Power over, power to and power with. Faces or forms of power?

Power over, power to and power with. Faces or forms of power? Pamela Pansardi University of Pavia IPSA World Congress 2016 - Poznan DRAFT - PLEASE DO NOT CITE WITHOUT PERMISSION Abstract. The expressions power over, power to and power with have recently drawn the attention of scholars interested in the analysis and evolution of power relations in both national and international politics. In this paper, I attempt an analysis of the three interpretations of the concept of power that aims at exploring the meaning and relevance of the three expressions. Different aspects will be taken into account: the reliance on normative or empirical criteria used for the distinction of the three concepts; their relations with other classifications of power, namely the distinction between hard, soft and smart power; lastly, the reasons according to which they should be considered as different faces of power, i.e., as different descriptions of a situation by a relevant point of view, or whether they should be analytically classified as consisting in different forms of power, i.e., as different ways in which power is exercised. The paper will accordingly attempt to investigate whether any single and unambiguous definition of the three concepts can be proposed. Introduction The attention to the distinction between the notions of power over, power to and power with is growing within the contemporary literature on power. While the success of the distinction is increasing, the clarity of the criteria that indicate what the distinction is about is concurrently blurring. Aim of this paper will be, on the one hand, to reflect on different interpretations of the distinction in order to shed some light on the different ways in which the three notions are currently used. On the other hand, it will consist in 1

the positive proposal of an interpretation of the distinction able to shed light on common misunderstandings and misuses of the three notions. Power to and power over The expressions power over and power to have appeared quite recently in theoretical debates about power. Their usage conforms, at a very general level, to the distinction between two interpretations of power: power over refers to an asymmetrical relation between two or more actors, while power to consist in the ability of the actor herself to carry out certain specific outcomes. Although different theoretical perspectives propose different criteria for the definition of the two concepts, it is possible to identify some common traits. The various usages of power over all agree in interpreting it as a relation between actors, specifically as a relation of social causation. A lot as been written about the connection between the concept of power and that of social cause: for some authors the two concepts share the same extension, according to others, power is instead to be considered the specific kind of social causation resulting from the intentions of the power-wielder. Power to is also explicated in terms of causation. However, the difference lies in the fact that while power over means causing behavior on the part of others, power to refers to the ability to cause certain outcomes or states of affairs. Moreover, power to is a dispositional concept: it refers to the capacity an actor possesses to bring about specific outcomes, not to her action of producing those results. By contrast, the concept of power over is described as both actual and potential: the phenomenon of power is identified both with the fact of an agent actually affecting the behavior of another, and with her having the ability to do so at some future time. Lastly, while power to is commonly interpreted as a property of individuals or of groups, power over has been attributed, by different approaches, to individuals, collectivities, institutions and social structures. It is worth noting that, though the origin of the distinction lies in theoretical efforts to clarify the very meaning of the concept of power, the expressions power to and power over have recently enjoyed an new lease of life within critical approaches and feminist theories, serving to indicate two distinct forms of power. In the latter cases 2

the distinction between the two expressions relies on normative criteria: while power over is thought of as necessarily conflictual and is mostly used as a synonym for domination, power to is regarded as a consensual and intrinsically legitimate form of power. Among the various contributions to the study of power, power over is undoubtedly the predominant interpretation. Conceptual analyses that are normally indicated with the expression power over define it as a relationship in which one actor is able to obtain a compliant behavior from another actor. They make reference to conceptions of power which, following the Weberian tradition, regard it as equivalent to social control and influence. In the contemporary literature about power, probably the most influential definition is the one offered by Robert Dahl, who claims that A has power over B to the extent that he can get B to do something that B would not otherwise do (1957, 202-203). In Dahl s account, power is assimilated to an actual or potential relation of social causation: it consists in a relation in which an actor intentionally successfully acts or is able to act to modify the behavior of another actor. The same interpretation of power in terms of power over is assumed by all the participants in the debate that Steven Lukes labeled the Three Dimensional Views of Power (1974, 5). In Power: A Radical View, Lukes analyzes Dahl s conception of power, together with that of two of his most important critics, Peter Bachrach and Morton Baratz. He labels them respectively the one and the two-dimensional views of power, and he himself proposes the so-called threedimensional view. Despite the differences among the three approaches, Lukes claims that they share a common core : they all consist in different applications of the same underlying concept of power, according to which A exercises power over B when A affects B in a manner contrary to B s interests (1974, 26-27). They diverge, instead, on the relevant interpretation of interests : while Dahl focuses his analysis on subjective interests understood in terms of preferences expressed through political participation, the second view also includes those subjective interests which are excluded from the political debate, and Lukes goes even further, focusing, in addition, on the real (or objective) interests of which individuals are not aware. Lukes analysis, then, explicitly highlights the conflictual nature of power. Moreover, it assumes power to be an intrinsically 3

evaluative concept: if exercising power means acting against the interests of someone, the concept of power virtually collapses into the concept of harming. Power over, then, in Lukes account, becomes a synonym of domination. It is worth noticing that in the second edition of Power: A Radical View, Lukes partially changes his mind. He recognizes that power over should be interpreted as a broader concept than that of domination (2005, 84): some exercises of power can play in favor of the interests of the subjects, or at least not damage them, as in the cases, for example, of mothering or teaching. Concerning the nature of the power-wielder, Lukes analysis consists in an attempt to bridge the gap between methodologically individualist and structuralist approaches to power. He suggests that power is not necessarily wield intentionally, but that it can be exercised by actors who are unaware of the consequences of their actions, or even through inaction. According to Lukes, then, locating power should consist, in these cases, in fixing responsibilities, i.e., in the identification of those who could have acted otherwise. The main interpretation of power in terms of power over has been challenged by a number of accounts aiming at clarifying, through the reconstruction of the etymological origin of the term and its current usage in ordinary language, the very meaning of power. The introduction of the expressions power over and power to is due to Hanna Pitkin, who, in the same years as those of the three-dimensional debate, offered an understanding of power in terms of ability (1972, 276). Pitkin underlines the need to distinguish between those situations in which an actor has power over another actor, which occurs only when that actor gets the other to do something, from those situations in which an actor has the power to accomplish something by herself. This second case, labeled power to, corresponds, in her view, to the prevailing use of the term. The very meaning of power, then, in her account, is to be interpreted in terms of being able to. The most comprehensive and influential analysis of power in terms of power to, however, is the one proposed by Peter Morriss in Power: A Philosophical Analysis. Power, on his interpretation, is explicitly defined as an ability to act and to bring about outcomes. In contrast with the prevailing literature, Morriss, gives evidence to the dispositional nature of power. Having power refers to certain relatively enduring capacities of individuals, to certain kinds of dispositional properties an individual 4

possesses even if they are never put into action: just as being soluble is a dispositional property of sugar, which it conserves even if it is never put into water, so power is a capacity of an individual even when it is never exercised (2002, 14). To further clarify the dispositional nature of the concept of power, Morriss distinguishes between ability and ableness an obsolete word virtually unused in contemporary English which represents the noun for able (2002, 80). Ability is the capacity of doing something under certain hypothetical conditions, while ableness is the capacity one has when such conditions actually occur: poor people are physically able to eat caviar, though they do not have the ableness to do so (2002, 81). The distinction between ability and ableness has nothing to do with the traditional distinction between actual and potential power, although it may recall it. It identifies two distinct meanings of being able to that is, of potential power the first referring to all the possible outcomes we could bring about if certain conditions occurred, the second referring to the narrower set of outcomes we are actually able to bring about under actual conditions. On the basis this distinction, Morriss recognizes that what is normally meant by social and political power is specifically a form of ableness, since the investigation of power in a society is concerned with those abilities which people possess in that specific time and place, with the current distribution of resources and opportunities. Under this interpretation, then, power does not consist in a social relation. This point is clarified by Morriss attempt to distinguish between the concept of power and that of influence. In doing so, he focuses on two possible meanings of power, those covered by the concepts of affecting and effecting (2002, 29). While affecting consists in having an effect on something, in altering it, effecting consists in bringing about something. According to Morriss, power bears no necessary relation to affecting other people; affecting covers the semantic field of the concept of influence. The meaning of power is instead concerned with effecting : power consists, then, in the ability to produce certain outcomes. It is the concept of influence that indicates a social relation, while power only implies the causation of certain results. For Morriss, then, power is power to. In fact, the usage of the expression power over is not common in everyday language: we normally ask ourselves which kinds of power people have and what they have the power to do; we very rarely wonder over whom people have power. 5

At the same time, when we say that A has power over B, we normally imply that A has the power to make B do a lot of important things, rather than only one thing or a few insignificant things. In this latter sense, the relation described by the concept of power over collapses into the idea of being in somebody s power. Other accounts of power do not consider power to and power over as competing concepts. They suggest, instead, that both should be included in order to offer a comprehensive understanding of power. For Keith Dowding, for example, power to and power over are to be interpreted as two analytically distinct parts of the overall definition of power. He refers to power to in terms of outcome power and to power over as social power (1991, 48). According to him, outcome power is the more basic interpretation of power: having power over someone necessarily includes some kinds of power to act. In addition, his approach underlines that both outcome and social power are related to the production of certain effects. While outcome power is defined in terms of the abilities an actor possesses to bring about certain outcomes, social power is interpreted as the ability of an actor to deliberately change the structure of incentives by which is meant the full set of costs and benefits attached to certain behaviors of another actor, in order to bring about certain outcomes. What characterizes social power, then, is the deliberate action of the power-holder that aims to change the incentive structure of the power-subject: also in the case in which an actor exercises his outcome power somebody else s incentive structure could be affected, but, in this case, this result will not be deliberately chosen. According to Dowding, then, social and outcome power are two possible forms that political power can take. But while the former consists in a relation between actors, the latter need not involve more than one actor: outcome power becomes political power only in the case in which its exercise produces side-effects that conflict with other actors interests. The usage of the expressions power over and power to has recently been taken up in normative political theory, in particular among feminist and empowerment theorists. Here, the usage of power to and power over does not refer to competing conceptions of the very notion of power, but generally indicates two different normative interpretations of power. In feminist theory, power to and power over are regarded as two totally distinct forms of power: power over is the illegitimate domination exercised 6

by men over women, while power to consists in the legitimate power which women can acquire in order to face male domination. This distinction is inspired by, and partly recalls, Hanna Arendt s interpretation of power. Arendt distinguishes between power and violence: every exercise of power of one person over another consists in a form of violence, and it does not deserve the label power ; power deserves this name only when it is legitimate, i.e., only when it consists in the ability of the group to act collectively (1970, 44). Some authors, proceeding from this normative perspective, propose a further distinction: they suggest a quadripartite analysis of power where by power over is meant the actual, illegitimate distribution of power in the society and power to, power with (the ability to act together) and power from within (the individual s awareness of her own capacities which motivate the action) are different aspects or phases of the process of empowerment (Townsend et al., 1999). In feminist theory, the most complete analysis of the distinction is that proposed by Amy Allen. Allen defines power over as the ability of an actor to constrain the choices available to another actor or set of actors in a nontrivial way (1998, 33). Following Lukes, she claims that the exercise of power over is not necessarily intentional, since it can be exercised in routine ways by actors who are unaware of their power: this is so in most cases of male-female power relations. Although in feminist theory the expression power over is mainly used interchangeably with that of domination, Allen proposes a distinction between the two concepts: domination is a subfield of the concept of power over, referring only to those occurrences of power over in which power is explicitly exercised in a way that is detrimental for the powersubjects. Power to, by contrast, is explicated in terms of the concept of empowerment: Allen suggests that the two terms can almost be used as synonyms. Power to is interpreted as the ability to attain certain ends; nonetheless, in her account this expression only indicates the power to act acquired by subordinate groups despite their subordination, particularly in the case of women against male domination. In her interpretation, power over and power to (together with power with, which consists in the collective exercise of power to ) are analytically different aspects of power which show a high degree of interrelatedness: in the case of women s empowerment, power to can be exercised by women in order to achieve a more legitimate distribution of power, 7

which includes the obtainment of some sort of power over men. However, these different aspects of power can all be interpreted on the basis of a more general and abstract concept of power: for Allen, as Pitkin and Morriss, having power means being able to (1998, 36). But, unlike the other two scholars, she reserves the usage of the expression power to only to a subfield of being able, specifically, to the power acquired by a subordinate individual or group through the process of empowerment. Power with In the last decades, the usage of the locution power with has emerged in a number of contributions to the study of power. Accounts of power with have been proposed predominantly by feminist theorists and social activists in order to describe a form of power that consists in the collective action of the members of a group. Power with, then, recalls Arendt s conception of power as the ability to act in concert (1970, 44), and in the same way it has been interpreted as a legitimate form of power. In this sense, it has been used mainly in opposition to the concept of power over, the latter being interpreted as the illegitimate power wielded by the dominant group over the powerless mass. Power with could be thought as a kind of power to since it consists in a capacity to act, but, diverging from power to, it refers to a capacity that is necessarily held collectively. Specifically, it is normally used by those advocating empowerment, and consists in that form of power which may arise when people try to resist and overturn a situation of domination. Certain feminist theorists, in particular those focusing on the idea of selfempowerment meaning the process in which the powerless succeeds on her own in acquiring the power to help herself have defined power with as the capacity to act together in order to obtain collectively what is impossible to obtain alone. For example, the sociologists Jo Rowlands and Janet Townsend, analyzing the processes of empowerment in groups of women in rural areas of Mexico, have defined power with as the power emerging from collaboration within the group. According to them, acting together not only empowers the group to obtain collective goals, but empowers every participant in the group, giving her a new awareness of her capacities through the experience of the sense of communion. In their theoretical analysis, then, power with is 8

both a result and a cause of another form of power: which is power from within, that consists in the capacity both of the individual women and of the group to motivate themselves in order to act. Social activist Julia Kraft, still connecting the concept of power with with the idea of empowerment, offers a wider interpretation of its role in the redistribution of power within societies. Power over, interpreted as domination, is the form taken by the actual distribution of power. The power of groups which fight against political oppression is, on the other hand, power with. But power with is not only the form that power acquires when oppressed groups struggle to obtain a redistribution of power. It is, on Kraft s account, the form of power which should completely replace power over, since it provides a more legitimate basis for the reallocation of power in society: acting together not only empowers people in order to overturn a particular situation of domination, but teaches them how to live collectively, sharing the equal ability to shape their lives. A more sophisticated account of power with is, however, the one put forward by Amy Allen. In her analysis of feminist interpretations of empowerment, Amy Allen relates the concept of power with to that of solidarity. She distinguishes between power over, power to and power with : the power of acting together cannot be described by the concept of power to, because it does not consist only in the possibility of obtaining certain specific goals, and cannot be described by the concept of power over, because it does not consist merely in acquiring power over those who are in a position of dominance. According to Allen, power with is a descriptive concept which does not necessarily refer to a legitimate form of power: it could be used to indicate, for example, the power exercised by a military group over a population. In feminist theories, on the other hand, Allen suggest that power with is to be used only to refer to that specific and legitimate kind of collective power that can be interpreted in terms of solidarity. In the process of empowerment, solidarity emerges when women act together organizing themselves in feminist groups or movements in order to foster the common purpose of overturning the system of domination. 9

Criteria for the distinction As emerging from the previous sections, the distinction between the notions of power over, to and with does not always rely on the same criterion. While a widely accepted distinction is the one between the relational nature of power over as opposed to the ability-based one of power to and power with, in recent literature the distinction frequently makes reference to normative criteria. In this section, my aim is to introduce three of the most relevant criteria for the distinction between the three concepts and to propose a critical discussion of their suitability in actually providing clear boundaries for the three concepts. 1) Extensional (value-free) criterion: - Power over consists in a relation of (intentional) social causation; - Power to and power with consist in mere abilities held by single individuals or groups of individuals to act or to bring about certain effects. As highlighted in the previous pages, the most common criterion for distinction refers to the opposition of a relational and an ability-based understanding of power. However, this same criterion provides less clear indications for the inclusion of a particular case within the one or the other umbrella than it appears at a first sight. Example 1: Is political decision-making power a case of power over or of power to? It is a case of power to because it consists in the capacity to bring about outcomes BUT it is also a case of power over because it consists in the ability to obtain compliant behaviors on the part of the relevant social others. In particular, in this case, either we assume that in Who Governs? what Dahl (1961) was investigating in asking who has more power in New Haven was power to (despite his own relational definition of power; 1957); or we assume that the power to make political decisions is equivalent to the power over the political community that must conform to those decisions (Pansardi 2012). The same reasoning can be applied to a number of cases which ranges from the ascription of political and social rights (i.e., the right to vote is at the same time an 10

individual ability and a social relation with all the individual affected by the results of the election) to individual capacities (i.e., my ability to buy a car is equivalent to my ability to cause the car-seller to sell it to me); this is so unless we consider cases of power to also those kind of mere abilities that are held by actors in a social vacuum as in the case of Robinson Crusoe to build a shack in a desert island (Pansardi 2012). 2. Normative criteria a. Conflictual/consensual - Power over is generally considered as a conflictual form of power; - Power to and power with are consensual forms of power. Power over can be considered as conflictual if we take Dahl s definition according to which A has power over B to the extent that he can get B to do something that B would not otherwise do (1957, 202-203). However, in certain cases we receive commands that are supported by the threat of (severe) sanctions even if we agree with the content of such commands or if we would have performed the same action or choice also in the absence of those commands. Accordingly, we must reflect on which one is the best criterion for explaining power over: if we rely on the conflict between power-holder and powersubject we must leave out from the extension of the concept cases in which the powersubject agrees on (or is indifferent with regard to) the content of the command, as in the case, for example, in which a police-man orders us to turn right to avoid an accident or road works. If, on the other hand, we define power over as including all cases in which the power-holder is able to cause actions on the part of the power subject, then we need to recognize that conflict is not a necessary condition for power over to be in place. Power to, on the other hand if we take seriously its definition in terms of ability does not involve any social relation and accordingly it cannot be claimed to be either conflictual or consensual in its own nature; only the effects on social others of a specific exercise of power to may result in conflicts with others. The relevant opposition here, however, is not so much the one between power to and power over, but the one between power over and power with. A diffused interpretation of power with, which follows Arendt s description of the opposition 11

between power, on the one side, and violence, on the other, refers to the idea of power as the ability to act in concert, understood as the main ideal on which to shape our democratic societies. The idea of power over, which Arendt calls violence, is an intrinsically negative idea. Accordingly, in order to have a legitimate society, the classical relations of power over which shape our societies should be dismantled, and replaced by a political structure based on the ideal of collective power, according to which every individual will be free from coercion and will be treated as an autonomous agent. In this sense, power with is an intrinsically positive notion, which describes a legitimate political structure in a society. On the other hand, power over defined as violence is always a negative feature of the social relations in a society. Accordingly, relations of power over in a society need to be condemned, and replaced by relations of power with. Power with, then, represent a view of power based on collective consent, a consent that is obtained through discursive practices which allow every individual in a society to act as an autonomous agent. Jurgen Habermas takes from Arendt the idea of collective power, and on it he bases his idea of communicative power. In his idea, the legitimacy of a particular society is based on the discursive practices that regulate the collective decision-making process. Communication, and rational argumentations are at the basis of the process of collective will formation, which is at the core of the idea of democracy. Communicative power, then, is power with, a power held collectively by all the members in a community. Just like in Arendt, then, power with should completely replace power over, seen as an intrinsically negative feature of the social relations in a society. Habermas writes: [ ] in contrast to Weber, who sees the fundamental phenomenon of power as the probability that in a social relationship one can assert one s own will against opposition, Arendt views power as the potential of a common will formed in noncoercive communication (Habermas, 1996: 147) The main characteristic of collective power, in Habermas s idea, is the possibility of collective decisions formation. Persuasion, through the means of rational 12

argumentation, as opposed to coercion, is the only power that should circulate in a legitimate political community. The collective decision-making process is ruled by unconstrained argumentation, which only aim is the collective and cooperative search for the truth. Habermas s conception of communicative power gives life to an understanding of democracy which is in clear opposition with the classical liberal understanding of democracy as preference aggregation: public deliberation, based on rational argumentation among speakers oriented towards a mutual understanding, is the only way to legitimize the outcomes of the collective decision-making process. Open argumentation does not only allow for the circulation of information, but also works for the transformation of individual preferences. In this way, accordingly, public deliberation allows for the formation of a collective will. Collective power or power with - then, is understood as the power exerted by the collectivity as a whole in the production of legitimate political decisions. It is non-coercive power since it allows every individual to be her own lawmaker, due to the cooperative enterprise the search for the truth. Power with is then the power which should replace all the other forms of decision-making which are nevertheless intrinsically coercive. Power with, accordingly, should replace power over. According to this first interpretation of power with, then, power with should be understood as an entirely consensual interpretation of power. It is, in particular, a kind of power which should replace the actual structure of power in our societies, shaped on the basis of relations of power over. Power with, is accordingly, the only legitimate and legitimizing kind of power. Power with corresponds, in this sense to the collective process of will formation in light of the production and of the legitimation of political decisions. In other words, power with can be said to correspond to a particular interpretation of the idea of democracy. However, a second interpretation of the notion of power with emerges from the literature, and does not seem to understand power with as a particular structure of power in a society, but rather, as a particular form of power. A power that is collective not in the sense that is both exercised by and over everybody in a community, but a power that is collective in the sense that it exists only if a particular group acts as a collective agent. Power with, in this sense can be described as the kind of power that arises when different 13

individuals in a group join their different forces at the aim of achieving a collective goal a goal that none of the participants could achieve individually. Just like, in order to build a house, different individuals skills are needed you need an architect, a builder, a plumber, an electrician and so on in the same way, in order to achieve determinate goals, different individuals need to join their forces. Then, just like a plumber would not be able to build an entire house by himself, in the same way a single individual may not the able to achieve a particular goal outside the collective action. In the case of an unequal distribution of power, then, power with can constitute the only strategy for self-empowerment of powerless groups. Power with is, in this sense, not to be opposed to power over, nor to be interpreted as a necessarily consensual form of power: it is consensual only in reference to those who participate in the power exercise, not in the reference to the potential power-subjects; for example, those who actually hold the largest share of power within the society (the dominant group). When different individuals join their forces together in order to fight for the achievement of a collective goal, being that goal the obtainment of a more equal distribution of power in a society, power with is nothing but a form of power over. b. Legitimate/illegitimate - Power over is a synonym of domination - Power to and power with are synonyms, respectively, of empowerment and collective empowerment. While the predominant use of the notion of power over seems still to understand power over as a synonym of domination, a number of authors argued against this equivalence, assuming that domination is a subtype of power over defined on the basis of: 1) its (negative) effects on the power-subjects (Allen 1999; Lukes 2005; Wartenberg 1990), recognizing, accordingly, that instances of power over can actually be beneficial for the power-subjects, as in the cases of mothering and teaching (Allen 1999; Wartenberg 1990); 2) the presence of a large degree of power differential and a degree of the stability of the relation under scrutiny (Pansardi 2013; Wartenberg 1990). Accordingly, the 14

correspondence between power over and domination is only contingent, and as a consequence, the illegitimate nature of power over can only be attested case by case. Power to and power with are, by contrast, considered as legitimate forms of power by scholars endorsing accounts of power inspired by the notion of empowerment. While power to consists in the ability to act that might/should be achieved by the single individual, power with consists in a collective understanding of empowerment. This two forms of power are accordingly legitimate given the current condition of oppression and domination lived by the groups in question. If we understand the distribution of power in a society on the basis of the opposition between powerful and powerless groups, power, is to be understood as power over, a power which may manifest itself in terms of the domination of certain individuals over others. But domination and power over are not synonyms: domination refers only to particular instances of power, defined not on the basis of a particular moralized interpretation of it for example, of its detrimental effects on the power-subjects interests but in terms of the asymmetry of the relation it entails: domination, then, describes a particular distribution of power over, and not power over itself (Pansardi, 2013). In a friendship, for example, the two parts exercise power over each other: they are able to influence each other s behavior, they are able to prevent the other from choosing certain options and so on. The relation between the two friends, based on the reciprocal love and respect they feel for each other, can be described as a relation of reciprocal power over, a relation of power in which domination takes no place. We would say, by contrast, that this relation resembles a case of balance of power. On the other hand, a situation like the one of the proletarian depicted in the Marxist interpretations of the labor market in capitalist societies can be described as a case of extremely unequal distribution of power over, and accordingly, as a case of domination. While in the case of friendship, the discontent friend can easily avoid to comply with the excessive requests of the other, and eventually, even choose to terminate the relationship, the proletarian has no other choice than to submit to will of the capitalist, whatever the requests of the capitalist will be. In the case of an unequal distribution of power, then, power with can constitute the only strategy for self-empowerment of powerless groups. Power with is, in this sense, not to be opposed to power over, nor to be interpreted as a consensual form of power. 15

When different individuals join their forces together in order to fight for the achievement of a collective goal, being that goal the obtainment of a more equal distribution of power in a society, power with is nothing but a form of power over. Accordingly, in this case, power with is not to be interpreted as a form of power which should replace power over in our society, but is to be understood as a strategy of the least powerful groups for the obtainment of a more equal distribution of power over. As Amy Allen writes in her critique to Iris Marion Young s interpretation of power in Justice and the Politics of Difference: Once we conceive power in relational terms, then not only does it no longer make sense to think of domination or oppression as a lack of some sort of stuff or good called power but it also no longer makes sense to think empowerment as a process of gaining or acquiring the same stuff. Rather, just as domination and oppression are understood in terms of social, cultural, economic and political relations that impede self-determination and self-development, empowerment must be understood in terms of social, cultural, economic and political relations that foster and promote these same capacities (Allen, 2008: 165-166). Accordingly, we can claim that both domination and empowerment are social relations, and specifically, relations of power over. Although the description of the idea of empowerment of a group subject to domination may suggest an interpretation of empowerment as the achievement of immunity to domination (Thomas, 2011: 446), such an immunity can only be guaranteed by the ability of the oppressed group to impede domination: in other words, it can only be guaranteed by the acquisition of power over the dominant group. Empowerment, then, can be understood as the achievement of a more equal distribution of power. It is worth noting that neither the notion of power with nor the one of empowerment are to be seen as intrinsically normatively positive concepts: just like, as Amy Allen notices, the power exercised by a military group over a population may be seen as a case of power with, in the same way the idea of collective empowerment of a specific group may indicate the capacity, for example, of fundamentalist religious movement to challenge secular institutions. Neither the idea of power with nor the one of 16

empowerment are, accordingly, in this sense, to be seen as intrinsically good forms of power. Moreover, the relation between the notion of power with and the notion of empowerment is only contingent: collective power and empowerment are not intrinsically related: again, since power with may indicate also the power exercised by a military group over a population, certain cases of power with are what we would normally describes as forms of domination. Conclusion In this paper, my intent was to show that the distinction between the notion of power over, power to and power with is more blurred than it appears at a first sight. Power over, power to and power with, although analytically distinct, present some significant overlappings. Accordingly, I discussed those which are, to my knowledge, the most relevant and diffused criteria for distinction of the three concepts, and I argued that either they only superficially explain the difference between the concepts or they are based on the weak premises. At the merely extensional, value-free level of the distinction, I argued that the distinction between power to (and power with) as ability-based and power over as relational is partially misleading if we look at the heart of the matter: unless we are referring to ability that are held in a social vacuum, all abilities held by individuals in a society correspond to powers that those individuals have over others (Pansardi 2012). Secondly, I took into consideration two normative interpretations of the distinction between the three concepts: the dichotomy conflictual/consensual power and the one between legitimate/illegitimate power. In reference to the conflictual/consensual distinction, I argued, firstly, that power over is not necessarily conflictual, since what characterizes a relation of power over is not the presence of a conflict between the powerholder and the power-subject on the content of the command to which the power-subject must conform, but the mere fact that the power-subject has no other choice to conform to the command also in the case she agrees to it or is indifferent to it. Secondly, I argued that, unless we follow Arendt s and Habermas s interpretation of collective power (power with) as a synonym of legitimate deliberative democracy, we have no reason to claim 17

that power with consists in a consensual form of power. Power with as well is equivalent to collective power over and accordingly may be exercised over other individuals in a situation of conflict. Lastly, with regard to the distinction between legitimate/illegitimate power, I have argued that power over and domination are not synonyms, and that the legitimacy of an exercise of power can only be attested case by case. On the other hand, I have argued that while power with is often thought of as collective empowerment, in terms of the achievement of a more equal distribution of power (over) within a society, power with can also indicate cases in which the group who is collectively exercising power is the dominant or the oppressive group. As a result, the distinction between power to, over and with is more blurred than it appears at a first sight and should not be thought of as one between different forms of power, but should be rather understood as a distinction between different faces of power, i.e., of different aspects of the phenomenon of social power that can be each in turn of particular interest for the study of the society and that may serve different theoretical or empirical purposes. References Allen, A. (1998). Rethinking Power. Hypatia, 13, 21-40. Allen, A. (1999). The Power of Feminist Theory: Domination, Resistance, Solidarity. Boulder (CO): Westview Press. Allen, A. (2008). Power and the Politics of Difference. Oppression, Empowerment and Transnational Justice. Hypatia, 23 (3), 156-172. Arendt H. (1970). On Violence. New York: Harcourt Brace. Bachrach P., Baratz M.S., Decisions and Nondecisions: An Analytical Framework, American Political Science Review, Vol. 57, 1963, pp. 641-651, reprinted in BELL R., EDWARDS D.V. and WAGNER R. H., (eds), Political Power: A Reader in Theory and Research, The Free Press, New York, 1969, pp. 100-109. 18

Bachrach P., Baratz M.S., Two Faces of Power, American Political Science Review, Vol. 56, 1962, pp. 947-952, reprinted in BELL R., EDWARDS D.V. and WAGNER R. H., (eds), Political Power: A Reader in Theory and Research, The Free Press, New York, 1969, pp. 94-99. Dowding, K.M., 1991. Rational Choice and Political Power. Edwan Elgar: Aldershot. Dowding, K.M., 1996. Power. Buckingham: Open University Press. Eyben, R., 2004. Linking Power and Poverty Reduction. In: Alsop, R., ed. Power, Rights and Poverty: Concepts and Connections. Washington: The World Bank. Flynn J. (2004). Communicative Power in Habermas s Theory of Democracy. European Journal of Political Theory, 3 (4), 433-454. Habermas J., (1986). Hannah Arendt s Communications Concept of Power. In Lukes S. (ed.), Power. Oxford: Blackwell, 75-89. Habermas J. (1996). Between Facts and Norms. Cambridge: MIT Press. Haugaard M. (1992). Structures, Restructuration and Social Power, Fordham: Ashgate. Haugaard M. (1997). The Consensual Basis of Conflictual Power. Constellations, 3 (3), 401-406. Haugaard, M., 2010. Power: A Family Resemblance Concept. European Journal of Cultural Studies, 13 (4), 419-438. Lukes, S., 1974. Power: A Radical View. London: MacMillan Press. Lukes, S., 2005. Power: A Radical View (Second Edition). London: MacMillan Press. Kraft, J. (2000). Power-with, not power-over, Peace News, June-August, 35. Morriss, P., [1988] 2002. Power: A Philosophical Analysis. Manchester: Manchester University Press. Pansardi P. (2012). Power to and Power over: Two Distinct Concepts of Power?, Journal of Political Power, 5 (1), pp. 73-89. Pansardi P. (2013). A Non-normative Theory of Power and Domination, Critical Review of International Social and Political Philosophy, 16 (5), 2013, pp. 614-633. Pitkin, H., 1972. Wittgenstein and Justice. Berkeley: University of California Press. Thomas, B. (2011). Unraveling and Discovering: The conceptual relations between the concept of power and the concept of empowerment. Critical Review of International Social and Political Philosophy, 14 (4), 443-463. 19

Townsend, J., Zapata, E., Rowlands, J., Alberti, P. & Mercado, M. (1999). Women and Power: Fighting Patriarchies and Poverty. London: Zed Books. Wartenberg T.E. (1990). The Forms of Power: From Domination to Transformation. Philadelphia: Temple University Press. 20