INTERPERSONAL POWER

In its broadest sense, interpersonal power refers to any cause of any change in the behavior of one actor, B, which can be attributed to the effect of another actor, A. It sometimes refers to the capacity to cause such change (Weber [1918] 1968), sometimes to actual use of that capacity (Dahl 1957; Simon 1953) but always to overcoming the "resistance" of B (Weber [1918] 1968), hence causing B to do something B would not otherwise do (Dahl 1957). Interpersonal power is therefore the power of one individual "over" another as opposed to an individual's power to do something, the capacity of an actor to attain some goal (as in Russell 1938). "Power over" always implies a relation between two actors rather than referring to an attribute of an actor. It is sometimes thought of as "micro" power and contrasts with "power to," which is attributed to collectives (Hawley 1963; Parsons 1963) and is thought of as "systemic" or "macro" power.

Adequate description of a power relation will typically refer to: (1) the bases of power (the bases of A's power over B are the resources that are possessed by A that are instrumental to the goals of B); (2) the means of power (the ways in which A uses these resources to change the behavior of B); (3) the strength of power (the costs to B if B does not comply with demands by A); (4) the costs of power (the costs of A of having to exercise power over B); (5) the amount of power (the extent to which A is able to get B to do something that B would not otherwise have done); (6) the scope of power (the acts with respect to which the amount of A's power over B is greater than zero); and (7) the domain of power (the persons over whom the amount of A's power is greater than zero). However, there is a great deal of disagreement over which of these constitute power and what kinds of bases, means, costs, and particularly compliance the word covers.

CONCEPTS OF POWER

In his work on power, Simon (1953) treats the effects rather than the causes of power and covers by the one word a whole family of concepts describing the effects of all the human causes of human conduct. Force, power, influence, authority, and manipulation all have a strong family resemblance to each other, and Dahl (1957), March (1955), and Simon (1953) have all treated them as one process. But French and Raven (1959) have pointed out important ways in which the dynamics of different kinds of power differ, and March (1966) has shown compellingly that treating them all as one unitary process leads only to a dead end.

The laws of force, for example, differ in a fundamental way from the laws governing power. In using force, A does not require B to choose between compliance or noncompliance. A kills B, imprisons B, drags B, wrestles B to the ground, but does not require choice by B. Threat of force requires a choice, but threat of force is a different process (Goode 1972). A threat may be futile if B has decided to die for a cause; the threat may not accomplish its purpose. Actual force will accomplish its purpose whether B chooses to comply or not. It simply removes B as a factor opposed to A's wishes.

Like force, manipulation does not require that B choose between complying or not complying with the wishes of A. A may control B by controlling the information at B's disposal, by preventing certain choices being open to B, or by activating motives of B known to lead B to do what A wants. In some sense, B may be said to make choices (hence the difference from force), but A does not require B to choose between compliance and noncompliance. A controls the conditions that govern how B analyzes the choice to be made.

Although there are more than one narrow senses of the term power, all are distinguished from force and manipulation by the fact that power involves a choice by B between compliance and noncompliance. But this is true also of influence and authority. What distinguishes power is that it involves external sanctions. In Blau (1964) B is coerced to do something, X, by threat of a penalty for noncompliance. In Festinger (1953), B is induced to do X by a promise of reward for compliance or coerced to do X by threat of a penalty for noncompliance. Both rewards and penalties are external to the actor. They make no internal change, no change in the actor's state of mind. B does not change views privately, even if B conforms publicly. Influence, on the other hand, persuades B that X is right according to B's own interests; hence B complies privately as well as publicly. Compliance is willing in the case of influence, "forced" (Festinger 1953) in the case of power. If compliance were not observable to A, B would still comply in the case of influence but would not in the case of power. Hence, power is highly dependent on observability, but influence is not (French and Raven 1959).

Authority differs from both: It refers to a claim by A, accepted by B, that A has a legitimate right to expect compliance by B, even if compliance runs counter to B's own preferences (Barnard 1938). It differs from influence because whether B likes or does not like X is irrelevant. It differs from power in that B is expected to comply (and, if B accepts A's authority, does comply) because it is right, not because it is expedient. If B accepts A's legitimate authority, then B complies with A's commands whether compliance is observable to A or not. Furthermore, the power that legitimate authority makes possible has different effects because it is legitimate. Power exercised outside the scope of legitimate authority creates "reactance" (Brehm 1966), but power exercised inside the scope of legitimate authority does not (French and Raven 1959).

French and Raven (1959) also distinguish reward from punishment power and hence inducement from coercion. Coercion causes reactance; inducement does not. But they are so indissolubly connected that it becomes difficult to treat them as distinct processes. A reward forgone is a cost, and a penalty forgone is a reward. Hence, withholding a reward is equivalent to imposing a penalty, and withholding a penalty is equivalent to giving a reward. In order to distinguish the two, one would have to be able to separate giving something from withholding something. This might be possible in principle, but not in the case of power. Power depends on the contingency of sanctions; that is, A must be able to give or withhold something, depending on whether B does not comply. It is difficult to have one without the other. Thus, probably the most satisfactory concept of interpersonal power is Festinger's forced compliance (1953), compliance that is public but not private, caused either by threat of penalty or promise of reward.

This still leaves the question of whether power is potential or actual. Weber ([1918] 1968) emphasized the capacity of one actor to overcome the resistance of another whether or not the capacity is actually used. Simon (1953) and Dahl (1957) have objected to making inferences from a potential that might or might not actually be used. Research in fact consistently shows that not all potential power is used (see below), and Dahl in particular has argued that power has an effect only if used. On the other hand, Friedrich (1937) has argued that B might comply with the preferences of A because of anticipated reward or punishment, even if A says or does nothing that overtly demands compliance by B. Bachrach and Baratz (1963) go further and argue that certain kinds of acts that involve B doing nothing (nondecisions) occur without B even needing to know what A will do, simply because B knows what A could do (a hypothesis confirmed by Zelditch and Ford, 1994). Both arguments imply that sheer existence of potential power has an effect whether used or not. The effect does not even have to be intended.

The issue of intentionality is another of the disputed questions in conceptualizing power, but not only can potential power have an effect without being used, there are even theories, such as those of Cartwright (1959) and Emerson (1962, 1972), in which "use" of power occurs without necessarily being intended by anyone.

The dispute over potential versus actual power refers more to causes, the dispute over broad versus narrow concepts of power more to effects. Thus, another ambiguity of power is whether one is referring to causes, effects, or the process relating the two.

THE CAUSAL APPROACH TO POWER

As process, power most typically refers to an exchange of "resources" (characteristics or objects instrumental to the attainment of goals). The only exception is the "causal" approach to power. Simon (1953) pointed out the similarity between the concept of "power over" and causality, leading to a longstanding tradition in which the former is defined in terms of the latter (as in Dahl 1957; Nagel 1975). But this led to so many difficulties that March (1966) concluded that the concept of power was superfluous. Although Nagel (1975) attempted to give the concept more explanatory power by narrowing it (to the effects of A's preferences on B), this led to little further advance, possibly because causal modeling gives one little insight into the process relating causes to effects.

But "process" theories themselves divide into four types: field theory; rational choice, or decision, theories of exchange; behavioral exchange theory; and a neo-Weberian "resistance" theory of exchange.

THE FIELD THEORY OF POWER

In field theory, power is the ability to activate forces in the life space of another in the direction of X (Cartwright 1959). The capacity to activate such forces depends on the motive base of B, thought of as a tension system, which, when activated by A, produces a vectored force that reduces the tension. Thus, power depends on the needs of B as much as on the wishes of A. It was this approach that gave rise to the concept of a "resource": Transferable resources give rise to power. But use of power by A faces not only opposing forces leading B away from X but also resistance due to the exercise of power itself. (This resistance was later termed reactance by Brehm 1966 to distinguish it from resistance due to B's dislike of doing X itself.) A's "control" over B is the outcome of the opposing forces acting on B. Thus, the outcome of the reduction of tension in the force field is distinguished from A's power in a manner corresponding to the distinction between actual versus potential power. In field theory, therefore, success—that is, actual compliance by B—is not the measure of A's power since there are many forces other than A affecting B's behavior.

Field theory, like the causal approach to power, treats the whole family of power concepts as power, excluding only physical force. But research in this tradition was much concerned with differences between different types of power, including reward power, punishment power, expert power, legitimate power, and referent power (power arising from B's attraction to A, which causes a desire to be similar to A). It was found that both reward and punishment power depend on A's ability to observe B's behavior, while the other forms of power do not. Coercion increased but reward decreased reactance. Legitimacy normatively constrained power so that using it outside its legitimate scope reduced its effect. Use of coercion when it is legitimate decreased reactance, while use of rewards when they are not legitimate increased reactance (French and Raven 1959). Because of reactance, use of coercion depends on restraints that prevent B from leaving the field.

EXCHANGE THEORIES OF POWER

The most important difference between field theory and exchange theories is that in the latter alternatives become the most important factor in the analysis of power. The reason is that, assuming voluntary relations, an actor may do things that she or he would prefer not to do because they are nevertheless preferable to any available alternative. Thus, an abused wife may not leave her husband because separation or divorce are either even less desirable or even more costly. But exchange theories themselves come in three somewhat distinct forms: rational choice, or decision, theory; behavioral exchange theory; and resistance theory.

Rational Choice Theories of Exchange and Power. Thibaut and Kelley (1959) offer perhaps the earliest rational choice model of the exchange process applied to power. They treat all interaction by analogy to the exchange of goods and services, each actor's participation in exchange being determined by choice among alternatives, the outcomes of which are characterized by their rewards and costs. Rewards are simply the positive values, costs the negative values, associated with the consequences of choosing an alternative. The value that results from the algebraic sum of rewards and costs is the payoff for a given act. Given interaction between two actors, A and B, payoffs are a joint function of the choices made by each. A matrix of joint payoffs for alternative actions by A and B formulates the conditions determining interaction between the two, in particular whether a relation forms and persists. Thibaut and Kelley treat payoffs in terms of their subjective value for the actors, hence in terms of their "utility" (although they do not use the term). In general, actors are assumed to maximize utility and continue in a relation only if for each actor a course of action is better than the best available alternative.

Thibaut and Kelley deal largely with stable relations rather than with particular acts. Power in a relation arises out of the "dependence" of the actors on each other. It can take either (or both) of two forms: A may be able to control the outcomes of B independently of any act by B, hence controlling B's fate, or A may be able to make rewards and costs contingent on B's behavior, hence controlling B's behavior. But an important principle of the theory is that control of an individual's fate can be converted into behavioral control.

There are many variants of the decision-theoretic approach to exchange, each differing with respect to how they treat the choice that underlies it. Value can be treated as objective (e.g., in monetary terms) or subjective (e.g., in terms of the meaning of money to the actor). The latter gives rise to a utility theory, such as in Thibaut and Kelley (1959), Blau (1964), and Bacharach and Lawler (1981). The relation between choice of an alternative and its outcome can be treated deterministically (as is true of all three of the utility theories just cited) or probabilistically. If value is treated objectively and outcomes probabilistically, one has an expected utility theory as in Harsanyi (1962). Among probabilistic theories, probability itself can be treated objectively or subjectively: In March (1955) the probabilities are objective, in Nagel (1968) and Tedeschi and colleagues (1973) they are subjective. In the latter case one has a subjective expected utility theory. In all these variants of decision theory, the central axiom is that the actor maximizes whatever the theory's criterion of choice.

Despite the variations among these theories, in general they predict that, given a conflict of interest, A is more likely to use power when more is at stake, when the cost of exercising power is less, when the likelihood of compliance is greater, and when there are fewer alternative means of obtaining compliance. The costs of exercising power include the relative depletion of A's stock of resources, the likelihood of retaliation, and the effect of reactance on subsequent relations with B. In general, the evidence supports these predictions, although "use" sometimes refers to whether or not A prefers to impose his or her will on B; sometimes to whether or not, if A does have such a preference, it is accomplished by manipulating rewards and punishments rather than by other means; and sometimes to whether or not A's behavior is overt and explicit, communicating threats or promises, or is covert. Rational choice theories will also in general predict that B is more likely to comply with A's preferences the larger the reward attached to compliance, the larger the penalty attached to noncompliance, the greater the credibility of A's promises of reward or threat of penalty, and the fewer the alternatives available to B. Again, the evidence largely supports these predictions. But it is well to keep in mind that with respect to both use and compliance, the evidence deals largely with voluntary relations.

Unlike other decision theories, subjective expected utility theory (especially Nagel 1968) makes use of Friedrich's (1937) "law of anticipated reactions," in which compliance occurs without directives, threats, or promises by A. In the case of punishment, not only is there no visible exercise of power by A, but compliance by B may appear to be willing because there is no visible resistance. In addition, if one takes into account that the law of anticipated reactions can be applied to A as well as to B, A may know in advance that B will comply and therefore know that it is unnecessary visibly to exercise power in order to achieve A's objective (Samuel and Zelditch 1989).

Bacharach and Lawler (1981) have pointed out that even though relative power in a relation is necessarily zero-sum, meaning that a gain by one actor implies loss by the other, relative power can be distinguished from absolute power, the sheer quantity of resources controlled by an actor. Hence, relative power can be distinguished from total power, the sum of the resources possessed by A and B. Total power, unlike relative power, is a variable that can increase for both actors at the same time. Bacharach and Lawler argue that total power has an effect independent of relative power, especially on the use of punitive power capabilities. Given a conflict of interest, if punitive power is an available tactic, the likelihood of its use by the more powerful actor increases as his or her relative power increases but decreases as total power increases.

Thibaut and Kelley note that all these ideas are more useful as the pattern of exchange becomes more stable; hence, the focus of exchange theory is often not only on the fact that power is relational, which is true even at the level of a unit act, but on the structure of relations arising from enduring patterns of repeated exchanges. The most influential theory of this kind has been Emerson's theory of power–dependence relations (1962, 1972). Though it originally grew out of Thibaut and Kelley's work, this theory has come to be associated more recently with behavioral exchange theory, a theory originating in the work of Homans.

Behavioral Exchange Theory. Homans (1961) behaviorized exchange theory by transforming rewards and costs into reinforcement contingencies and the process of exchange into mutual operant conditioning. Many fundamental concepts of decision theory, however, survive in behavioral exchange theory. Choices are determined by maximizing profit, which is the difference between rewards and costs. Costs in Homans take a form more like costs in economics than in Thibaut and Kelley, becoming opportunity costs, that is, profit forgone by choosing X over not-X. But, as in Thibaut and Kelley, exchange occurs at a point at which for both actors exchange is more profitable than each actor's best available alternative. Thus, A uses power in Homans—where "use" means attempts to direct the behavior of another—when it is more profitable than any alternative available to A, and B complies when compliance is more profitable than noncompliance. Homans, however, gives more emphasis than do Thibaut and Kelley to the effect of satiation on the value of a reward: Accumulation of rewards brings diminishing returns. Hence, poor actors are more responsive than are rich ones to offers of any given level of reward for doing an otherwise undesired act.

Emerson's power–dependence theory, perhaps the most influential theory of interpersonal power, was originally a utility theory of relations like Thibaut and Kelley's (Emerson 1962). But Emerson (1972) followed Homans in behaviorizing exchange theory, except that, unlike Homans, Emerson again dealt with relations rather than acts. Power–dependence theory is formulated in terms of two actors whose social relations entail ties of mutual dependence. "Power" is the amount of B's resistance that can be overcome by A: It is potential rather than actual power that is used to describe the relation. Power is a function of dependence, which arises from the control by one actor of resources on which another depends for achieving his or her goals. Dependence varies directly with an actor's motivational investment in goals but inversely with the availability of alternative sources of resources outside the relation. The "power advantage" of one actor over another is a function of the net balance of each one's dependence on the other. If the net balance is zero, power is equal, or "balanced." A relatively unique feature of Emerson's theory is that balanced power is assumed to be stable, while imbalance creates pressures to change the power relation in the direction of balance. There are four kinds of balancing operations. B may (1) reduce motivational investment in the goals mediated by A, withdrawing from the relation; (2) increase the number of alternative sources of the resource (extend networks); (3) increase A's motivational investment in goals mediated by B (e.g., by offering status to A); or (4) deny to A alternative sources of resources mediated by B (coalitionformation). Attention to these operations leads one to distinguish use of power from change of power—increasing or decreasing dependence in the relation. These assumptions about balance have had a major impact on the study of organizations (beginning with Thompson 1967), but other exchange theories have tended to reject the balance assumption, especially Blau (1964), who instead treats the balance mechanisms as means by which A maintains or increases power over B.

"Use" in power–dependence theory comes to mean something quite different from what it means in decision or field-theoretic formulations: It refers now to asymmetries in the outcomes of reciprocal exchange. It is assumed that use of power, in the more usual sense, increases until an equilibrium is reached, at which point the less powerful actor is receiving no more benefits than the best alternative source could provide. While power imbalance does in fact increase asymmetry of exchange, it has been consistently found that its use is suboptimal. It is constrained in part by search costs (Molm 1987s) and in part by commitment to the other and concern for equity (Cook and Emerson 1978). Furthermore, use of punishment power depends not only on punishment capability but also on reward capability: Those weaker in reward power are, other things being equal, more likely to use punishment power (Molm 1997). Molm argues they have more to gain and less to lose by the use of coercion than those with more reward power: It is muggers, who gain nothing at all without coercion, who have nothing to lose by using it.

Molm also finds that, given mutual dependence, the effectiveness of coercion is directly proportional to its use, providing that its use is: (1) instrumental rather than symbolic, that is, a specific consequence of a specific behavior rather than a personal challenge; (2) highly contingent on the undesired behavior; and (3) consistent, that is, the punishment of equivalent behavior is equivalent. Coercion is perceived as more symbolic if it is noncontingent, inconsistent, or threatened but not actually used, increasing the side effects of coercion (resistance, negative affect, retaliation, loss of the relation). If it is perceived as instrumental, on the other hand, a high probability of punishment for a specific behavior, because it is effective in the short run, actually increases the net frequency of rewards in the long run, while "weak"—that is, sporadic—coercion decreases the long-run net frequency of rewards and increases the side effects of coercion.

The most significant development in the behavioral exchange theory of power has been its extension to networks of dyadic relations. Interpersonal power is not necessarily confined to a dyadic relation, as Coleman (1973) has shown. But except for Coleman, interpersonal power has usually been treated as dyadic. More complex structures are possible, however, by connecting dyads into networks (Emerson 1972). Study of such networks has rapidly developed as a central preoccupation of research on interpersonal power, especially preoccupation with the question of determinants of power at a position within a network. Most of this research has been concerned with negatively connected networks. A negative connection is one such that exchange in one relation decreases the value of exchange in another, while a positive connection increases its value. Thus, if A must choose which of two offers of exchange to accept from two others, B and C, AB is negatively connected to AC because choice of AB excludes exchange between A and C. Subsequent research has concentrated on the effects of varying structures of negatively connected networks. Centrality of a position turns out not to predict asymmetry of exchange very well; hence, much theory and research has been directed at discovering what alternative concept of power of a position does predict asymmetry of exchange. Cook and Emerson (1978) proposed "vulnerability" as an alternative and defined it as the effect that removing a position would have on the total quantity of resources available for exchange in a network. An intuitively appealing concept, this idea works well for many networks, though Markovsky and colleagues (1988) have objected that there are certain structures for which it does not accurately predict asymmetries. Their alternative conception derives from resistance theory.

Resistance Theory. Willer's (1981) resistance theory objects to the importance of satiation in behavioral exchange theory and as an alternative goes back to Weber's concept of power as overcoming resistance. Willer conceives of interaction as exchange of sanctions. The value of a sanction is a function of its objective value multiplied by its quantity, but value is not itself a function of quantity; hence, there is no satiation effect. Willer focuses instead on the "preference alteration state" of the actor (Pa), consequent on sanctions, and defines resistance as the ratio of two differences: the best possible outcome minus the value of Pa is divided by Pa minus the worst possible outcome B will accept, called Pa at confrontation. Exchange occurs at equiresistance.

The emphasis on resistance leads the theory to incorporate coercion and conflict more easily within the same framework as voluntary exchange. But in addition it leads to a different understanding of power at a position in a network. As in Cook and Emerson (1978), power at a position depends on alternatives, but Markovsky and colleagues (1988) propose a graph-theoretic index of power based on exclusion—B, bidding for exchange with A, is "excluded" if, having no alternatives, B does not exchange at all if outbid by another. The difference between the two is that in Markovsky and colleagues, exclusion is relative to the number of exchanges sought, while in Cook and Emerson actors seek to make a single exchange at a given point in time. It is "advantageous" in resistance theory to be connected to positions that have few alternatives relative to the number of exchanges sought and disadvantageous to be connected to positions that have many alternatives relative to the number sought. In exclusive networks, an odd number of positions is advantageous because in them a position has alternatives or a partner's alternatives have alternatives. An even number is disadvantageous because a position has one or more rivals for exchange with a partner. The ratio of advantageous to disadvantageous paths leading out from a point in a network (counting overlapping paths only once) gives the graph-theoretic index of power at that point in the network. The logic of the argument can be extended to inclusion as well as to exclusion. An inclusive connection is one such that exchange in one cannot occur until exchange in another also occurs. Inclusive relations, for example, increase thepower of peripheral positions in a network. However, in mixes of the two, exclusion overwhelms inclusion and power is as centralized as in exclusive networks (Szmatka and Willer 1995).

Exclusion is more certain in some networks than others, and the magnitude of outcome differentials is sensitive to the certainty of exclusion. Outcome differentials are greater in "strong power" networks, in which the less powerful have no alternatives if they reject the terms of exchange of the more powerful, than in "weak power" networks, in which no position is assured of excluding another without cost to itself (Markovsky et al. 1993). Markovsky and colleagues argue that, in general, the strength of power is inversely proportional to the density of a network because the greater the density of a network, the more alternative opportunities for exchange. Hence, outcome differentials should be inversely proportional to the density of a network. They also show that weak power networks are more sensitive to strategy. Strong power maximizes outcome differentials even if the powerful are passive. In weak power networks, structural forces keep the division of profits more nearly equal, but strategy makes a difference: The more experienced demand more if they are in powerful positions and yield more if they are in less powerful positions, thus magnifying outcome differentials. Weak power networks are also more sensitive to the actual experience of exclusion (Skvoretz and Willer 1993): Potential exclusion, determined by structure, determines outcomes in strong power networks, but (given full information) outcomes are sensitive to both the potential and actual experience of exclusion in weak power networks.

(see also: Decision-Making Theory and Research; Exchange Theory)

references

Bacharach, Samuel B., and Edward J. Lawler 1981 Bargaining: Power, Tactics, and Outcomes. San Francisco: Jossey-Bass.

Bachrach, Peter, and Morton S. Baratz 1963 "Decisions and Nondecisions: An Analytical Framework." American Political Science Review 57:632–642.

Barnard, Chester I. 1938 The Functions of the Executive. Cambridge, Mass.: Harvard University Press.

Blau, Peter M. 1964 Exchange and Power in Social Life. New York: Wiley.

Brehm, Jack W. 1966 A Theory of Psychological Reactance. New York: Academic Press.

Cartwright, Dorwin 1959 "A Field Theoretical Conception of Power." In Dorwin Cartwright, ed., Studies inSocial Power. Ann Arbor: University of Michigan Press.

Coleman, James S. 1973 The Mathematics of CollectiveAction. Chicago: Aldine.

Cook, Karen S., and Richard M. Emerson 1978 "Power, Equity, and Commitment in Exchange Networks." American Sociological Review 43:721–739.

Dahl, Robert A. 1957 "The Concept of Power." Behavioral Science 2:201–215.

Emerson, Richard M. 1962 "Power–Dependence Relations. American Sociological Review 27:31–41.

——1972 "Exchange Theory." In Joseph Berger, Morris Zelditch, and Bo Anderson, eds., Sociological Theories in Progress, vol. 2. Boston: Houghton Mifflin.

Festinger, Leon 1953 "An Analysis of Compliant Behavior." In Musafir Sherif and Milbourne O. Wilson, eds., Group Relations at the Crossroads. New York: Harper.

French, John R. P., and Bertram Raven 1959 "The Bases of Social Power." In Dorwin Cartwright, ed., Studiesin Social Power. Ann Arbor: University of Michigan Press.

Friedrich, Carl J. 1937 Constitutional Government andPolitics: Nature and Development. New York: Harper and Bros.

Goode, William J. 1972 "The Place of Force in Human Society." American Sociological Review 37:507–519.

Harsanyi, John C. 1962 "Measurement of Social Power, Opportunity Costs, and the Theory of Two-Person Bargaining Games." Behavioral Science 7:67–80.

Hawley, Amos 1963 Community Power and Urban Renewal Success." American Journal of Sociology 68:422–431.

Homans, George C. 1961 Social Behavior: Its ElementaryForms. New York: Harcourt Brace Jovanovich.

March, James G. 1955 "An Introduction to the Theory and Measurement of Influence." American PoliticalScience Review 49:431–451.

——1966 "The Power of Power." In David Easton, ed., Varieties of Political Theory. Englewood Cliffs, N. J.: Prentice-Hall.

Markovsky, B., J. Skvoretz, D. Willer, M. Lovaglia, and J. Erger 1993 "The Seeds of Weak Power: An Extension of Network Exchange Theory." American Sociological Review 58:197–209

Markovsky, Barry, David Willer, and Travis Patton 1988 "Power Relations in Exchange Networks." AmericanSociological Review 53:220–236.

Molm, Linda 1987 "Linking Power Structure and Power Use." In Karen S. Cook, ed., Social Exchange Theory Newbury Park, Calif.: Sage.

——1997. Coercive Power in Social Exchange. New York: Cambridge University Press.

Nagel, Jack H. 1968 "Some Questions About the Concept of Power." Behavioral Science 13:129–137.

——1975 The Descriptive Analysis of Power. New Haven, Conn.: Yale University Press.

Parsons, Talcott 1963 "On the Concept of Political Power." Proceedings of the American Philosophical Society 107 (3):232–262.

Russell, Bertrand R. 1938 Power: A New Social Analysis. London: Allen and Unwin.

Samuel, Yitzhak, and Morris Zelditch, Jr. 1989 "Expectations, Shared Awareness, and Power." In Joseph Berger, Morris Zelditch, Jr., and Bo Anderson, eds., Sociological Theories in Progress: New Formulations. Newbury Park, Calif.: Sage.

Simon, Herbert 1953 "Notes on the Observation and Measurement of Political Power." Journal of Politics 15:500–516.

Skvoretz, J., and D. Willer 1993 "Exclusion and Power: A Test of Four Theories of Power in Exchange Networks." American Sociological Review 58:801–818.

Szmatka, J., and D. Willer 1995 "Exclusion, Inclusion, and Compound Connection in Exchange Networks." Social Psychology Quarterly 58:123–131

Tedeschi, James T., Barry R. Schlenker, and Thomas V. Bonoma 1973 Conflict, Power, and Games. Chicago: Aldine.

Thibaut, John W., and Harold H. Kelley 1959 The SocialPsychology of Groups. New York: Wiley.

Thompson, James D. 1967 Organizations in Action. New York: McGraw-Hill.

Weber, Max (1918) 1968 Economy and Society. Guenther Roth and Claus Wittich, eds.; trans. Ephraim Fischoff et al. New York: Bedminster Press.

Willer, David 1981 "Quantity and Network Structure." In David Willer and Bo Anderson, eds., Networks,Exchange, and Coercion. New York: Elsevier/Greenwood.

Zelditch, M., and Ford, J. 1994 "Uncertainty, Potential Power, and Nondecisions." Social Psychology Quarterly 57:64–73.

Morris Zelditch, Jr.