The Domestication of Power

Traditional political philosophy, curiously enough, had not addressed the problem of power for centuries. Of course, I do not mean problems with the actually reigning Power, I mean the problem of power as such. Eventually, we should not be surprised at this, since political activities in general were captured and reflected by legal and (moral) theological terms. Aristotle's Politics, as is well known, is much more a catalogue of forms of constitutions, embedded into a moral philosophy, than a conception of politics on an independent footing. Even Machiavelli, whom many hold to be the founder of political science, and who indeed was raptured, so to speak, about power, did not write a philosophical treatise on this subject. The vocabulary of political philosophy still relies heavily on concepts borrowed from moral philosophy and from social sciences. Historically, the most important concepts have been the nature of man, the emergence of norms, the social contract, laws and the natural law, the state of nature, justice, passions, interests and reason. In Hobbes's theory power is considered to be a mere passion (lust for power), the Leviathan is more a legal category than a power entity. The souvereign has much more to do with the social contract then with power.

The first concept that could be given a purely political understanding was liberty. The ideological struggles within and after the French revolution, the first secular revolution in history revolved around this concept. Obviously, whatever meaning liberty was attributed to, the chief evil was to be the lack of liberty or freedom. [1] To simplify matters, I shall distinguish between two ways about how the lack of freedom can be interpreted. According to the first interpretation freedom is the rule of reason, wherefore the lack of freedom is identical with the lack of reason, or with the rule of irrationality. The other interpretation, which will interest us, equates the lack of freedom with arbitrary power, in other words, with tyranny. Much attention has been devoted to the relation of freedom to reason, less to the problem and meaning of tyranny. This is quite interesting, though of course there have been some notable works written on this subject, for instance Marx's Economic-Philosophical Manuscripts contain interesting insights about the exploitation of man by man, that draws back to Hegel's discussion on the relationship between master and servant. But Marx - and marxism - remained - paradoxical as it may sound - uninterested in political categories, including freedom as well.

Before I go into detail, I need to mention one more philosophical development. The disappointments with the results of the French revolution led many thinkers to a skeptical stance toward reason and induced them to endorse a more individualistic interpretation of freedom. And Nieztsche, famous for being the par excellence philosopher of power, wrote in an era when individual freedom had also become a problem. His solution, as it were, was nothing else but an attempt to reconcile the oppositional concepts of freedom and power, of course, in a strictly noncollectivist conception. The collectivist conceptions arose from other angles.

Why is it important to see that freedom and power have clearly discernible meanings, reflecting on clearly discernible phenomena, to be studied by moral and political philosophy? I hope to be able to argue for these allegations in an indirect way. I shall maintain the traditional view that for political philosophy freedom and power are oppositional concepts, and that for sociology and moral philosophy power is an interpersonal relationship, but I shall not argue for these claims. I am going to say more about why and how it is a mistake and, indeed, why it is harmful not to accept these claims. I shall proceed as follows. First, I shall argue that Max Weber in his Economy and Society made two significant mistakes when he defined power. The first one was of a philosophical, the second of a sociological sort. Second, I shall briefly survey the major approaches to power that all rest on Weber's definition - and on his mistakes. Thirdly, I shall discuss the relationship between social science and political philosophy in terms of these approaches, and I shall argue that without a sound political and moral philosophical understanding of power we shall be defenseless against power itself.

Max Weber defines power in the following way: "'Power' is the probability that one actor within a social relationship will be in a position to carry out his own will despite resistance, regardless of the basis on which this probability rests." (53) This definition is objective, short, yet accurate and impartial, devoid of all philosophical worries; it appears as it were a concise summary of our most direct impressions about power. But Weber goes on immediately to define domination, which differs from power in that it "is the probability that a command with a given specific content will be obeyed"; and, more importantly for us, that "the concept power is sociologically amorphous. [...] The sociological concept of domination must hence be more precise..." (ibid).

There are two questions here. First, it is unclear whether Weber wants to say that the concept of domination is an adequate concept to capture the phenomenon of power as well or it can cover it only partially. Second, it is unclear what is amorphous : is it power or is it the concept of power? Literally, Weber asserts that it is the concept of power that is amorphous; yet the justification he offers suggests that he meant the phenomenon of power as being amorphous (from a sociological point of view). The difference seems a nuance, yet its consequences are far-reaching, for two reasons. First, it results (and has indeed resulted) in superficiality. Most social and political scientists, including the representatives of the pluralist school who did much empirical researches in the sixties and seventies about the power relations of communities, towns, firms, institutions and political systems, simply disregarded the fact that Weber has no conception of power, only a definition of it. They ritually cited Weber's definition and based their researches on it, not taking into account that the Weberian sociology is about domination and not about power. Second, Weber's suggestions do in fact allow for the most diverse interpretations. If we take a closer look at Weber's text, there seem to be two possible ways to make sense of power and domination within the same conceptual framework. Unfortunately, both interpretations are seriously flawed.

First, we may infer from his definitions that although every form of domination rests on power, but power refers to a much broader realm of social phenomena, which can only partially be investigated by means of sociology. Weber's own introductory notes explaining his methodology give some support to this interpretation. Second, one may argue that the concept of domination can more easily be inserted into the overall conceptual scheme of Weberian sociology, but it refers to the same realm of social phenomena as power does. Weber refers here to the concept of command, later he refers to the concepts of claims of legitimation and of domination. The first interpretation can be called - strictly for the purposes of this essay - a "realist", whereas the second a "nominalist" understanding. However, whichever interpretation we come to accept, we come to a deadlock: either power is something excluded from sociological research (and possibly left for philosophical speculations), or it is undistinguishable from domination (and possibly from other concepts). In both cases, power is set free. It is uncontrolled by sociology - and uncontrolled by philosophy, either.

"Power" and "domination" are both "probabilities". Though it is sufficiently obscure how power and domination could be "probabilities", what we should note is the explanation that follows in both cases. Power works, according to Weber, within a social relationship which presupposes two actors, one who wants the other to obey, and the other who in fact obeys. Domination, on the other hand, does not seem to be presupposing social relationships so defined: it works in a passive mode ("a command ... will be obeyed"). Thus, there is a significant difference between power and domination. As a matter of fact, Weber might have been unaware of the significance of this difference (which we still can hold to be a mistake); later developments in sociology suggest, however, that the conceptual link connecting micro-level social actions with macro-level social phenomena is seriously missing. If power is taken to be a micro-level interpersonal phenomenon, and domination a macro-level structural phenomenon, connecting them with each other seems a much more difficult business than Weber could have assumed.

Weber's definition of power, as I already explained, has had a tremendous impact on social science. The inner tensions have also been transmitted. The realist and the nominalist positions typically intertwine. However, what demarcates the two major views of power is not this distinction, but the difference between the micro-level approach and the macro-level approach. Since neither approach reflects on the distinction I made, neither of them can uphold a consistent conception of power. I shall discuss some examples.

The micro-level approach to power yields individualist, whereas the macro-level one yields collectivist conceptions. The latter ones typically employ holistic methodologies, and power works as a mechanism or as a social entity on which individuals have a limited influence (and even if they could exert some influence on it, they would be uninterested in it). The individualists conceptions rely on methodological individualism, that is usually associated with rational choice theories. Power is for them an interpersonal relation (so at least they claim), or a commodity to change or a utility to enjoy. It may be argued that for the postmodern conceptions of power another rubric should be set aside, because they oppose both the individualist and the collectivist approaches. Since my position is that these controversies are only superficial, I am not concerned with whether this new category can be justified or not; at any rate, I shall discuss two examples belonging to this group separately. Let us consider the collectivist set.

For Hannah Arendt, power is the opposite of violence. Power is the collective capacity of society to attain common goals. Power appears and disappears, basically as a function of people's trust in the political system to secure and carry out the common goals. Violence, on the other hand, is the demolition of power. Society cannot rest on violence, but on power. In Arendt's depiction power is a profoundly positive feature of human society. Arendt is a realist, insofar as she thinks that power is a real phenomenon that cannot be investigated by purely sociological means. And she is a nominalist, insofar as she tends to see power as a collective capacity and nothing else. The problem of how the collective capacity can be derived (through what mechanism it is brought about) is not addressed by Arendt.

According to Jürgen Habermas, Arendt's conception of power is incomplete, because Arendt stresses only the generative (social) aspect of power, and disregards its operational-technical (political) aspect which means struggle, competition and "power motivations". Habermas shares the collectivist view of Arendt, in that for him power is a medium of the political system, and he incorporates into his theory the Arendtian observation that power exists only via consent (of course, Habermas takes consent to be communicatively produced). In other words, Habermas's conception of power amends the Arendtian view with an aspect of power that is intended to make the conception more realistic, but it basically preserves the structure of his predecessor's conception.

In Talcott Parsons's famous system theory power is assigned a "lubricative" function, being the medium of the goal-attaining subsystem which converts the outputs (performances) of other subsystems into power (which is the form of these outputs that can utilized within the goal attaining subsystem) and converts the outputs of the goal-attaining subsystem for other subsystems. Parsons talks about "generalized" mediums, which is a reference to the initial model which was a two-person (i.e. interpersonal) game where power generates from. However, Parsons is even less interested in the problem of how to derive social structures from micro-interactions. Parsons was among the first scholars to introduce the Weberian sociology in the US, and in respect of his conception of power he seems to have remained faithful to the Weberian ambiguities. Power is derived from micro-level interpersonal relations, though Parsons thought, contrary to Weber, that power is not an amorphous phenomenon that cannot be investigated by means of sociology (the realist element); but did not see any relevant difference between power and domination - authority in the Parsonsian terminology -: authority is chrystallized or fixed power (the nominalist element).

Niklas Luhmann offered an even more detailed conception of power as a symbolically generalized communicative medium. At crucial points he follows Parsons, but like Habermas, he amends the Parsonsian conception with the concept of communication. Communicative media have a somewhat different function for Luhmann; they emerge to resolve problems of reducing complexity which historical progress inevitably brings forth.

In the case of Parsons and especially Luhmann we must note that power is a medium that has functional equivalents. If we dig into the micro-level structure of power, truth, trust, love, money and norms, we find that each medium serves the purpose of imposing the will of Ego on Alter, which is the best way to reduce complexity. [2] This is not the place to discuss these theories in detail, it is just crucial to see that on the basis of Weber's original definition these media are but cases of pure power. And if they are, then we must think of power as an all-embracing and the prevalent feature of every social form of human existence. To accomplish this picture, we also have to note that the need to reduce social complexity in many cases quite literally "forces" the individuals to invent and apply such media. This means that power is less a product of interpersonal relations, and more a product of historical progress, to which we all are subjected. Thus power is not something we produce, but something that produces us. Now since each medium emerges via communication, and since each medium is reducible to power, we get the result that power and communication are by and large identical. This is the conclusion that takes us to the third approach to power, which I shall discuss below in more detail. To return to Weber, the upshot of Luhmann's communicative conception of power can be expressed in the following way: power is not only an ubiquitous social phenomenon that can be examined by sociological means, but it is the "most real" phenomenon of social life. On the other hand, it is not only domination that becomes identical with power, but even trust, love, truth and the rest must be seen and understood as but mere faces of power. The concept of power is itself a power phenomenon, and (the phenomenon of) power works by means of concepts.

Let us now switch to the individualist conceptions. The analysis of power is carried out on an individual level, usually within interpersonal relations, in small groups (committees) or networks. The starting point is the original Weberian definition of power. It is assumed that the phenomenon of power can be exhaustively and clearly described by means and in terms of sociology (?). A big advantage of these conceptions seem to be that they virtually refuse to explain how macro-level phenomena can be derived from micro-level interactions, sparing a lot of energy. Unfortunately, however, there are some other difficulties arising from the other premises these conceptions employ which make them equally inconsistent like the former conceptions.

The major problem is this. Whereas power is studied within interpersonal relations, it is attributed to individuals. Power refers to a property that individual actors are possessing. Sometimes it is called a capacity to act in certain ways, usually in the face of resistance, which is usually supposed to be produced by another human being. In network analyses and in exchange theories power is explicitely taken to be a commodity with which everybody can make business, either as being able to "buy" obedience, or as being able to "sell" obedience. There are three obvious difficulties with this conception. First, it allows too much room for free deliberation and choice as to comply or not. True, in many cases we do ponder about questions of complying or not complying, we do make calculations whether resistance or compliance would pay more. But in many other cases we do not calculate, but make moral decisions (disregarding the personal benefits and harms). Second, an exchange relationship presupposes that what is exchanged, is valuable for both parties. Compliance, however, is rarely (though not never) something valued by the one who complies. Third, even if power were a commodity, its value would be always up to the particular exchange relationship. But this is not the case with normal, marketable commodities. In other words, what the person exerting power wants is, again, rarely "obedience" as such, but obedience from this or that person. Further, the "costs" of (producing) power are determined subjectively (see Barry's analysis), which again makes the analogue of power to marketable commodities fail.

In conclusion, I claim that power evaporates in the hands of rational choice theorists and network analysists. The main cause for this is a similar mistake that Weber made: power is taken to be something else what it is. It is neither a commodity, nor utility, although it has some features that make it look like as if it were a commodity or utility. It is neither a substance, nor a kind of perception. Ironically, the individualist approach tends to materialize power in the same way as the collectivist approach does, and and the same time it tends to read power into the most various types of interpersonal relations. In other words, the nominalist view destroys the mistaken realist approach.

The third approach is the postmodern one. I mention two conceptions, the most famous ones: that of Michel Foucault and that of Pierre Bourdieu. It seems impossible to sum up either of them in a few lines; both authors have written numerous books on the subject. As a matter of fact, especially in the case of Foucault, our job is not that hopeless. For Foucault's language is itself a professed attempt to disentangle himself from his subject, his main thesis being that language itself is permeated by power. The many definitions that he gave about power serve this simple purpose, his talking or discourse on power is nothing but a constant reflection on the ubiquity and multifaceted nature of power. Power has "extension", but it also "flows", yet it exists only in action; it is available to each of us, yet we cannot take hold of it at will; power, domination and force refer to the same phenomenon, yet they are not the same; power has a "function", but maybe only "effects"; power constitutes us, and we constitute power. The inflation of metaphors is a very powerful rhetorical tool, extensively used by Baroque authors, though Foucault would by quite surprised at this characterization. It tells a lot, however, that the typical debate between adherents of Foucault's conception and his opponents very soon turns out to be a debate between deafs. And this does not come as a surprise, since the Foucauldian strategy to find power is a constant re-definition of power, and this is, we must admit, the only available strategy, given the premises. Put it in an ironic way, Foucault has no conceptual power to find anything else but power.

Bourdieu's conception of power is a close relative of Foucault's theory. Bourdieu focuses much more on the tryadic relationship of language, power and individual. His basic proposition seems to be that language represents, expresses and confirms power. Power dresses into symbolic terms. The individuals' are more or less defenselessly exposed to this, the reality of power relations is the primary, the actual holders of power are the secondary. The result is that within the tryad of power, language and individuals it is power that "explains" and the language and individuals are what "is explained". The obvious question is how we learn about power, how it is possible to speak about power. To this question I see three possible replies. First, power created us in the way that we have some (limited) knowledge about it (either we learn about it or we have an innate concept of power). This is the "theological" argument. The consequence of it is that there is no way to reflect on power in the classical sense, since whatever we come to realize, is power-dependent. The second reply would be that our concepts of power do not simply denote the phenomenon of power, but carry it and make it real. The consequence is that language and power become identical, the original tryad collapses. The third reply, however weird it may sound, is that we do not learn about anything through language and conceptual analyses and the individuals do not speak at all: what happens is that power acts; language and discourse is but power in action .

Let me summarize. Although Weber was right in supposing that - sociologically speaking - power was an interpersonal relationship. But what we labelled as the realist and nominalist intepretations of the relationship between the concept and the phenomenon of power, together with Weber's inattention to the problem of the two levels of sociological explanation rendered all conceptions resting on Weber's definition incapable of producing a consistent sociological theory of power. Instead, what they have produced, is a philosophical nonsense. Or rather, let us admit that what is a philosophical nonsense, can still exist, and call it a philosophical monster.

Let me, therefore, return to political philosophy. It is quite interesting that all the conceptions - Arendt's being a possible exception - that we have discussed have been advanced by social scientists and not by political philosophers (in the traditional sense). As a matter of fact, for the most part of our century political philosophy has been regarded as outmoded by social sciences, for three reasons. First, politics became a profession that needed professional and not philosophical knowledge about society, economics, international relations, etc. Second, political philosophy was considered to be "ideology per se", to be studied by professional means. Third, political philosophy has always been a normative field, which many scholars hold - not incorrectly - to be a subfield of moral philosophy. And normative questions have been declared to be unscientific. And despite the revival of political philosophy from the seventies, no serious philosophical treatise has been published on power. My job here is to merely demonstrate in what ways the social scientists' understanding of power has changed the - virtual - philosophical meaning of this phenomenon.

As I explained in the Introduction, the last message on power coming from political and moral philosophy was that it constrains freedom, both individual and common. Power as such is something repugnant, even abominable. It represents a constant tendency toward tyranny, or, to use Lord Acton's phrase, toward corruption. This holds for individuals and communities alike. What power exactly consists in, was not discussed, with one exception. I have in mind the conception of "will to power", which is part of a rather fragmentary philosophy, yet it clearly reflects on a radical change: power and individual freedom make their first steps together.

Weber surely had not in mind these Nietzschean ideas when he formulated his definition of power. Yet his account has proved to be as important from a political philosophical point of view as was Nietzsche's philosophy. As is well known, Weber put a great emphasis on the impartiality of social science, which implies that the social scientist also has to bracket his or her ideological or normative commitments. His definition of power was conceived of in this spirit: it is impartial and devoid of philosophical concerns. Impartiality made it possible that power was understood as one phenomenon of society among others; being devoid of philosophical concerns implied that power was not examined jointly with freedom. Both features are, in my view, highly relevant.

For Arendt, as we have seen, power is an emphatically valuable phenomenon; since she was a philosopher, she could still afford normative judgments. But she did not note, despite her overall critical stance toward modernity, that her positive assessment of power is the very essence of modernity with respect to power. Her thesis that power is a collective capacity of society is, from political philosophy's point of view, an identification of power with freedom.

Habermas seems to have brought back in the dark side of power, but the price for that is a deep inconsistency in his theory. For if power flows from intact and pure communicative action, but as a medium of the political system it can threaten the life-world, and potentially the pureness and intactness of communication, then it is hard to see in what ways such power can be resisted. And if the power of the political system can be resisted at all, the resistance must be based on a kind of power itself. There is a conceptual problem here which has two causes. First, the Arendtian conflation of power and freedom appears again. Second, unlike Arendt, whose theory implies that the decision as to whether power is still intact in society or it has already turned into violence must be made by the philosopher, Habermas assumes that this job is to be done by the social scientist. His theory does not simply provide us with a conception of communicative action, but also with the means to assess the actual conditions of communication. The scientifically purified concept of power, bereft of normativity is assigned a normative task. Hence, we have an ambiguous attitude to power: we are called to love it, because it belongs to normality; but we also are called to be suspicious about it, because it threatens normality.

Luhmann does not have this ambiguity between prescriptions of science and of morality. His theory is fully within the Weberian tradition of impartiality. But his theory contributes to the unintended (let us be fair) ontologization of power. As I pointed out, he makes power identical with not only freedom, but ultimately with all interpersonal relationships. Power is, so to speak, the very form of social existence of human beings. Foucault deepens these insights in that he stresses not only the passive mode of working of power, namely, maintaining society, but also the active, constitutive, creative aspect of power. For Luhmann power is valuable especially from society's point of view; for Foucault power is even more valuable from our, i.e. from the individuals' point of view. Power is necessary not only as something that helps us reduce the complexity around us, but it is even more necessary to form our lives autonomously. Power is not simply freedom, but freedom in a modern sense: it is a prerequisite of autonomy, which is the hallmark of modern morality, too. Power is, thus, not something to be controlled by norms and moral principles. On the contrary, moral norms and principles ought to be designed such that they can effectively defend, foster and cherish power. What was once a fearful and repugnant feature of human existence, has by now become a beloved condition of free and autonomous life. Power has been succesfully domesticated.

I do not want to pretend that the process what turned the meaning of power upside down was exclusively a result of Nietzsche's philosophy and Weber's sociology. Not only power became identical with freedom, freedom became identical with power, too. That would be a different story. I only claim that what happened to power and to some other concepts, was partly due to social science's acting in the role of political philosophy. And this was made possible by bad philosophies. Weber said that the concept of power was amorphous, but he went on writing as if he had said that the phenomenon of power had been amorphous. A minor mistake, indeed. Together with some other philosophical developments, however, it helped social science to ignore the difference of power from other phenomena and from other concepts, and in effect, to ignore the difference of concepts from phenomena in general. Since if power is an amorphous phenomenon, it could be detected everywhere: even beyond concepts. This is the realist error. And if power is an amorphous concept, it can be captured by any other concept: be that domination, authority, violence, force, freedom, love, trust and so on. This is the nominalist error. And if power is to be explained by sociological means, or it is what explains other things in a sociological way, it necessarily loses its interpersonal aspect. I tried to demonstrate that both the realist and the nominalist interpretations of Weber's definition have had fatal consequences, and that the initial definition of power as an interpersonal relationship has to be virtually abandoned, in each conception of power discussed above.

It is a difficult question, and I only have some tentative ideas how to answer it, how and in what ways social science can fulfill the function of moral and political philosophy. What from the story of power we can infer seems to be this. If power is something to which only social science has access, it must say something about the morality of power. Now since the highest norm of social science is bracketing normative assessment, it finds itself in a rather awkward situation. Habermas and Foucault have taken some philosophical responsibility for their scientific conclusions; others have refused to do so, of course, in the name of the highest norm. But power is something that simply does not allow us to remain morally indifferent. I take the Habermasian and the Foucauldian conceptions of power as clear evidences for this. Power must be assessed morally. The problem is, that it cannot be done as a concept of social science. Power cries out for moral and political philosophy. But if we want to evaluate it in moral terms, we cannot do two things. We cannot suppose that only power exists and we cannot suppose that if power is something intrinsically bad, it does not have an opposite which is intrinsically good. In other, more moderate, words, we must insist that power is real, but it has clear boundaries. And we must hold that we can have a concept of power that allows for other concepts that stand in specific relations to the concept of power.

Finally, I want to say some words about the problem whether the social scientific understanding of power could change and has in fact changed the phenomenon of power. This is a tricky question, since if power is an interpersonal relationship, it must somehow be connected with the way the actors understand it, with the meaning the actors attribute to it. But in this way we seem to get on the track that takes us to the identification of power with language. If, on the other hand, we hold that power is a Platonic idea, untouched by the doxa, we attribute to it a kind of existence that excludes interpersonality.

The right answer to this question lies in admitting, on the one hand , that the understanding and interpretation of power can influence its nature, since speaking and discoursing is as real a phenomenon as power itself, and there is no reason to deny that they can influence one another. What's more, the purpose of this paper is exactly to show that the misunderstanding and misinterpretation of what power consists in itself has had harmful consequences on political and social thinking of not only scholars, but of non-professionals as well. For if power is something that we are not to judge from a moral point of view, our sense of and desire for freedom will also be weakened, be that personal or political freedom. And what is an even more fearsome consequence, we shall adapt our moral principles to this situation and take power to be a prerequisite of our personal or common freedom, and think that there is nothing else to strive for.

On the other hand , if there can be misunderstandings and misinterpretations as to what power means, there must be understandings and interpretations that conform to the true nature of power. And I relied in my reasoning on two observations: first, that power is an interpersonal relation, and second, that it needs to be assessed morally. Unless we stick to these assertions, we shall not be able to capture power correctly and we shall end up with bad philosophies and absurd theories. If this can be done only at the price of abandoning all social science approaches to power, letting it to be subject of moral and political philosophy, then I am firmly convinced that this price is worth paying.

[2] Luhmann's terminology is not easy to follow. However, if we manage to translate a proposition such as "the selection mode of Ego becomes a motivation for Alter" into more conventional terms, we shall face the following statement: "What Ego wants, Alter will do". Philosophical refinements are possible, but Luhmann's theory is not philosophy, and I do not see in what ways such refinements could render this theory philosophically consistent.