p.1 Over the course of a 15-year period, a group of social exchange theorists (Blau 1964; Emerson 1962, 1972b; Homans [1961] 1974; Thibaut and Kelley 1959) developed a theory of power that differed markedly from prevailing conceptions in the social sciences. Their view of power derived from exchange theory's emphasis on the ties of mutual dependence that underlie all social structures.

p.1 Emerson... argued that power exists in all social relations... that its use involves no necessary intent to harm, coerce, or even influence; and that both parties in a power relation... will act out their respective roles even if they have no awareness of power governing their interaction.

p.2 Emerson's (1962, 1972b) development of the exchange conception of power, commonly referred to as "power-dependence theory," strongly influenced the way that sociologists and other social scientists think about power. Two of his central ideas are now widely accepted among those who study power: that dependency is the source of one actor's power over another, and that power is an attribute of a relation, not an actor.

p.2 Heath (1976) contended that the exclusion of coercion from exchange was an unnecessary restriction. He argued that voluntary (mutual reward) exchanges and coerced exchanges are fundamentally the same and that both could be explained by exchange principles.

p.2-3 Those who control rewards or punishments for others have power over them, derived from others' dependence on them for obtaining things they desire or avoiding things they dislike. Thus, conceptually, the two bases of power are mirror images of each other.

p.5 I study coercive power as part of everyday life... In this research, coercive power is used - or not used - by actors seeking to improve their outcomes in social relations that offer the potential for both reward and punishment.

p.10 if the use of coercive power is not structurally induced, but must be used strategically to create contingencies that influence other actors' behaviors, then what conditions constrain or encourage its use? [JLJ - a good question. I am interested as well.]

p.15 When an actor has possessions or behavioral capabilities that are valued by other actors, they are resources in that actor's relations with those others... social exchange theory reminds us that resources depend on their value to others and, therefore, they are attributes of relations, not actors.

p.29 It was originally Thibaut and Kelley's (1959) insight, however, that the structural condition that provides the basis for social exchange - the mutual dependence of actors on each other - also provides the basis for power.

p.30 power is an attribute of a relation, not an actor, derived from an actor's control over resources that are valued by another actor... power is a potential, inherent in the structure of mutual dependencies created by the differential access that actors have to others who control resources they value... only by defining power as a potential can it be treated as an attribute of an actor's position in a structure of dependence relations... power is nonzero-sum

p.35 The use of power can be structurally induced by power advantage, regardless of actors' intent to use power or to influence another's behavior, or it can be strategic.

p.36 Thus, structurally induced power use results when asymmetries in dependencies produce unequal probabilities of exchange with alternative partners.

p.40 Wrong (1979) argues that for power to exist as a potential, A must have the capacity to affect B's outcomes and B must believe that A has this capacity and would exercise it. The potential for power lies in anticipated reactions. [JLJ - I would argue that power as a concept is useful in forming and pondering scenarios of "what do I do now?" in a complex world, becomes real due to an entanglement of schemes, and ultimately is useful in helping a sentient determine how to go on. Power can appear real, but after careful considering and planning, a situational or clever way to avoid a display of power or its effects could possibly be constructed. Power is ultimately an expression of being - forceful or otherwise, intelligently directed or otherwise, strategically effective or otherwise, communal or solitary in nature, - intended to influence a desired outcome.]

p.47 When people do remain under the coercive control of others, Homans maintained, they do so only if they cannot escape.

p.51 A distinctive feature of many definitions of coercion, both macro and micro, is the requirement of threat... Some theorists define coercive power only as threatened sanctions

p.60 The psychological research on punishment and coercion... shows that coercive power can also be an effective means of influencing others who control desired outcomes.

p.66 I expand the concept of a transaction to include any contingent two-way exchange of outcomes. The reciprocity that is one of the defining characteristics of exchange applies to both rewards and punishments: just as benefits beget reciprocal benefits, so does harm beget reciprocal harm.

p.66 Mutual punishment consists of an exchange of punishing outcomes, again variable in frequency and reciprocity. Many theorists call mutual punishment conflict... coercion is explicitly asymmetrical in nature, involving the potential flow of punishment from one actor and the potential flow of reward from another... A coerces B by punishing B when B fails to reward A, and withholding punishment when B's rewards to A are forthcoming.

p.79 The use of undergraduate students as subjects is a frequent target of criticisms of laboratory experiments.[JLJ - they need the money for beer - you need to publish articles for tenure - what could be a better exchange arrangement?]

p.94 Emerson (1972b) proposed that structural power produces predictable exchange outcomes regardless of actors' information about the power structure or their relative positions in it.

p.101 in lieu of structurally induced power use, coercion must be used strategically

p.139 The incentive to coerce derives... from dependence on another for rewards.

p.140 No intent to use power or to influence another's behavior is necessary; the same structural conditions that provide actors with the capacity to use power also provide the incentive to use power, whether knowingly or not... He [Emerson] argued, in effect, that power use is an unintended by-product of actors seeking their own self-interest in structurally imbalanced relations

p.140-141 The effect of structural power rests on the assumption that actors follow the basic behavioral procedures... by (1) initiating reward exchange with other actors who control resources they value, (2) increasing exchange patterns that are more rewarding and decreasing those that are less rewarding, and (3) changing behavior when rewards decline.
 These principles assume only that actors respond, consciously or not, to the consequences of their own behaviors. Actors need not be aware of their power over others, or make any conscious effort to influence others.

p.141 In the absence of information about the power structure, actors should initially explore exchanges with... partners [JLJ - yes, this is a useful strategy in playing a social game]

p.142 actors are "using power" whenever they withhold rewards from another.

p.143 because structure - not intent - determines power use, he [Emerson] believed the end result would be much the same regardless of whether power use was structurally induced or strategically enacted... two actors... Their relative power positions determine the best outcomes they can hope to obtain, however, and the optimum strategies for achieving those outcomes.

p.144 Emerson was correct in arguing that whether power is induced structurally or used strategically, the result in rewards-based networks will be much the same. Power-imbalanced relations will produce asymmetrical exchange, in favor of the power-advantaged actor... the effective use of power strategies requires that actors apply behavioral principles to influence others... This task requires at least some information about the power structure, close monitoring of the other's behavior, and the consistent, contingent giving and withholding of rewards.

p.145 Now assume, as before, that actors begin by exploring exchanges with all available partners. [JLJ - this seems to be what happens when actors are unaware of the power consequences of their position. When the consequences become known, they merely have to respond to the demands of the position]

p.160 Unlike reward power, coercion is not an unintended byproduct of actors seeking exchange with alternative partners. Instead, it must be used strategically, by actors punishing their partners' failure to reward. The effective use of power strategies requires more information, skill, and behavioral monitoring than structurally induced power use.

p.161 Two important differences distinguish strategic power use from the structurally induced use of power with which Emerson was concerned. First, actors do not merely respond to the consequences of their own behaviors; they create contingencies that produce consequences for other actors' behaviors (Thibaut and Kelley 1959). As the term "strategy" suggests, such actions are typically purposive. With few exceptions, actors who use power strategically impose cost knowingly.
 Second, strategic power use is based not on the principle of reciprocity that underlies power-dependence theory's assumption about the initiation of exchange, but on the application of behavioral principles... actors administer rewards and punishments contingently. In particular, they impose cost (by administering punishment or withholding rewards) on partners who fail to provide sufficient benefits... Actors who use power strategically respond to others' behaviors, but they also attempt to influence them.

p.162 Game theorists attempt to transform uncertainty (unknown probabilities) into risk (known probabilities), by making assumptions about other actors' preferences and motives that allow subjective probabilities to be attached to their behaviors... Most decisions appear to emerge from processes more typical of adaptation, or trial-and-error learning, than rational calculation

p.163 strategies are dynamic and adaptive. Both actors' expectations and their behavioral strategies evolve over time, as they experience the consequences of their own actions and observe their partners' actions. Initial expectations about others' behavior may be based on knowledge of the outcome structure of beliefs formed from past experiences, but these expectations are modified in the light of actual experience. As actors adapt to what each is doing, they mutually shape each other's strategies. Effective strategies are kept, and those that work poorly are discarded or modified.

p.167 strategic power use, like structurally induced power use, should vary with the structure of dependence.

p.170 retaliation is not an irrational response. If an actor can effectively suppress a partner's attempts at coercion, by swift retaliation, she can avoid the costs of both continued coercion and compliance.

p.185 To use punishment to coerce rewards, actors must make their punishment contingent on the other's failure to reward.

p.220 I contend that coercive power will be more effective when it is used frequently and consistently, to punish a partner's failure to provide rewards in exchange, and that increasing the use of coercion will increase the effects of structural variations in punishment power.

p.220 Heath contended that whether the choice was between greater or lesser rewards, or greater or lesser losses, the same principle applies: individuals will choose a better outcome over a poorer one.

p.222 Punishment is most effective when it is combined with rewards, just as reinforcement is most effective when it is combined with punishment

p.229 My analysis, derived from the behavioral principles that underlie exchange theory, predicts that coercion will be more effective if it is used more: the partner's reward exchange should increase as a linear function of the probability of contingent punishment.

p.243 the most important factor for the effectiveness of coercion is its contingency: the probability of punishment for undesirable responses... must be greater than the probability of punishment for desirable responses... The greater the difference, the more effective the coercive strategy. In addition, the probability of punishment for undesirable responses must be consistent

p.244 When coercion is applied consistently and contingently, it is a powerful means of obtaining greater benefits from an exchange partner... inconsistent coercion is ineffective

p.250 strategic power use requires actors to create contingencies of reward or punishment that produce consequences for other actors' behaviors - that is, to use their exchange resources selectively to reward or punish their exchange partners' behaviors. Such actions are typically purposive attempts to change the partner's behavior and improve the actor's benefits from exchange. As a form of strategic power use, coercion is a behavioral solution to an unsatisfactory exchange relation - an effort to increase the partner's rewarding by contingently applying punishment.

p.251-252 To the extent that actors anticipate the consequences of power use, the relative gains or losses attached to these alternative outcomes can affect decisions to initiate power strategies.

p.260 explaining the use of coercive power requires expanding exchange theory's conception of power to include strategic power use.

p.260-261 Coercion... is... behaviorally motivated by conditions that produce insufficient rewards. [JLJ - Perhaps warfare is just the physical side of diplomacy, and the intention is to coerce an opponent to make diplomatic or territorial concessions. We bash our opponent, yelling, "sign this!" until he signs it. Then the war is over. In a social game, coercion seeks as reward, the resignation of an opponent.]

p.261-262 As Heath (1976) argued 20 years ago, coerced exchanges and mutual reward exchanges follow the same logic in their effects: actors will increase the frequency of their reward exchange either to obtain rewards or to avoid punishments. The learning principles that underlie the effects of reward power on exchange also mediate the effects of coercive power on exchange.

p.262 As this analysis suggests, we can bring coercive power within exchange theory's scope by extending its analysis of power to include strategic power use.

p.264 although it is true that coercion is a potentially powerful means of influencing others, it is weak when used inconsistently or under structural conditions that fail to support it.

p.266 coercion is not structurally induced by a coercive power advantage, but purposively enacted as a strategy to increase rewards... actors... Their use of punishment is not an unintended side effect of other actions or other exchanges. Rather, it is a purposive effort to change a partner's behavior, by creating contingencies that produce negative consequences for the partner's failure to reward them.

p.268 When punishment is administered contingently and consistently, coercion is a powerful means of getting what one wants... Because coercion is not structurally induced, but must be used strategically, its effectiveness depends much more on the motivation, skill, and persistence of individual actors... Successful coercion requires diligence in monitoring another's behavior, skill in applying punishment contingently, and the willingness to accept short-term losses... in return for uncertain long-term gains.

p.271 In purely coercive structures, the absence of mutual reward dependence removes risk altogether. Muggers and terrorists obtain no rewards without coercion, and so they lose nothing by using coercion. [JLJ - Molm is referring to the observed behavior of subjects in her study who displayed reluctance to coerce or punish the behavior of others, even if such coercion would lead to (admittedly small) financial rewards. Her point is that muggers or terrorists have no choice but to coerce - they get their rewards this way.]

p.273 To the extent that actors can escape another's punishing actions, coercion becomes impossible

p.287 [instructions to students playing the exchange game that produced the key studies of this book] YOU SHOULD TRY AT ALL TIMES TO MAKE AS MUCH MONEY AS YOU CAN [JLJ - Molm realizes that these are students, who have yet to enter the business world, where after a few years of backstabbing colleagues, greedy bosses, poor-quality work products knowingly delivered to the customers, etc, that this lesson-in-values will already have been learned and put into daily practice. The alternative explanation is that the results from her experiments do not match what Molm wants to publish as theory, so she is coercing her reluctant students into being more aggressive in the game to confirm her soon-to-be-published ideas. Well, what would you expect from someone writing a book, "Coercive Power in Social Exchange", that she would not be coercive to her bored and beer-money-seeking students?]