Turner On Power

European Journal of Social Psychology Eur. J. Soc. Psychol. 35, 122 (2005) Published online in Wiley InterScience (www.interscience.wiley.com).
DOI: 10.1002/ejsp.244
Agenda 2005 Explaining the nature of power: A three-process theory

JOHN C. TURNER*
Australian National University, Australia
Abstract Power is an inescapable feature of human social life and structure. This paper addresses the nature of power. The standard theory is that power is the capacity for inuence and that inuence is based on the control of resources valued or desired by others. However, there have always been problems with this theory and new ones have appeared. The paper summarizes the standard theory and its problems, outlines the different meanings of power and presents a new theory emphasizing group identity, social organization and ideology rather than dependence as the basis of power. It proposes that power is based on persuasion, authority and coercion. A key point is that the theory changes the way these processes have been understood by reversing the causal sequence of the standard theory. The latter argues that control of resources produces power, power is the basis of inuence and that mutual inuence leads to the formation of a psychological group. The three-process theory argues that psychological group formation produces inuence, that inuence is the basis of power and that power leads to the control of resources. Implications of the theory for social change, coercion, prejudice and the extent to which power is a social evil are briey noted. The challenge is to study how power emerges from and functions within social relationships with a denite social, ideological and historical content rather than reifying it as an abstract external force producing generic psychological effects. Copyright # 2005 John Wiley & Sons, Ltd. Few would deny that power is central to human affairs or a key concept in the social sciences. Roberts (2003) argues that it is the basis of society: The ability of one person to make a hundred others do his [sic] bidding is the basic building block upon which all collective human endeavour is based (p. xv). Keltner, Gruenfeld, and Anderson (2003) state that it is a basic force in social relationships (p. 265). It appears to be a universal and indispensable feature of social organization, at work in all political, organizational and institutional life and, in some views, in every social relationship. Every group, organization or society must solve the problems of power to achieve its goals or risk failure, dysfunction or even extinction. But what exactly is power? How should it be dened, what are the processes through which it operates, what are their distinctive effects on attitudes and behaviour, and
*Correspondence to: J. C. Turner, School of Psychology, Australian National University, Canberra, ACT 0200, Australia. E-mail: John.Turner@anu.edu.au Contract/grant sponsor: Australian Research Council; contract/grant numbers: DP0342645.
Copyright # 2005 John Wiley & Sons, Ltd.
Received 25 August 2004 Accepted 25 August 2004
John C. Turner
how is power gained and lost? These questions dene the problem of the nature of power and it is this problem1 I shall seek to address in this Agenda article. Social psychologys rst attempt to solve the problem, which is still the standard theory, emerged from the group dynamics tradition in the 1950s as it translated Lewinian notions of own and induced forces in the lifespace into a detailed analysis of social inuence processes (Festinger, 1953). This theory denes power as the capacity to inuence and argues that it is based upon the inuencing agents control of resources desired or valued by the target. Power is the potential to pret, 1996; French & Raven, inuence and inuence is the exercise of power (Bass, 1990; Fiske & De 1959; Moscovici, 1976; Pfeffer, 1992; Turner, 1991). One consequence is that power has tended to be studied as an adjunct to social inuence rather than in its own right. In recent decades this has changed. Critiques of the dependence theory of inuence have appeared which reject the equation of power with inuence (Moscovici, 1976, 1980; Turner, 1991) and research on power in the context of intergroup relations has provided new perspectives (e.g. Fiske, 1993; Jost & Major, 2001; Reicher & Levine, 1994a, 1994b; Sachdev & Bourhis, 1985; Stott & Reicher, 1998; Sidanius & Pratto, 1999; Turner & Haslam, 2001; Tyler, 2001). It has become possible to redene the nature of power as having more to do with the basis of organized, collective action than to do with a dependence relationship. In this article I try to provide a new way of thinking about power, one more consistent with the currents of contemporary research. I point to problems with the standard theory and then outline a different causal perspective on the origins of power. Whilst what is proposed will bear similarities to standard ideas, the basic causal analysis will be different, with the result that both general implications and specic empirical predictions will differ.
THE STANDARD THEORY OF POWER AND ITS PROBLEMS The standard theory is not one specic formal theory but rather a set of general assumptions about the relationship between power and inuence which is shared by all the classic theories of social inuence (Deutsch & Gerard, 1955; Festinger, 1950, 1953, 1954; French & Raven, 1959; Kelman, 1958). The basic ideas are that power is the capacity to inuence other people, that it is conferred by the control of resources (positive and negative outcomes, rewards and costs, information, etc.) that are desired, valued or needed by others and which make them dependent upon the inuencing agent for the satisfaction of their needs or reaching their goals, and that different types of resources confer different types of power leading to different kinds of inuence. A further important idea widely shared and relevant to the meaning of power is that some kinds of inuence are more informational or cognitive in character, reecting needs for information or to reduce uncertainty, and lead to private acceptance (private attitude change in line with the inuence attempt), whereas other kinds are more social or goal-oriented in character, embodying group pressure or compulsion, and lead only to public compliance. If one of powers meanings is the idea of compulsion, force or domination in contrast to persuasion, then the standard theory that power is the capacity to inuence also contains a sub-theory that outward compliance without inner conviction is a function of particular kinds of non-informational resources (social approval, rewards and costs) controlled by the inuencing agent. The standard theory has been extremely successful. It has been applied to a huge range of phenomena and is still after 50 years the only theory of power that ever gets a mention in social psychology (e.g. see Keltner et al., 2003, for a recent restatement). Nevertheless, there have always
1 I shall address the problem solely from a social psychological viewpoint. It is not possible to consider the contributions of other social sciences in this article, which does not diminish their relevance.
Eur. J. Soc. Psychol. 35, 122 (2005)
Explaining power
been problems with the standard theory and over the last 30 years additional ones have emerged. The most important can be summarized as follows. First, the eld has never agreed on one view about how the dependence-inuence relationship works and the differences between the classic theories are not trivial. The seminal theorists disagree about whether there is one (Festinger, 1950), two (Deutsch & Gerard, 1955), three (Kelman, 1958) or ve, six or more (French & Raven, 1959; Raven, 2001) kinds of inuence process. They disagree about the role of group cohesion in compliance. For Festinger (1950, 1953, 1954) being attracted to the group leads to private acceptance of its inuence by members and public compliance reects dislike for the inuencing group. For Deutsch and Gerard (1955) attraction to the group leads to normative not informational inuence (i.e. compliance). French and Raven (1959) and Kelman (1958) explain compliance in terms of rewards and costs which are distinguished from both informational processes and attraction to others (referent power or identication). Even more fundamentally they disagree about the relationship between informational and social processes of inuence. Festinger (1950, 1954) regards them as inseparable and assumes that people form groups because of their need for information. Deutsch and Gerard (1955) split informational from group inuences, while French and Raven (1959) and Kelman (1958) identify social processes of inuence which are neither informational nor merely compliance (i.e. a kind of acceptance based on emotional ties). Also highly confusing is that the latter theorists both specify rewards and costs as the basis of compliance as if other kinds of power did not also involve rewards and costs. In dependence theory all inuence is shaped by rewards and costs (see Keltner et al., 2003). In informational inuence or social reality testing, for example, one shifts towards people with information or expertise because they provide information which is rewarding in reducing ones uncertainty. How the rewards and costs specied as the basis of compliance in these theories differ from the rewards and costs in other kinds of inuence is never explained. Second, the basic idea that inuence reects dependence is highly problematic (see Moscovici, 1976, and Turner, 1991, for detailed critiques). If one considers the popular dual-process theory of Deutsch and Gerard (1955), for example, which contrasts an informational process with normative inuence in which one conforms to the expectations of others to gain their approval, neither process provides an adequate description of the phenomena they are meant to explain. Private conformity is not a function of people who lack information shifting to people who have it. In fact, even people confronting an unambiguous stimulus (as in the Asch paradigm) become uncertain when similar others disagree with them and what reduces uncertainty is subjectively valid information, not information in the abstract. The perceived validity of information is always a function of social and relational factors such as the perceived source of a message, the degree to which it has consensual support and the degree to which the target denes the source as a positive reference group, that is, the degree to which it is in line with ingroup norms. So-called informational inuence is not purely cognitive but also social and normative. There is in fact no way of dening persuasive or valid information independently of the social context within which it is apprehended. The same information which persuades one group will fail to persuade another. One groups expert is anothers crank. One does not accept inuence from experts because of the information they provide (if one is not an expert, how can one judge its quality?), but accepts the information as valid because one denes them as an expert (Moscovici, 1976). Similarly, the idea that normative inuence in the Asch paradigm is non-informational group pressure is inconsistent with classic ndings such as the importance of unanimity rather than group size in producing conformity and the fact that conformity is still strong even when group surveillance is removed. Basic ndings and indeed basic phenomena (conformity in the classic paradigms, group polarization, minority inuence) do not t and remain unexplained by the dual-process theory (Turner, 1991). These and other considerations imply that neither cognitive nor group aspects of inuence can
Copyright # 2005 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 35, 122 (2005)
John C. Turner
be reduced to dependence for information or social approval. Moscovici (1976) argues that social conict, not informational dependence, is the crucial process in persuasion. Turner (1991) argues that it is reality testing in the context of a shared social identity. A third issue is that the standard theory does not sit easily with the facts of historical and social change. It seems to rule out social change, innovation and minority inuence (Moscovici, 1976). It implies that inuence ows only in one direction, from the top down, from the haves to the havenots, from those with power to those without it. But there are countless historical examples of social movements, from the scientic (Freud) to the political (Hitler) to the religious (Christianity), in which power and control of resources did not precede but followed from processes of inuence. New movements can gain adherents despite often lacking resources, expertise and prestige. The latter are often the badges of success rather than preconditions. If the standard theory is true, then how is it that social change from below ever takes place, how is it that subordinate groups for example, do come, under certain conditions, to reject the legitimacy of the social order and oppose it? A fourth problem already touched on relates to the role of psychological group formation. The standard theory is an important component of the social interdependence analysis of the psychological group (see Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). It is more fully understood as an implicit causal sequence that runs from resources to power to inuence to psychological group formation. It is assumed that control of resources gives people power, dened as the capacity to inuence peoples attitudes, beliefs and behaviour, and that inuence between people who depend upon each other leads to shared social norms and values. It is also assumed that positively interdependent people tend to become cohesive (attracted to each other) and cooperate. Thus mutual dependence between people is assumed to produce a psychological group, a collection of individuals sharing mutual relations of cooperation, cohesion and inuence, which have become stabilized over time into a social structure of roles and norms. In this view, then, power and inuence precede causally and theoretically the formation of a psychological group (Turner, 1991; Turner et al., 1987). But a close examination of the standard conformity paradigms suggests that the psychological group is a precondition of inuence, not simply an outcome. For example, what is experienced as uncertainty, producing a need for information, cannot be dened solely with reference to the objective, physical stimulus situation. Uncertainty arises as a social product of disagreement between people who expect to agree in the given situation. It is where people who expect to agree nd that they disagree that they become uncertain and seek to resolve matters. Self-categorization theory (Turner, 1985) argues that people expect to agree where they dene themselves as members of the same group, in terms of the same social identity, and confront the same shared stimulus situation. If we are the same in relevant respects and judging the same reality, then we ought to agree. On the other hand, if we are different, then there is no reason to experience uncertainty about the validity of ones judgement simply because we disagree (e.g. David & Turner, 1996). Thus group formation is a precondition for the experience of uncertainty which sets in train the processes of informational inuence. It is also a precondition for the experience of subjective validity (condence in the validity of ones views), for it is only when similar others agree that one can feel condent that ones views reect the objective public reality rather than personal biases or systematic error. One accepts information and one seeks approval from those who share ones beliefs, values and objectives, from members of ones ingroups. What this points to is the importance of group formation rather than dependence as the basis of power and the hypothesis that group formation reects identity not dependence. Finally, power and inuence are not properly distinguished in the standard theory. They are confounded in the general view that all inuence reects a process of submission to the power of resources (at the same time as, confusingly, some kinds of dependence are supposed to be more specically to do with submission to power than others because they produce compliance). Nowhere in
Explaining power
the standard theory is there a place for the sharp antagonism between inuence and power commonly found in everyday discourse, as persuasion versus naked force or domination. The attempt to deal with this issue through the idea of compliance based on reward and coercive power (French & Raven, 1959; Kelman, 1958) fails theoretically since all inuence is based on rewards and costs in the dependence formulation. There is also the problem that the conventional concept of compliance does not capture the idea of naked force or compulsion as a strongly conictual process. It is still seen as a form of positive inuence in which the powerless are shaped by the powerful, as a form of going along with the group. It is also worth noting that although coercive power is grouped with reward power as a component of compliance, it functions very differently and produces different results (Festinger, 1953; French, Morrison, & Levinger, 1960). Further, research on attitude change in the reinforcement and dissonance traditions has long shown that both rewards and costs can affect private attitudes, each able to increase or decrease attitude change under specic conditions, which is problematic for this whole approach to compliance (as public conformity). Even in the power literature, rewards and costs can sometimes increase and sometimes decrease private attitude change (e.g. Leet-Pellegrini & Rubin, 1974; Veenstra, Turner, Reynolds, Haslam & Burgess, 2004). Neither is supposed to happen in the compliance formulation. Although compliance based on rewards and costs appears as a commonsense process under the inuence of the standard theory, there is some doubt as to whether the process as dened actually exists. There is also evidence for the idea that what compels outward compliance may ensure private rejection (i.e. attitude change in the opposite direction to that intended; e.g. Brehm & Brehm, 1981; David & Turner, 1996; Moscovici, 1980; Raven & French, 1958; Veenstra et al., 2004). The standard theory barely acknowledges the notion of private rejection since it is inconsistent with its basic thrust that control of rewards and costs provides power and power is the production of intended effects. A theory of the nature of power must explain how and when the demands of the powerful are rejected as well as when they are effective. In sum, in the standard theory all forms of inuence, from the private to the public, reect power based on control of resources and social interdependence for positive outcomes is seen as the basis of the psychological group. But, in fact, there is no unanimity about how different types of resources produce different types of inuence and the general analysis is problematic. Persuasion is assumed to ow from informational dependence on others to satisfy some purely individual need when the very need for and denition of information arises from group and social relationships. Persuasion is not a process of submitting to people with superior resources, but one of negotiating and validating reality collectively within ones reference groups. The standard view also paints a picture of compliance which fails to distinguish between going along with people one likes and being coerced against ones will. All inuence is conceptualized as if it were a kind of submission, moving in the direction dictated by those with resources, because one has no choice, because one is dependent, at the same time as real coercion (force, domination, compulsion) is lost sight of. The standard theory fails to deal adequately with the role of psychological group membership and the facts of social and historical change as well as with the nature of persuasion and coercion. To go beyond the standard theory we need to begin by considering what a theory of power is meant to explain. What do we mean by power?
THE MEANINGS OF POWER pret (1996) comment, we use the Dening power has long been a murky business. As Fiske and De term in ordinary parlance condent that we know what it means until we are asked to dene it. It has
John C. Turner
proved difcult to differentiate consistently from related constructs of inuence, compliance, control, dominance, authority, status and rank. Part of the problem is that there are several legitimate meanings. An adequate theory must be clear about what it is trying to explain and this in turn requires that the various meanings of power be spelled out in relation to each other. This is the task to which we now turn. The most general meaning of power found in the literature and everyday thinking is that it is the capacity to cause effects, to have an impact on or change things, to do work, either in the physical or social world. At this level of abstraction power belongs to things as well as people and affects things as well as people. A sub-category is power as the capacity in some way to affect people or society, to cause them to do things that they would not otherwise have done. A person, institution, physical event or idea may be described as powerful because it is having a big impact on society; it affects what people do and think and how they live. This category can be additionally restricted to effects produced by people, intended or unintended. The idea of a person or group having an impact on others, causing them to think, feel or act in ways they would not have otherwise, I shall term power as impact. A further sub-category is power as the capacity of people to have an intended inuence on others and this is how most theorists have dened social power. It is the denition that permeates the standard theory. It rests on an analogy between the processes by which people inuence others, to produce desired changes in their psychological and behavioural states, and the way in which people can exert some physical power over objects. But there is also a kind of power which only emerges from human social relationships, from the capacity of people to organize themselves into groups, institutions and societies. People acting in concert, cooperating, coordinating and unifying their actions, developing emergent capacities as members of social systems, are able to have an impact on both the physical and social worlds through their interrelatedness that would be impossible if they were purely individual beings. Power from this perspective is the capacity to affect the world, including others, through inuencing and controlling people to carry out ones will, to act on ones behalf, as an extension of oneself. It is social power in the sense that it is a peculiarly social means of having an impact, not because the impact is on people rather than things. The emphasis here is on getting people to act in line with ones desires, beliefs, orders, instructions, commands, suggestions, etc. in order to have an impact on the world through them (see Figure 1). Power to affect the world and power through people are obviously related. The more people, groups, organizations one can get to act as collective embodiments of ones will the bigger the impact one can have on the world and so reference to power through often connotes impact on and vice versa. Part of the general argument to be made is that power through people provides the resources for impact on the world. Whilst accepting this implication, we shall nevertheless for the moment dene power through people as the core social psychological problem of power. How do individuals or groups inuence and control other people to carry out their wishes, suggestions or commands (Roberts, 2003)? The standard theory tends to obscure the difference between interpersonal inuence and both the power to affect the world through collective organization and action and the power to create and direct the collective will of people. It fails to distinguish power through people and impact on them. The former can be seen as a subcategory of the latter logically but psychologically the meaning of the process is very different. The problem of social power is not simply how one affects others, I suggest, but how one gets them to act on ones behalf, to carry out ones will, to become part of a psychologically cohesive social organization. Power through can be subdivided further. There is power as inuence as opposed to control. This is inuence in the sense of persuasion (cognitive change, conversion), the power to get people to act in line with ones desires by persuading them that the desired judgement, decision, belief or action is correct, right, moral, appropriate. If one can persuade others of the correctness of some belief or the
Explaining power
Eur. J. Soc. Psychol. 35, 122 (2005)
Figure 1.
The nature of power through people
John C. Turner
rightness of some action, then they are likely to act on it as a matter of their own volition, as free, intrinsically motivated and willing agents. The capacity to inuence and persuade, to change peoples private attitudes, values and beliefs, is one basis of power through and although often regarded as soft is in fact, I shall argue, the decisive process. Inuence in contrast to control is not the same as the exercise of social power in general but is the much more specic idea of changing peoples private beliefs and attitudes through persuasion. Power as control is here dened as the capacity to get people to do what one wants where they are not persuaded of or are uninterested in the validity of the specic belief or act. I suggest it takes two forms. Legitimate authority is control based on the acceptance by the target of ones right to prescribe their beliefs, attitudes or actions. One may not be persuaded that some action is right but one is persuaded that some person, group or role has the right to prescribe it, by virtue of their position within the group structure, by virtue of the group norms and values which legitimate such control. Such control is based on voluntary deference to and private acceptance of ingroup authority. It is not experienced as an oppression of self or loss of power, but as voluntary submission to a collective will pret, 1995; Tyler & with which one identies and can result in a sense of empowerment and pride (De Degoey, 1995). Finally there is control as coercion, attempting to control others against their will. In this case some individual or group, being unable or unwilling to exert inuence or authority over a target, resorts to deploying the human and material resources it controls to constrain, block, compel and manipulate the targets behaviour without any effort at persuasion or appeals to legitimate authority. Coercion is an unusual or one might say a limiting case. It is power through in the particular sense that it is directed to making the target an extension of ones will on an involuntary basis. It attempts to get the target to act in line with ones own will by forcing them to act against their own, by restricting or opposing the freedom of the target to pursue its own goals. It shades into power as impact, in affecting but not converting the target, and betokens a more or less conictual stance which can rapidly degenerate into open conict. Traditionally the distinctions between persuasion and control and between voluntary and involuntary action have been equated in the contrast between private acceptance and compliance, but in fact these are not the same. Acting without knowing what is the correct response can be voluntary. Being controlled is not the same as acting against ones will. Thus neither control nor coercion as here dened can be equated with the traditional notion of compliance. The latter obscures the crucial process underlying social organization that people are able to act in terms of a collective rather than personal will. Others may prefer different terms to those above but it is the substance of the analysis which matters. All the meanings of power in current use are specied in relation to each other. As we move from power in the most general sense down to coercion, we have illustrated the particular characteristics of each meaning and the level at which it has diagnostic value. Having suggested that the core problem to be addressed is power through, we can now outline and explain the processes of persuasion, authority and coercion by means of a unifying causal scheme.
A THREE-PROCESS THEORY OF PERSUASION, AUTHORITY AND COERCION The proposed causal scheme turns the standard theory completely upside down (see Figure 2). It is power and resource control which are assumed to follow from inuence rather than vice versa. Inuence in turn is seen as a reection of social identity not dependence. The unifying framework here is self-categorization theory, which provides an explanation of psychological group formation in terms of self-categorization rather than dependence (Turner, 1985). The proposed scheme advances
Explaining power
Eur. J. Soc. Psychol. 35, 122 (2005)
Figure 2.
The standard and three-process theories of power
10
John C. Turner
hypotheses which are not only counter-intuitive but paradoxical from the standard perspective. What is supposed to be strong power turns out to be weak and the recipe for success turns out to be a prescription for failure. The theory has three aspects. First, it argues that causality runs in the exactly opposite direction from the standard view, that psychological group formation, understood as the development of a shared social identity, gives rise to inuence between group members, that mutual inuence is the basis of power (the capacity to persuade and/or control others to carry out ones will) and that it is power which enables people to gain and control resources. In the standard theory inuence depends upon power. In this theory power and control of resources depend upon group inuence. Second, it explains that the link between inuence and power needs to be understood in terms of three basic and interrelated processes, inuence itself (here meaning persuasion), authority and coercion. The idea that inuence is the basis of power is both a general proposition and a more specic one with varying meanings depending on the specic process of power being discussed. Third, the theory distinguishes between the different meanings of power in terms of these three processes. Only the general scheme can be put forward in the space available. It is worth noting at the outset, however, that the idea that psychological group formation reects the development of social identity and not dependence for resources is a key proposition of self-categorization theory which has much evidence in its favour. It is now widely accepted that shared social identity plays a crucial role in psychological group formation and in group processes of inuence, cooperation and cohesion (Turner, 1999). This does not mean, however, that group formation is purely subjective and unconstrained by social reality. People become a psychological group as they shift to dening themselves as members of a social category rather than in terms of personal identities as a function of their goals, expectations, background theories, knowledge and ideologies and the social comparisons they make with others within the specic social situation (see Turner & Bourhis, 1996; Turner & Onorato, 1999; Turner, Oakes, Haslam, & McGarty, 1994, for discussion of how group identity forms). Also, the theory proposes that self-categories exist at many different levels of inclusiveness. Thus a point which applies to the self at one level can also be applied to higher and lower levels. The same theoretical points can be made about a leader and a group, a subgroup and an organization, and an organization and society. People inuence and control others through persuasion, authority and coercion. These processes are distinct but can all be dened in relation to group identity and the inuence process which ows from it. Persuasion is the process of inuence outlined in self-categorization theory (Turner, 1987). It is explained as arising from the collective attempt by a group to develop a consensual response to some stimulus situation. Neither group formation nor inuence is based upon resource dependence. It is assumed that where group members agree about some stimulus their judgement is validated by the group consensus and becomes normative. Thus a judgement is assumed to be informational, to provide evidence about reality, precisely to the degree that it has ingroup consensual support. Where members disagree, they become subjectively uncertain, since similar perceivers confronting the same stimulus ought and expect to agree. To reduce uncertainty, which is cognitively inconsistent and impedes condent action, they must either differentiate the stimulus situation (we are talking about different things), categorize themselves as different (our goals are different) or engage in mutual persuasion to reach agreement (if we approach it in this way we can see that our views are actually compatible). Since the perceived validity of any judgement is a function of the degree to which it embodies an ingroup consensus or norm, it also follows that where group members seek to inuence each other they will tend to be more persuasive to the degree that they are perceived as relatively prototypical of the emerging consensus (Turner, 1987). Moreover, the relative prototypicality of a member will vary with group identity which in turn varies with the social comparative context within which the group is dened (Turner et al., 1994). Both the persuasive power of the group and of individuals within it are a function of group identity and consensus.
Explaining power
11
This idea that relative prototypicality is the basis of individual differences in relative inuence or leadership (Turner, Reynolds, Haslam, & Veenstra, in press) has proved extremely important. It was developed to explain group polarization (Turner, 1987), showing how extremists gained in inuence over moderates as the group dened itself in contrast to an outgroup (e.g. McGarty, Turner, Hogg, David, & Wetherell, 1992), has been used to explain how leadership emerges in crowds where there is no organization (Reicher, 1987), how minorities can inuence majorities as they gain in relative prototypicality in certain social contexts (David & Turner, 1999) and how political leaders become persuasive by using rhetoric to shape the degree to which they are perceived as prototypical by an audience of a common identity (Reicher & Hopkins, 1996a, 1996b). More recently, the relative prototypicality hypothesis has generated much research in the area of organizational leadership (see Haslam, 2004, for an overview). This is a very brief summary of the whole analysis, which is of much importance to power. Also highly relevant is the self-categorization work on group consensualization, schisms and changes in the social identity of crowds as a function of the actions of the police (e.g. Haslam, Turner, Oakes, McGarty, & Reynolds, 1998; Sani & Reicher, 1998; Stott & Reicher, 1998). In essence, persuasion and leadership ow from a shared social identity and changes in the relative prototypicality of individuals, groups and roles explain how power can be won or lost independently of the resources that they control. Where persuasion is the power to get people to believe that certain things are correct, authority is the power to control ingroup members because they are persuaded that it is right for a certain person to control them in certain matters. Thus legitimate authority is a product of inuence and the formation of norms within the group. Authority is based on ingroup norms that a person, role or group has the right to prescribe appropriate beliefs, attitudes or behaviour in certain areas. The group agrees (by custom, experience, or formal decision) that they ought to follow a specic person or position (that it has the role of leadership or authority) to express their collective will and short-circuit what might be futile, divisive and time-consuming arguments about what is the right course of action. It is assumed that an individual can gain authority from their consistent and successful leadership where he or she persuades the group of the validity of their point of view over a series of tasks and acceptance of that leadership becomes normative, or that authority can be conferred directly by formal agreement, custom or the norms inherent in group identity. A persons authority varies exibly with the identity, norms and goals of the group, its beliefs and the situation in which it nds itself, in the same way that conformity to any ingroup norm varies. Given an established authority, all the factors which enhance conformity to the group enhance acceptance of authority. Given inuence within the group, any factor which makes an individual more inuential than another will tend over time to confer authority (including their perceived relative prototypicality, Sunshine & Tyler, 2003; Weber, Mummendey, & Waldzus, 2002). Authority is not direct persuasion but groups confer authority in order to get things right. Thus acceptance of leader authority carries a presumption that the leader is likely to be right about the matter in hand and can lead to validation and internalization of the leaders view under certain conditions. Authorities can have the direct right to prescribe private beliefs but this will depend on the nature of the group and the specic authority conferred. Because the power of authority ows from its designation as expressing the collective will of the group, obedience can produce pride and a sense of empowerment when ones collective identity is salient. This analysis is in line with French and Ravens (1959) concept of legitimate power and Tylers (2001) relational model of authority. It agrees with the former that authority is based on social norms and values and with the latter that acceptance of authority is not a social exchange or dependence process (although it does assume that authority serves collective self-interest, that it must get things right to be able to lead effectively). Tylers model stresses the importance of procedural fairness for encouraging identication (respect and pride) with authorities, enhancing
12
John C. Turner
perceived legitimacy and increasing compliance. This seems fully compatible with the present analysis. An advantage of the present framework is that it points to a whole range of other factors relevant to identication with authorities and acceptance of their control as an ingroup norm (e.g. the ideology and goals of group members, the social comparative context, their history of success or failure for the group, the degree to which the authorities are perceived as more or less prototypical of the relevant identity). Coercion is authority in a dark mirror. It is dened here as the attempt to control a target against their will and self-interest through the deployment of human and material resources to constrain and manipulate their behaviour. This process is as close as one gets to the standard theory but is not the standard theory. Coercion here is not French and Ravens (1959) reward and coercive power. It is not inuence based on resource control, but a conictual attempt at control given that one cannot inuence. It is dened more by the conictual relationship than the deployment of resources. Authorities commonly deploy resources to inuence behaviour, which is accepted as legitimate by group members because of the nature of the source. The nature of the power process denes the meaning and value of the outcomes, not vice versa. Being rewarded by an institution one admires is a positive experience just as being punished by a legitimate authority can engender shame and guilt. Being coerced by rewards and costs, however, can easily lead to their subjective re-evaluation as bribes and insults or the glory of a martyrs death by the target so as to frustrate completely the intention of the source. What is positive and what is negative cannot be dened outside of the social and psychological nature of the power relationship. Further, in coercion, deploying resources is at the end of the inuence-power sequence. Coercion depends on inuence and authority in that the source must have inuence and authority over those willing to be its coercive agents (Goldhagens, 1997, willing executioners). A government which directs the police or armed forces to exert force against a target must have inuence and authority over the police or armed forces. History is full of instances where coercion ceases because of loss of inuence and authority over coercive agents by the directing source (e.g. in 1943 in occupied Denmark, the Nazi authorities attempt to extend their nal solution to Danish Jews was a resounding failure as countless Danish citizens refused to comply with the round-up but instead helped the great majority to ee to neutral Sweden in what Kershaw, 2000, describes as the most remarkable rescue action of the war, p. 604). The resources being deployed are themselves nearly always people, organized, with specic expertise, supported by infrastructure and materiel, trained to use specic social procedures and cultural technologies (from the police to mass communications to universities). The actual deployment of resources is only misleadingly described as a process of dependence, as if the target willingly tries to please those that can satisfy his or her needs. Coercion being against the will of the target is an inherently conictual process, to a mild or more extreme degree. Deploying resources here is much more like stopping, physically preventing, attacking, compelling the target, as in a social conict, than like offering a reasonable bargain. Coercion tends to be confused with legitimate resource control (Bass, 1990). The control of resources in coercion depends on power through (the coercive agents) but in relation to the target is actually close to power in the sense of impact on. It affects the target, sometimes in the way intended, but it is not getting the target to act as a voluntary extension of ones will. In fact, it does the very opposite. Coercion, being an attempt to restrict the freedom of the target, makes the target aware of their difference from and disagreement with the source, it leads to disidentication, increased social distance and rejection of the source and private rejection of the position being advocated and hence reduces any residual inuence and authority of the source.2 It
The syndrome of effects hypothesized to follow from coercion in this section are an amalgamation of theory and data from a range of sources. Copyright # 2005 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 35, 122 (2005)
2
Explaining power
13
requires surveillance, which in turn enhances the targets perceived loss of freedom and fosters mistrust (Kramer, 1999). It is a divisive, destabilizing and counter-productive means of control. The more coercion is used the more it must be used, since it undermines inuence and authority and leads to attitude change away from the source at the same time as it provokes resistance and reactance to the loss of freedom (Taylor, 2000). The more it is used the more it brings into being a countervailing source of power as the targets develop a collective identity dened by their rejection of coercion and the goal of defeating the coercive agents who threaten their freedom. The rewards and costs which encouraged obedience at the outset quickly lose their sting as coercion is persisted in and increased. Freedom from coercion becomes more important and whatever advances ones freedom is redened positively, even when before it was a cost. The ineffectiveness of such a strategy should be selfevident. What chance compliance when the target is being motivated to go the other way, evade surveillance and undermine the power of the coercive source at every opportunity? At the extreme, coercion threatens the power of the source itself, since it brings into being an enemy dedicated to its downfall. Coercion is the power one uses when one does not have power. One resorts to power as impact when one does not have power through (inuence and authority in relation to the target), but even here one must have inuence and authority of some kind (over coercive agents). But the paradox is apt. Coercion is the weakest and least effective process of power dened as getting the target to act as an extension of ones will. That the opposite is frequently thought is because coercive impact can be extreme and dramatic and because it always implies power through those that carry it out. Destroying an enemy is a sign of great power in both senses, but it is also a confession of complete failure in terms of inuencing and controlling the actions of the enemy. Returning to the general causal scheme of from group to inuence to power to resource control, it can be elaborated in several ways. Group identity unies and empowers people by giving them a common self-interest and vantage point. It produces inuence and inuence enables a group to act as a unied, coordinated, organized body. Group identity and inuence give people the power of collective action and cooperative endeavour, a power to affect the world and pursue shared goals much greater than any member wields in isolation. Inuence also makes possible authority whereby a group can function effectively and efciently through centralized leadership and control, accepted by the members without the necessity or impossibility of seeking everyones agreement on every relevant matter. Then there is the fact that inuence and authority create the possibility of coercion since only through inuence and authority can the human and material resources be deployed to constrain others against their will. Even where one person confronts just one other, an attempt at coercion reects a judgement of relative power, which is always inuenced by peoples judgements about the likely reactions of others, whether they will support or oppose the actor and target, which in turn is a matter of their identities, norms and values. Finally, there is the critical point that it is inuence, its presence or absence, its effectiveness or not, its intended or experienced use, that determines whether control will take the route of authority or coercion. The same commands can be experienced as either legitimate or coercive. Authority can be and frequently is transformed into coercion and coercion can be and frequently is cloaked in legitimacy and transformed into authority. What matters is the subjective experience of the target. Does he or she feel that the source is seeking to persuade on the basis of a common identity, values and interests, with no ulterior motives, or not? All these are routes whereby inuence is the basis of power. That such power leads to the creation and control of human and material resources has been argued in passing. What are called resources often on closer inspection turn out to be people, organized for particular purposes and subject to some social authority. Even material resources almost never exist independently of the organized bodies of people who use them,
14
John C. Turner
the civil, legal, political and economic institutions which give them meaning and specify the conventions of their use. One dramatic (but not in fact rare) example of the general point must sufce (Cohen, 2003). In 1947 Ben-Gurion, Chairman of the Jewish Agency in Palestine (the state of Israel in waiting), expecting attack from Arab states upon declaration of Israeli independence, undertook a major review of all Jewish military forces and concluded that they were completely inadequate to the waging of full-scale regular warfare against professional armies. The problems were political, organizational, attitudinal, technical and material. Against the opposition of many, he set about using his authority and inuence (and a degree of extreme coercion against selected Jewish targets he dened as undermining his authority and impeding his strategy) to re-organize, train, recruit, manufacture weapons, equip and re-fashion the Israeli armed forces. He led and he mobilized others to follow, setting up programmes to transform the situation. His policies expanded Israeli military forces dramatically. In February 1948 on the eve of Israels independence its forces numbered roughly 16,000; by December 1948 they numbered 92,000. From April 1947 to March 1949 ries went from 10,000 to 60,000, submachine guns from 2,000 to 31,000, artillery pieces from 25 (at the end of 1947) to 500, etc. Israel won its war. Nothing about this expansion was spontaneous or easy. It came from deliberate policy backed up by the power of the group and the authority it gave BenGurion. There is no question that the hard military resources came from and were created by the power of the group and not vice versa. Established power may appear to depend upon resources, but once social conict and instability enter the picture such resources can quickly evaporate or change sides and rising groups create, attract and develop resources as a function of the inuence they exert and the organized activity it makes possible (Pfeffer, 1977, 1992). As relations of identity, ideology, inuence and collective self-interest change, what were dened and created as, for example, the military resources, the tools, of one group can change into their new masters, replacing those who supposedly controlled them (e.g. the New Model Army in the English Civil War, Fraser, 1997, or the Prussian Army in the Great War, Craig, 1968; see Hanson, 2001, for a discussion of the social and cultural nature of military power). The three-process theory turns the standard theory on its head and solves its problems. It is compatible with the facts of and provides many new insights into social change; it unies the informational and group aspects of the inuence process in an analysis supported by much research; it properly acknowledges and species the causal role of the psychological group in inuence and power; and it rejects the confusion inherent in the traditional notion of compliance, sharply differentiating coercive force from the workings of authority whilst showing how they are theoretically linked as divergent reactions to social identity and inuence. A caveat is that it is not being argued that coercion does not have its place in the armoury of leadership. Any group facing complex tasks and changing realities is likely to require all three processes of power to organize its activity and preserve its integrity. The art of leadership is often to coordinate and balance their use. No leader is likely to persuade every single member of the group of the rightness of some action; no authority fails to provoke its critics and dissidents. For the group to function there must be a means of dealing with those who seek to reject or redene core values and pursue interests different from the group as whole. Coercion as a highly selective adjunct to inuence and authority can be vital to preserving the very existence of the group, but as a means of neutralizing and excluding the determinedly recalcitrant, not as an effective means of power through the group (but see below) and never as a substitute for inuence and authority over the group as a whole. The mistake is to follow the standard theory, to confuse the resources provided by power with its basis and wield all leadership as a form of unilateral control.
Explaining power SOME IMPLICATIONS OF THE THREE-PROCESS THEORY Social Change and the Basis of Power
15
The idea that resource control is the basis of power tends to imply that differences in power between individuals and groups are relatively static and enduring. So long as one controls sufcient resources, it seems, one has power and those without resources have little option but to submit. It is difcult to see how power ever changes hands in this view. In practice there are many examples from real life of relatively rapid gains and losses in power where individuals and groups without initial resources become more powerful and those with overwhelming resources suddenly lose power. The three-process theory makes perfect sense in this context since it assumes that power reects group identity and that all self-categorizing is relational and dynamic, varying with social comparisons within and between groups, the specic social context, and the collective goals, values and beliefs of group members. Thus for example, as intergroup relations change, so that cooperation between groups is replaced by conict, then group identity is likely to polarize to emphasize differences from the conicting outgroup and more extreme, conictual members will tend to become more prototypical of each group than will more moderate members. In consequence, the more extreme members will gain in inuence and authority over moderates. History records endless examples of groups which rapidly change their leaders in line with changes in group identity as they enter into conict or conclude a peace with an enemy and where seemingly secure control of resources suddenly evaporates as the collective goals, beliefs and even mood of the group change. The general implication is that power differences in society are constantly shifting and that power change is as normal as stability. The power of leaders and groups depends on identity, organization and ideology and these foundations are always being built up or torn down, being developed creatively or deteriorating in indifference, as a function of partisan interests, collective experience, new tasks and problems and the endless battle between belief and reality. The power of leaders will rise or fall with any factor which makes them more or less representative of ingroup identity and authority. Changes in the collective goals, beliefs, attitudes, circumstances and even mood of the group, in fact any factor which leads the ingroup to dene itself and its collective interests differently, can have implications for which members will have inuence and power irrespective of the resources they control. What matters for social change is that some subordinate group can develop a distinct identity through which to develop its own goals, values and beliefs and contest the power of the hitherto dominant group, that it should come to reject as illegitimate the authority of the social order which subordinates it (Branscombe, Schmitt, & Harvey, 1999; Simon & Klandermans, 2001; Spears, Jetten, & Doosje, 2001; Tajfel & Turner, 1979). The three-process theory helps to understand how this happens (where a dominant group, for example, acts selshly rather than for the collective good, violates core societal values and beliefs, shows disrespect, arrogance and coercive brutality). One aspect of this knot of issues is the process of coercion. The Nature and Effects of Coercion Coercion is the paradigmatic power process in the standard theory. It evokes an image of control which cannot be resisted because it is based on overwhelming resources, an image of dependence in extremis. However, the standard theory gets coercion wrong in reducing it to compliance, which confuses two different forms of control. Once coercion is seen as a distinct process, as proposed here, we nd that it has quite unique effects. Far from being impossible to resist, a useful if unpleasant form of inuence, it is coercion which is the least effective and most counter-productive form of power (understood as getting people to do what one wants). It is, I have suggested, the mirror image of authority. It is
16
John C. Turner
triggered by a lack of inuence and authority and undermines residual inuence and authority even further. The coercion of a target tends to increase social distance from and disidentication with the source, undermine trust and feelings of control, self-respect and pride, induce private rejection of the inuence attempt and engender resistance and reactance in which the goal of rejecting control and restoring freedom of action gradually breeds social conict with the intention of frustrating control. Coercion persisted in, without the cloak of legitimacy, tends to produce private attitude change away from the coercer, reactance, conict and the emergence of a countervailing force. Ultimately, coercion creates its own enemy. Surveillance to maintain control like all other strategies aimed at manipulating and constraining the will of the target likewise frustrates their own purpose (Veenstra et al., 2004). They create an enemy where there could have been an ally and ensure a waste of resources to try to get an enemy to do what one is persuading them (through negative inuence) that they should not. How much more efcient to change private attitudes, so that others work willingly, actively and freely, requiring no burdensome apparatus of external control, to achieve the ends that one desires? The weakness of coercion is in trying to use it as a form of inuence. Selective coercion can certainly be useful if the aim is to destroy, impede or constrain an enemy, but this is more like power as impact than power through. The problem arises when people confuse the two on the basis of the standard theory and produce the very opposite of what they intended. But coercion in any form is a threat to the power of the coercer as we shall discuss presently. This analysis of the effects of coercion raises the questions of when coercion is likely to be effective as control and when it is likely to produce at the same time private rejection and reactance. Coercion is hypothesized to be more effective (albeit in a limited, short-term fashion) where the response required is minimally self-dening (ego-involving), the target individualized, the resources deployed are detached as far as possible from any perceived coercive intent (e.g. presented as objective consequences rather than deliberate human pressure), are positive rather than negative, and cloaked in whatever legitimacy is plausible. The general rule is where possible to disguise coercion as legitimate authority and minimize the threat to the targets perceived freedom. Conversely, the more that any coercive act is perceived as against ones own will and interests, emanating from an opponent rather than an ally or partner, and embodies an intended threat to ones freedom of action, the more surely will it produce private rejection and reactance and over time foster the contrasting collective identity and countervailing beliefs that produce sustained oppositional power and collective conict. Since this theoretical conception of coercion has never been dened in any previous theory of power these hypotheses await systematic testing (but see Kramer, 1999; Reicher & Levine, 1994a, 1994b; Veenstra et al., 2004). It will be important to test them in a temporal framework, showing that repeated coercion experienced as such does gradually produce an oppositional stance and identity in the target which erodes the power of the leader and that paradoxically the authority of the leader could disappear more rapidly the more resources and hence the more perceived coercive threat he or she displays. An analogy to these ideas is provided by historical studies suggesting that the deployment of great military power freed from all civil and political restraint (Cohen, 2003; Craig, 1968) or not based on the civic spirit and loyalty of the soldiers (Hanson, 2001) has often been the speediest route to military disaster as political support has disappeared, cohesion and discipline evaporated and enemies multiplied.
Prejudice The present analysis proposes a novel explanation of the link between power and prejudice. It suggests that prejudice can be seen as a strategy for maintaining and enhancing the power of a dominant group
Explaining power
17
in society whilst it undertakes coercion of some other group. Where one coerces a target, I have suggested, the result is rejection and resistance by the target, ones inuence and authority over the target is directly undermined. If the experience of a specic target becomes a collective experience of ones group or society, then coercion will directly undermine ones power over the group or society. But it follows from self-categorization theory that any human being can identify with another and experience their emotions collectively on any number of bases and any feelings of empathy, sympathy, of shared social identity, with the target by the group will therefore undermine the authority of the coercive leadership over the group which is the source of its power. It is obvious that where coercion is exercised against any dissident subgroup by coercive agents who are members of the same society or subscribe to belief in a common humanity, there is a real danger that the agents will identify with their victims unless this outcome is prevented. Coercion can therefore only be exercised by a dominant group without undermining its own power by negatively categorizing and stereotyping the target as different from the group as a whole, to break down and prevent any identication with the former by the latter. This is especially necessary where the target is a dissident member or subgroup of the very group which is the source of ones power. It is most important that those carrying out the coercion feel no identication or they may be transformed from agents to opponents of ones power. This strategy can be taken a step further. If one can categorize the target as a threat, danger, enemy to the group as a whole, creating an implicit or explicit intergroup conict, and involve other group members psychologically in the coercion of the target, then, far from undermining ones authority, coercion may work to enhance ones legitimacy and power, appearing as if for the collective good. At the extreme, the strategy of demonizing the target to legitimize its coercion can become a strategy of creating and achieving power over the group by inventing or nding a common enemy that redenes the group identity to place one at its core and ones opponents beyond the pale (thus Hitlers invocation of race and the Jewish threat to replace the class ideology of his social democratic and Marxist opponents). Far from being merely an expression of power, therefore, prejudice can provide and maintain power. This is related to but different from the idea of prejudice as a rationalizing justication of power differences or as scapegoating to avoid blame or that creating intergroup conict necessarily enhances loyalty to leaders. This is prejudice to avoid the destabilizing effects of coercion or to gain power through redening group identity. This analysis is in the intergroup conict of interest tradition in that it takes for granted that a dominant group which has power to pursue its interests in society as a legitimate authority will nevertheless to some degree always face opponents from subordinate groups who reject its policies and which it must deal with through coercion. The prejudice does not cause the coercion, which is undertaken for reasons of political and economic self-interest (slavery, for example, was a highly protable institution for the slave-owners). The prejudice is used by the dominant group to deal with the political dangers to its authority which would arise in society from those who would be opposed to its coercion but for the prejudice. Thus it is the coercion which makes the prejudice necessary and politically the prejudice is directed not just against its victims but also against any of the dominant groups own supporters whose loyalty might be less than rock solid. A power analysis of prejudice requires that it be put in the context of a multi-group social system in which there is a constant struggle between groups to dene society in such a way as to strengthen their own power and position.
The Evils of Power Power has had a bad press. It is believed to corrupt, lead to abuse, stereotyping and prejudice (Keltner et al., 2003; Lee-Chai & Bargh, 2001). These notions ow from the confounding of power with the dominance/submission relationship implicit in dependence and the failure to distinguish coercion
18
John C. Turner
from other processes in the standard theory. I suggest that power as persuasion or authority is neither oppressive nor maladaptive nor inherently detrimental to social cohesion. The very opposite in fact since the coordination and centralization of group action through legitimate authority empowers group members to achieve their goals. For a group to pursue its collective will there must be a power structure through which group identity and goals are realized. The issue is what kind of power process is at work? Legitimate authority should not reduce members feelings of power since it expresses a collective identity, not a perceived hierarchical difference in which members are subordinated. Authority can degenerate into coercion, just as coercion can be transformed into authority, and it is important to study how these changes occur. But it is also important to recognize that the absence or weaknesses of power structures can be highly frustrating and detrimental to social cohesion and effective social organization (Pfeffer, 1992; Reicher & Haslam, 2004). The three-process theory provides a useful analysis for understanding how authority can degenerate and lead to abuse, since authority provides the opportunity and temptation for reliance on coercion and at the same time may under certain conditions encourage those in authority to differentiate themselves from followers and develop partisan, separate interests. It is easy to see how the nature of coercion is likely to brutalize the authority that wields it and produce a cycle of conict and even violence in which mutual hatred, delegitimation and self-justication thrive (Bar-Tal, 1990). It is also easy to see how a dominant group which has developed interests which conict with those of subordinate groups and society as a whole will engage in deception, manipulation, coercion and terror to maintain its position since by denition it cannot appeal to the common good other than dishonestly. But if one considers whether there is anything psychologically inevitable about the corrupting effects of power on those that wield it, some caution is in order. The theories which imply such generic psychological effects all adopt, behind the supposed variety of inuence, one narrow social vision of the power relationship, that of power as dominance and submission based on dependence (e.g. Keltner et al., 2003). But power cannot be reduced to one such process; it takes a variety of forms. Power is an emergent property of human social relationships, not something that stands outside of them, and power relations can take as many concrete and nuanced forms as the social relationships they express. If one looks at the evidence for the supposed generic effects of holding power we nd that these effects are either restricted to just one kind of power process or disappear as soon as the values, beliefs, goals of pret & Fiske, 1999; the powerful are varied and/or the nature of the social relationship is varied (De Kipnis, 1972; Lee-Chai, Chen, & Chartrand, 2001; Overbeck & Park, 2001; Reynolds, Oakes, Haslam, Nolan, & Dolnik, 2000; Rind & Kipnis, 1999; Stott & Drury, 2004; cf. Ellemers, van Rijswik, Bruins, & de Gilder, 1998). Kipnis (1972), for example, denes power as something like coercion in our terms in contrast to a supposed no-power condition of persuasion and authority. He actually shows therefore that the effects of holding power vary with the type of power process (as do Rind & Kipnis, 1999). How people react to holding power is not just a function of some abstract relation of dominance, but of the identity, beliefs, values and theories of the people involved and the political and social content of their relationship. Leadership can be about dedication, responsibility, service and self-sacrice as well as privilege and rewards, working for the collective good, not greed (e.g. Lee-Chai et al., 2001; Overbeck & Park, 2001). The beliefs, theories and values of the group dene what is legitimate and what unjust, what is appropriate organization and procedure, what is proper accountability, how leader-follower relations should be organized and conducted, what should be tolerated and what should be deed. Thus the issue of corruption, of when it is perceived, of what makes it acceptable or unacceptable, whether it can or even should be prevented, has a social, political, historical and ideological dimension, which is in fact always at work in the experimental demonstration of supposedly generic effects.
Explaining power
19
A strength of the three-process theory in this respect is its focus on group identity, organization and ideology as bases of power, since the emergence of immorality and abuse in its exercise cannot be divorced from these things. From a social psychological perspective, no power is absolute, all power is constrained, both what can be done and what should not be done reect the substantive beliefs and values of people with specic identities in a denite place and time. These brief remarks hardly do justice to the issue of the evils of power, but they do illustrate the rich possibilities of a fresh approach to power in considering a question of such import.
CONCLUSION The standard theory of power tends to conate all the varieties of inuence into one relationship of dependence for resources. Power is provided by control of resources and those without such resources are in the power of those who have them. In this view power is like some abstract commodity which exists in principle outside of social relationships and which can be used by the lucky few to impose their will on such relationships. I have tried to suggest that this is a reication of what is actually going on. Power is an emergent property of specic social and psychological relations between people and these relations shape the form it takes. It emerges from group formation, social organization and the shared beliefs, theories and values (the culture, ideologies, etc.) which shape social and personal identity and perceived self-interest. These are the foundations of persuasion, authority and coercion, the processes by which people are able to get others to carry out their will and have a collective impact on the world. An implication is that power relations always have a denite social, relational and ideological content which directly affects how power is gained, lost and used. Leaders gain power not by possessing resources in the abstract but by standing for, representing, believing, working for something, or being perceived to do so, with which at least some others concur. Power is never absolute but always socially constrained and conferred. The social psychological study of power therefore must put it into the context of the group and societal relationships from which it emerges and the theories which deal with such relations.
ACKNOWLEDGEMENTS I would very much like to thank Kate Reynolds, Kris Veenstra, Alex Haslam, Steve Reicher and Natalie Taylor for their help in writing this article and the Australian Research Council for funding support (DP0342645).
REFERENCES
Bar-Tal, D. (1990). Causes and consequences of delegitimization: Models of conict and ethnocentrism, Journal of Social Issues, 46, 6581. Bass, B. M. (1990). Bass & Stogdills Handbook of leadership: Theory, research, and managerial applications (3rd ed., pp. 225273). NY: The Free Press. Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving pervasive discrimination among African Americans: Implications for group identication and well-being. Journal of Personality & Social Psychology, 77, 135149.
20
John C. Turner
Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. NY: Academic Press. Cohen, E. A. (2003). Supreme command: Soldiers, statesmen and leadership in wartime. London: Simon & Schuster. Craig, G. A. (1968). The politics of the Prussian Army 16401945. London: Oxford University Press. David, B., & Turner, J. C. (1996). Studies in self-categorization and minority conversion: Is being a member of the outgroup an advantage? British Journal of Social Psychology, 35, 179199. David, B., & Turner, J. C. (1999). Studies in self-categorization and minority conversion: The ingroup minority in intragroup and intergroup contexts. British Journal of Social Psychology, 38, 115134. pret, E. (1995). Vicarious feelings of personal control and the social categorization of powerful others. De Unpublished paper. University of Grenoble. pret, E., & Fiske, S. T. (1999). Perceiving the powerful: Intriguing individuals versus threatening groups. De Journal of Experimental Social Psychology, 35, 461480. Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social inuences upon individual judgment. Journal of Abnormal & Social Psychology, 51, 629636. Ellemers, N., van Rijswik, W., Bruins, J., & de Gilder, D. (1998). Group commitment as a moderator of attributional and behavioural responses to power use. European Journal of Social Psychology, 28, 555573. Festinger, L. (1950). Informal social communication. Psychological Review, 57, 271282. Festinger, L. (1953). An analysis of compliant behaviour. In M. Sherif, & M. O. Wilson (Eds.), Group relations at the crossroads (pp. 232256). NY: Harper & Bros. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117140. Fiske, S. T. (1993). Controlling other people: The impact of power on stereotyping. American Psychologist, 48, 621628. pret, E. (1996). Control, interdependence and power: Understanding social cognition in its social Fiske, S. T., & De context. European Review of Social Psychology, 7, 3161. Fraser, A. (1997). Cromwell: Our chief of men. London: Arrow Books. French, J. R. P., & Raven, B. H. (1959). The bases of social power. In D. Cartwright (Ed.), Studies in social power (pp. 150167). Ann Arbor, MI: Institute of Social Research. French, J. R. P., Morrison, H. W., & Levinger, G. (1960). Coercive power and forces affecting conformity. Journal of Abnormal & Social Psychology, 61, 93101. Goldhagen, D. J. (1997). Hitlers willing executioners: Ordinary Germans and the Holocaust. London: Abacus. Hanson, V. D. (2001). Why the West has won: Carnage and culture from Salamis to Vietnam. London: Faber & Faber. Haslam, S. A. (2004). Psychology in organizations: The social identity approach (2nd ed.). London: Sage. Haslam, S. A., Turner, J. C., Oakes, P. J., McGarty, C., & Reynolds, K. J. (1998). The group as a basis for emergent stereotype consensus. European Review of Social Psychology, 8, 203239. Jost, J. T., & Major, B. (2001). The psychology of legitimacy. NY: Cambridge University Press. Kelman, H. C. (1958). Compliance, identication, and internalization: Three processes of attitude change. Journal of Conict Resolution, 2, 5160. Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition. Psychological Review, 110, 265284. Kershaw, I. (2000). Hitler 193645: Nemesis. London: Penguin. Kipnis, D. (1972). Does power corrupt? Journal of Personality & Social Psychology, 24, 3341. Kramer, R. M. (1999). Trust and distrust in organizations. Annual Review of Psychology, 50, 569598. Lee-Chai, A. Y., & Bargh, J. A. (2001). The use and abuse of power. Philadelphia, PA: Psychology Press. Lee-Chai, A. Y., Chen, S., & Chartrand, T. L. (2001). From Moses to Marcos: Individual differences in the use and abuse of power. In A. Y. Lee-Chai, & J. A. Bargh (Eds.), The use and abuse of power (pp. 5774). Philadelphia, PA: Psychology Press. Leet-Pellegrini, H., & Rubin, J. Z. (1974). The effects of six bases of power upon compliance, identication, and internalization. Bulletin of the Psychonomic Society, 3, 6870. McGarty, C., Turner, J. C., Hogg, M. A., David, B., & Wetherell, M. S. (1990). Group polarization as conformity to the prototypical group member. British Journal of Social Psychology, 31, 120. Moscovici, S. (1976). Social inuence and social change. London: Academic Press. Moscovici, S. (1980). Towards a theory of conversion behaviour. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 13, pp. 209239). NY: Academic Press. Overbeck, J. R., & Park, B. (2001). When power does not corrupt: Superior individuation processes among powerful perceivers. Journal of Personality & Social Psychology, 81, 549565.
Eur. J. Soc. Psychol. 35, 122 (2005)
Explaining power
21
Pfeffer, J. (1977). Power and resource allocation in organizations. In B. M. Staw, & G. R. Salancik (Eds.), New directions in organizational behaviour (pp. 235265). Chicago, Ill: St. Clair Press. Pfeffer, J. (1992). Managing with power. Boston, Mass.: Harvard Business School Press. Raven, B. H. (2001). Power/interaction and interpersonal inuence: Experimental investigations and case studies. In A. Y. Lee-Chai, & J. A. Bargh (Eds.), The use and abuse of power (pp. 217240). Philadelphia, PA: Psychology Press. Raven, B. H., & French, J. R. P. (1958). Legitimate power, coercive power, and observability in social inuence. Sociometry, 21, 8397. Reicher, S. (1987). Crowd behaviour as social action. In J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S. Wetherell, Rediscovering the Social Group: A self-categorization theory (pp. 171202). Oxford: Blackwell. Reicher, S., & Haslam, S. A. (2004). Rethinking the psychology of tyranny. University of St. Andrews. In submission. Reicher, S., & Hopkins, N. (1996a). Constructing categories and mobilising masses: An Analysis of Thatchers and Kinnocks speeches on the British miners strike 19845. European Journal of Social Psychology, 26, 353 371. Reicher, S., & Hopkins, N. (1996b). Seeking inuence through characterising self-categories: An analysis of antiabortionist rhetoric. British Journal of Social Psychology, 35, 297312. Reicher, S., & Levine, M. (1994a). Deindividuation, power relations between groups and the expression of social identity: The effects of visibility to the out-group. British Journal of Social Psychology, 33, 145 163. Reicher, S., & Levine, M. (1994b). On the consequences of deindividuation manipulations for the strategic communication of self: Identiability and the presentation of social identity. European Journal of Social Psychology, 24, 511524. Reynolds, K. J., Oakes, P. J., Haslam, S. A., Nolan, M., & Dolnik, L. (2000). Responses to powerlessness: Stereotyping as an instrument of social conict. Group Dynamics: Theory, Research and Practice, 4, 275290. Rind, B., & Kipnis, D. (1999). Changes in self-perceptions as a result of successfully persuading others. Journal of Social Issues, 55, 141156. Roberts, A. (2003). Hitler and Churchill: Secrets of leadership. London: Weidenfeld & Nicolson. Sachdev, I., & Bourhis, R. Y. (1985). Social categorization and power differentials in group relations. European Journal of Social Psychology, 15, 415434. Sani, F., & Reicher, S. (1998). When consensus fails: An analysis of the schism within the Italian Communist Party. European Journal of Social Psychology, 28, 623645. Sidanius, J., & Pratto, F. (1999). Social dominance. Cambridge: Cambridge University Press. Simon, B., & Klandermans, B. (2001). Politicized collective identity. American Psychologist, 56, 319331. Spears, R., Jetten, J., & Doosje, B. (2001). The (il)legitimacy of ingroup bias: From social reality to social resistance. In J. T. Jost, & B. Major (Eds.), The psychology of legitimacy (pp. 332362). NY: Cambridge University Press. Stott, C., & Drury, J. (2004). The importance of social structure and social interaction in stereotype consensus and content: Is the whole greater than the sum of its parts? European Journal of Social Psychology, 34, 1123. Stott, C., & Reicher, S. D. (1998). Crowd action as intergroup process: Introducing the police perspective. European Journal of Social Psychology, 28, 509530. Sunshine, J., & Tyler, T. R. (2003). Moral solidarity, identication with the community, and the importance of procedural justice: The police as prototypical representatives of a groups moral values. Social Psychology Quarterly, 66, 153165. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conict. In W. G. Austin, & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 3347). Monterey, California: Brooks/Cole. Taylor, N. (2000). When disagreement strikes: Incorporating compliance, conversion, resistance and reactance into a model of social response. PhD thesis. Australian National University. Turner, J. C. (1985). Social categorization and the self-concept: A social cognitive theory of group behaviour. Advances in Group Processes, 2, 77122. Turner, J. C. (1987). The analysis of social inuence. In J. C. Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S. Wetherell (Eds.), Rediscovering the social group: A self-categorization theory (pp. 6888). Oxford: Blackwell. Turner, J. C. (1991). Social inuence. Milton Keynes, UK: Open University Press. Turner, J. C. (1999). Some current issues in research on social identity and self-categorization theories. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social identity: Context, commitment, content (pp. 634). Oxford, UK & Cambridge, USA: Blackwell.
Eur. J. Soc. Psychol. 35, 122 (2005)
22
John C. Turner
Turner, J. C., & Bourhis, R. Y. (1996). Social identity, interdependence and the social group. In W. P. Robinson (Ed.), Social groups and identities: Developing the legacy of Henri Tajfel (pp. 2563). Oxford: Butterworth Heinemann. Turner, J. C., & Haslam, S. A. (2001). Social identity, organizations and leadership. In M. E. Turner (Ed.), Groups at work: Advances in theory and research (pp. 2565). Hillsdale, NJ: Erlbaum. Turner, J. C., & Onorato, R. (1999). Social identity, personality and the self-concept: A self-categorization perspective. In T. R. Tyler, R. M. Kramer, & O. P. John (Eds.), The psychology of the social self (pp. 1146). Mahwah, NJ: Lawrence Erlbaum Associates. Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Basil Blackwell. Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. A. (1994). Self and collective: Cognition and social context. Personality & Social Psychology Bulletin, 20, 454463. Turner, J. C., Reynolds, K. J., Haslam, S. A., & Veenstra, K. (in press). Reconceptualizing personality: Producing individuality by dening the personal self. In T. Postmes, & J. Jetten (Eds.), Individuality and the group: Advances in social identity. London: Sage. Tyler, T. R. (2001). Social justice. In R. Brown, & S. Gaertner (Eds.), Handbook of social psychology: Intergroup processes (Vol. 4, pp. 344364). Oxford, UK: Blackwell. Tyler, T. R., & Degoey, P. (1995). Collective restraint in social dilemmas: Procedural justice and social identication effects on support for authorities. Journal of Personality & Social Psychology, 69, 482497. Veenstra, K., Turner, J. C., Reynolds, K. J., Haslam, S. A., & Burgess, N. M. (2004). Power and social identity: The differential effects of ingroup and outgroup leadership on power and compliance. Australian National University. In submission. Weber, U., Mummendey, A., & Waldzus, S. (2002). Perceived legitimacy of intergroup status differences: Its prediction by relative ingroup prototypicality. European Journal of Social Psychology, 32, 449470.
Eur. J. Soc. Psychol. 35, 122 (2005)

Turner On Power

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Turner On Power

Uploaded by

Copyright:

Available Formats

European Journal of Social Psychology Eur. J. Soc. Psychol. 35, 122 (2005) Published online in Wiley InterScience (www.interscience.wiley.com).

Agenda 2005 Explaining the nature of power: A three-process theory

Copyright # 2005 John Wiley & Sons, Ltd.

Received 25 August 2004 Accepted 25 August 2004

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

The nature of power through people

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

The standard and three-process theories of power

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

Copyright # 2005 John Wiley & Sons, Ltd.

Eur. J. Soc. Psychol. 35, 122 (2005)

You might also like