly, that the degree to which groups engage in political pressure is a proxy for the strength of their interest. The collective action calculation demonstrates that this notion is false. Collective action is more problematic for large than for small groups, and for those with diffuse rather than concentrated interests, while the relative costs of collective action can vary massively across different sets of people. Lack of mobilization does not demonstrate that people do not have a strong interest in mobilizing.

Radical critics make the same error. They commit the political power or “blame fallacy” (Dowding, 1991, pp. 89–95). Bachrach and Baratz (1970), Crenson (1971), Lukes (1974, 2005), and Gaventa (1980) all assume, at least in part of their analyses, that because some group is not acting in its own interests, another group is acting contrary to those interests. But that is not the case. People can be powerless all on their own. If working together is costly, especially (though not exclusively) where free riding seems to be an option, then groups that share an interest can fail to organize (Olson, 1965/1971; Dowding, 1996, ch. 2). Without a careful analysis of the structural position of a set of people, the potential costs and benefits they face, we can make no inferences about their interests or their potential power. We can also apply this logic to the role of deep structure in creating interests and preferences.

What makes collective action so central to any understanding of power is the complex counterfactual nature of its analysis. Consider any dictator. He can only maintain power by marshalling a coalition of support. He rewards his supporters with side-payments gained through exploiting the majority. Often, successful dictators set up moving coalitions with different institutions doing very similar jobs, so each is rival to the others and can be used to ensure that no other agents can form a new winning coalition to bring down the dictator. A dictator is successful by creating incentives for agents to support him and by ensuring that the majority does not have a strong enough incentive to bring about radical change. However, these coalitions could always coalesce differently to bring him down. We see in revolutionary times that if the mass behaves radically enough it can swiftly defeat dictators, often bringing over some of the dictator’s supporters. Counterfactually, any of these moves against a dictator could occur at any given time.

That is the logic of collective action. We can act together to get what is in our collective interests, but all too often we do not. As well as incentives to act together, there are incentives not to so act. In revolution, some will die in the streets, so that others may gain. The likelihood of success will determine how great the risks people will take. Studying the incentives that lead people to coalesce is the only way to go beyond comparative static analysis and see the dynamic aspects of social and political power. These require careful theoretical analysis, together with detailed understanding of actual collective action and coordination strategies in both revolutionary and conventional situations. We need to understand what incentivizes people institutionally, morally, and rhetorically to engage in collective acts.

An aspect of those structures is tacit coalitions of interests that work together to serve diverse interests. The growth machine and neo-Marxist analyses show how democratic leaders can favor growth and finance strategies because they bring short-term electoral advantage. There need be no explicit bargained agreements between financiers and politicians, as long as politicians see that securing the finance interests brings short-term benefits to constituents. We saw from the global financial crisis of 2008 that incompetence and obviously corrupt behavior (albeit not technically legally corrupt) are not punished by governments if they believe that reconstructing the entire global finance system is not a process that will bring them immediate electoral rewards (Bell & Hindmoor, 2015).

Actors whose interests are looked after by the political system without their actually engaging in activity to forward those interests have been called “systematically lucky” (Dowding, 1991, 1996). The term is perhaps best applied to those who do not have the power to secure their own interests; but finance capital, which of course also has great power, does not always need to use that power. Actors can also act now to secure systematic luck in the future (Dowding, 1999, 2016, ch. 7; Hindmoor & McGeechan, 2013).

Some object to the powerful being considered lucky since they consider that it lets them off the hook (Barry, 2002, 2003; Lukes & Haglund, 2005; see Dowding, 2003, 2016, ch. 6, for response). The critique is largely normative and comes back to blame. One aspect of structural forces is not simply that one has acted to bring about some outcome, but what one could have done to change some outcome. Generally speaking, we assign greater blame to commission than omission. However, omitting to act can sometimes be more blameworthy than acting: not acting to save a drowning baby is worse than littering. Either way, we need to look at the incentives to act or not act and the damage done thereby. To bring about change we do not have to assign blame. We might think that it is the structure of incentives that is to blame. Politically, we should then act to change those structures. At that stage, those who gain from the existing social, economic, and political structures, whether through their own ability to wield power or through the systematic luck generated by tacit coalitions, will come forward to defend their interests.

Virtually all the distinctions and claims made about social power can be viewed and analyzed though the collective action problem. Even the deep structures that create internal incentives to behave in one manner or another can be seen as an internal collective action problem between the different interests we have as agents playing different roles in society. Our roles as parents, bureaucrats, anglers, and poker players can provide different sets of interests that might conflict over social policy. How we weigh those interests internally, and how we might act in inconsistent ways as we play our lives in those different roles, can be viewed as an internal power game, where the agent tries to form a coherent coalition of interests. We also play games with ourselves over time, as we weigh our immediate interests against future ones. For example, Ainslie (1992) analyses decisions we make as games played by our different interests over our life cycle. So our desire to eat too much competes with our desire to be healthy, and we try to commit our future selves to one behavior rather than another by joining fitness clubs, by the food we buy for the week ahead, and so on.

Social and political power is a complex concept that can be studied in different ways depending on the specific research questions. Comparative static analysis of the powers in different agents in different institutional and structural situations is best conducted by looking at the relative resources of those agents. Dynamic analysis of actual power plays needs careful investigation of the strategies that agents adopt, the specific luck they enjoy at given moments, and the context in which they act. In both cases, awareness of the incentives that are provided by the structural situation of the agents is necessary for thorough interpretation of their powers and why they do not use them. Understanding the collective action problem and the ways in which actions can be coordinated is the key to understanding at both the structural and the agential level.

References

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York: Harper & Row.Find this resource:

Ainslie, G. (1992). Picoeconomics: The strategic interaction of successive motivational states within the person. Cambridge, U.K.: Cambridge University Press.Find this resource:

Allen, A. (1996). Foucault on power: A theory for feminists. In Susan Hekman (Ed.), Feminist interpretations of Michel Foucault. University Park: Pennsylvania State Press.Find this resource:

Allen, A. (1999). The power of feminist theory: Domination, resistance, solidarity. Boulder, CO: Westview.Find this resource:

Altemeyer, R. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press.Find this resource:

Arendt, H. (1958/1998). The human condition. Chicago: University of Chicago Press.Find this resource:

Bachrach, P., & Baratz, M. (1962). Two faces of power. American Political Science Review, 56(4), 947–952.Find this resource:

Bachrach, P., & Baratz, M. (1970). Power and poverty: Theory and practice. New York: Oxford University Press.Find this resource:

Baldwin, D. A. (2015). Misinterpreting Dahl on power. Journal of Political Power, 8(1–3), 209–227.Find this resource:

Barry, B. (2002). Capitalists rule, OK? Some puzzles about power. Politics, Philosophy and Economics, 1(2), 155–184.Find this resource:

Barry, B. (2003). Capitalists rule, OK? A commentary on Keith Dowding. Politics, Philosophy and Economics, 2(3), 323–341.Find this resource:

Bartky, S. (1990). Femininity and domination: Studies in the phenomenology of oppression. New York: Routledge.Find this resource:

Bassett, K. (1996). Partnerships, business elites and urban politics: New forms of governance in an English city? Urban Studies, 33, 539–555.Find this resource:

Bassett, K., & Harloe, M. (1990). Swindon: The rise of a growth coalition. In Michael Harloe, Chris Pickvance, & John Urry (Eds.), Place, policy and practice: Do localities matter? London: Unwin Hyman.Find this resource:

Bell, S., & Hindmoor, A. (2015). Masters of the universe but slave of the market: Bankers and the great financial meltdown … and how some banks avoided the carnage. Cambridge, MA: Harvard University Press.Find this resource:

Blencoe, C., Brigstocke, J., & Dawney, L. (2013). Editorial: Authority and experience, Journal of Political Power, 6(1), 1–7. Special Issue.Find this resource:

Braham, M. (2006). Measuring specific freedom. Economics and Philosophy, 22, 317–333.Find this resource:

Butler, J. (1993). Bodies that matter: On the discursive limits of “sex.” New York: Routledge.Find this resource:

Butler, J. (1997). The psychic life of power: Theories in subjection. Stanford, CA: Stanford University Press.Find this resource:

Castells, M. (1977). The urban question. London: Edward Arnold.Find this resource:

Castells, M. (1978). City, class and power. London: Macmillan.Find this resource:

Chalmers, D. J. (2011). Verbal disputes. Philosophical Review, 120(4), 515–566.Find this resource:

Chong, A. (2015). Smart power and military force: An introduction. Journal of Strategic Studies, 38(3), 233–244.Find this resource:

Clark, Terry N., & Goetz, E. G. (1994). The anti-growth machine: Can city governments control, limit or manage growth? In Terry Nichols Clark (Ed.), Urban innovation: Creative strategies for turbulent times. Thousand Oaks, CA: SAGE.Find this resource:

Clegg, S. R., Courpasson, David, & Phillips, Nelson. (2006). Power and organizations. London: SAGE.Find this resource:

Crenson, M. (1971). The unpolitics of air pollution. Baltimore, MD: Johns Hopkins University Press.Find this resource:

Dahl, R. A. (1957). The concept of power. Behavioral Science, 2(3), 201–215.Find this resource:

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

Dahl, R. A. (1961b). The behavioral approach in political science: Epitaph for a monument to a successful protest. American Political Science Review, 55, 763–772.Find this resource:

Dahl, R. A. (1968). Power. In D. L. Sils (Ed.), International encyclopedia of the social sciences. Vol. 12. New York: Free Press.Find this resource:

Dahl, R. A. (1986). Rethinking who governs? New Haven Revisited. In Robert Waste (Ed.), Community power: Directions for future research. Beverly Hills, CA: SAGE.Find this resource:

Dahl, R. A., & Stinebrickner, Bruce. (2003). Modern political analysis. Upper Saddle River, NJ: Prentice-Hall.Find this resource:

DeLeon, R. (1992). West Coast city: Progressive politics in San Francisco 19751991. Lawrence: University of Kansas Press.Find this resource:

Domhoff, G. W. (1978/2010). Who really rules? New Haven and community power reexamined. Brunswick, NJ: Transactions.Find this resource:

Domhoff, G. W. (1986). The “growth machine and the power elite.” In Robert Waste (Ed.), Community power: Directions for future research. Beverly Hills, CA: SAGE.Find this resource:

Domhoff, G. W. (2013a). Who rules America? The triumph of the corporate rich. New York: McGraw-Hill.Find this resource:

Domhoff, G. W. (2013b). Who rules America? Power and politics. New York: McGraw-Hill.Find this resource:

Dowding, K. (1991). Rational choice and political power. Aldershot: Edward Elgar.Find this resource:

Dowding, K. (1994). The compatibility of behaviouralism, rational choice and “new institutionalism.” Journal of Theoretical Politics, 6(1), 105–117.Find this resource:

Dowding, K. (1996). Power. Buckingham: Open University Press.Find this resource:

Dowding, K. (1999). Shaping future luck. Journal of Conflict Processes and Change, 4(1), 1–11.Find this resource:

Dowding, K. (2001). Explaining urban regimes. International Journal of Urban and Regional Research, 25(12), 7–19.Find this resource:

Dowding, K. (2003). Resources, power and systematic luck: Reply to Barry. Politics, Philosophy and Economics, 3(2), 305–322.Find this resource:

Dowding, K. (2006). Three dimensional power: A discussion of Steven Lukes’ Power: A Radical View second edition. Political Studies Review, 4(2), 136–145.Find this resource:

Dowding, K. (2008). Power, capability and ableness: The fallacy of the vehicle fallacy. Contemporary Political Theory, 7(3), 238–258.Find this resource:

Dowding, K. (2012). Why should we care about the definition of power? Journal of Political Power, 5(1), 119–135.Find this resource:

Dowding, K. (2016). The philosophy and methods of political science. London: Palgrave.Find this resource:

Dowding, K. (2017). Power, luck and freedom: Collected essays. Manchester, U.K.: Manchester University Press.Find this resource:

Dowding, K. Dunleavy, Patrick, King, Desmond, Margetts, Helen, & Rydin, Yvonne. (1999). Regime politics in London local government. Urban Affairs Review, 34(4), 515–545.Find this resource:

Elkin, S. L. (1987). City and regime in the American Republic. Chicago: University of Chicago Press.Find this resource:

Fainstein, N., & Fainstein, S. (1986). Regime strategies, communal resistance and economic forces. In Susan Fainstein, R. C. Hill, Dennis Judd, and M. Smith (Eds.), Restructuring the city: The political economy of urban development. New York: Longman.Find this resource:

Flyvbjerg, B. (1998). Rationality and power: Democracy in practice. Cambridge, U.K.: Cambridge University Press.Find this resource:

Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge, U.K.: Cambridge University Press.Find this resource:

Foucault, M. (1977). Discipline and punish: The birth of the prison. Trans. Allan Sheridan. New York: Random House.Find this resource:

Gal, S. (1995). Language and the “arts of resistance.” Cultural Anthropology, 10(3), 407–424.Find this resource:

Gallarotti, G. M. (2011). Soft power: What it is, why it’s important, and the conditions for its effective use. Journal of Political Power, 4(1), 25–47.Find this resource:

Gallarotti, G. M. (2015). Smart power: Definitions, importance, and effectiveness. Journal of Strategic Studies, 38(3), 245–281.Find this resource:

Gaventa, J. (1980). Power and powerlessness: Quiescence and rebellion in an Appalachian Valley. Oxford: Oxford University Press.Find this resource:

Gramsci, A. (1971). Selections from the Prison Notebooks. Ed. and trans. Quintin Hoare and Geoffrey Nowell Smith. London: Lawrence and Wishart.Find this resource:

Harding, A. (1991). The rise of growth coalitions, UK-Style? Government and Policy, 9, 295–317.Find this resource:

Harding, A. (1994). Urban regimes and growth machines: Towards a cross-national research agenda. Urban Affairs Quarterly, 29, 356–382.Find this resource:

Hardy, C., & Leiba-O’Sullivan, S. (1998). The power behind empowerment: Implications for research and practice. Human Relations, 51(4), 451–483.Find this resource:

Harsanyi, J. C. (1962a). Measurement of social power in N-person reciprocal power situations. Behavioral Science, 7(1), 81–91.Find this resource:

Harsanyi, J. C. (1962b). Measurement of social power, opportunity costs, and the theory of two-person bargaining games. Behavioral Science, 7(1), 67–80.Find this resource:

Harsanyi, J. C. (1976). Essays on ethics, social behavior and scientific explanation. Dordrecht: D. Reidel.Find this resource:

Hayward, C., & Lukes, S. (2008). Nobody to shoot? Power, structure, and agency: A dialogue. Journal of Power [now Journal of Political Power], 1(1), 5–20.Find this resource:

Hindmoor, A., & McGeechan, J. (2013). Luck, systematic luck and business power: Lucky all the way down or trying hard to get what it wants without trying. Political Studies, 61(4), 834–850.Find this resource:

Hunter, F. (1953). Community power structure: A study of decision makers. New York: Anchor Books.Find this resource:

Irvine, J. T. (1990). Registering affect: Fleteroglossia in the linguistic expression of emotion. In Catherine A. Lutz and Lila Abu-Lughod (Eds.), Language and the politics of emotion. Cambridge, U.K.: Cambridge University Press.Find this resource:

List, C., & Spiekermann, K. (2013). Methodological individualism and holism in political science: A reconciliation. American Political Science Review, 107(4), 629–643.Find this resource:

Logan, J., & Molotch, H. (1984). Tensions in the growth machine: Overcoming resistance to value-free development. Social Problems, 31, 483–499.Find this resource:

Logan, J., & Molotch, H. (1987). Urban fortunes: The political economy of place. Berkeley: University of California Press.Find this resource:

Lukes, S. (1974). Power: A radical view. London: Macmillan.Find this resource:

Lukes, S. (2005). Power: A radical view: The original text with two major new chapters. Basingstoke: Palgrave Macmillan.Find this resource:

Lukes, S. (2015). Robert Dahl on power. Journal of Political Power, 8(1–3), 261–271.Find this resource:

Lukes, S., & Haglund, L. (2005). Power and luck. Archives Europeénnes de Sociologie, 46(1), 45–66.Find this resource:

Martin, J. L. (2001). The authoritarian personality fifty years on: What questions are there for political psychology. Political Psychology, 22(1), 1–26.Find this resource:

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

Morriss, P. (2002). Power: A philosophical analysis. 2d ed. Manchester, U.K.: Manchester University Press.Find this resource:

Mossberger, K., & Stoker, G. (2001). The evolution of urban regime theory: The challenge of conceptualization. Urban Affairs Review, 36(6), 810–835.Find this resource:

Nye, J. (1990a). Bound to lead: The changing nature of American power. New York: Basic Books.Find this resource:

Nye, J. (1990b). Soft power. Foreign Policy, 80 (Fall), 53–71.Find this resource:

Nye, J. (2003). The velvet hegemon: How soft power can help defeat terrorism. Foreign Policy, 136 (May–June), 74–75.Find this resource:

Nye, J. (2004). Soft power: The means to success in world politics. Washington, DC: Public Affairs Institute.Find this resource:

Nye, J. (2011). Power and foreign policy. Journal of Political Power, 4(1), 9–24.Find this resource:

O’Conner, J. (1973). The fiscal crisis of the state. London: St. James Press.Find this resource:

Olson, M. (1965/1971). The logic of collective action: Public goods and the theory of groups. 2d ed. Cambridge, MA: Harvard University Press.Find this resource:

Oppenheim, F. (1981). Political concepts: A reconstruction. Oxford: Basil Blackwell.Find this resource:

Orr, M. E., & Stoker, G. (1994). Urban regimes and leadership in Detroit. Urban Affairs Quarterly, 30(1), 48–73.Find this resource:

Pansardi, P. (2011). Power to and power over. In Keith Dowding (Ed.), Encyclopedia of power (pp. 521–524). Beverly Hills, CA: SAGE.Find this resource:

Pettit, P. (1997). Republicanism: A theory of freedom and government. Oxford: Oxford University Press.Find this resource:

Pettit, P. (2008). Dahl’s power and republican freedom. Journal of Power [now Journal of Political Power], 1(1), 67–74.Find this resource:

Pitkin, H. F. (1972). The concept of representation. Berkeley: University of California Press.Find this resource:

Polsby, N. W. (1980). Community power and political theory. 2d ed. New Haven, CT: Yale University Press.Find this resource:

Raz, J. (1986). The morality of freedom. Oxford: Oxford University Press.Find this resource:

Ricci, D. (1971). Community power and democratic theory: The logic of political action. New York: Random House.Find this resource:

Rodriguez, E. (2005). Hannah Arendt’s concept of power and violence. Houston, TX: University of Houston.Find this resource:

Rothman, S. R. (2011). Revising the soft power concept: What are the means and mechanisms of soft power? Journal of Political Power, 4(1), 49–64.Find this resource:

Savitch, H., & Thomas, J. (1991). Conclusion: End of the millennium big city politics. In H. Savitch and J. Thomas (Eds.), Big city politics in transition. Newbury Park, CA: SAGE.Find this resource:

Scott, James C. (1985). Weapons of the weak: Everyday forms of peasant resistance. New Haven, CT: Yale University Press.Find this resource:

Scott, James C. (1990). Domination and the arts of resistance: Hidden transcripts. New Haven, CT: Yale University Press.Find this resource:

Stone, C. N. (1980). Systemic power in community decision-making: A restatement of stratification theory. American Political Science Review, 74(4), 978–990.Find this resource:

Stone, C. N. (1988). Pre-emptive power: Floyd Hunter’s “community power structure” reconsidered. American Journal of Political Science, 32, 82–104.Find this resource:

Stone, C. N. (1989). Regime politics: Governing Atlanta 19461988. Lawrence: University Press of Kansas.Find this resource:

Stone, C. N. (1993). Urban regimes and the capacity to govern: A political economy approach. Journal of Urban Affairs, 15(1), 1–28.Find this resource:

Winner, L. (1980). Do artifacts have politics? Daedalus, 109(1) (Winter), 121–136.Find this resource:

Wolfinger, R. (1960). Reputation and reality in the study of community power. American Sociological Review, 25(4), 636–644.Find this resource:

Woodward, J. (2003). Making things happen: A theory of causal explanation. Oxford: Oxford University Press.Find this resource: