POWER . The term kratophany literally rendered is "the appearance of power." Mircea Eliade, however, who made this a technical term in English, used it to indicate an appearance of the sacred in which the experience of power dominates. Thus, that every kratophany must be, at the same time, a hierophany ("appearance of the sacred") is certain by definition, while the converse is less clear; indeed, assent to it will hinge upon the degree to which one regards the concept or experience of power to be an irreducible part of the concept or experience of the sacred.
That the idea of power is central to much religious experience can be seen by means of a simple mental exercise: try to imagine hierophany without the elements of awesomeness, authority, or effectiveness. Most will agree that it is possible to imagine intellectual constructs such as truth or value without power, but hierophany seems to require more. Here is one difference between philosophy and religion, between the intellectual grasp of an idea and the experience of a sacred reality: the religious experience involves the whole personality and not merely the intellect. It includes the emotions as well as less obvious aspects of human awareness such as the kinesthetic sense and deep instinctual and symbolic structures. Finally, it may be that the sense of reality and the sensing of power are inextricably combined into what is experienced as a unity that might be labeled "real presence." As a category of modern physics, power can be described as a potentiality, or a potential ability to do "work," which in turn implies the expenditure of energy to change the distribution of energy in a given system, just as water piled up behind a hydroelectric dam has great potential for generating electricity because of its advantageous location with respect to the direction of gravitational forces. Unlike water, however, the sacred always remains potential even after awesome power has been expended, and it is this mysterious characteristic of being an inexhaustible source of power that in part gives to hierophany its paradoxical tendency both to attract and to repulse.
The normal reactions to sacred power within a given culture can conveniently be classified under the rubrics of mana and taboo. Mana implies a positive attitude toward power within an object or symbol or person—power that can be appropriated for useful purposes. Taboo implies the opposite, namely, power in an object or symbol or person that must be avoided for safety's sake or at least hedged about with special "insulating" rites before it can be made useful. Examples are amulets and charms, holy books, saints' relics, and living sacred persons. Infraction of such governing rules constitutes sacrilege and usually brings down cultural or cultic sanctions upon the guilty, or even the direct intervention of sacred power itself.
Perhaps the most important, because clearest, example of the role played by power in religion can be seen by examination of the meaning of cosmogonic myths and of what appears to be the psychological reality that informs them, namely, the universal experience of the prestige of origins. Here, above all, is demonstrated the positive side of sacred power in its intrinsic creativity. Here is the power to bring a world into being, to shape reality, and thereby to found human cults and cultures. It is literally true that within cosmogonic myths everything that happens is a unique demonstration of creative power, since everything that happens does so for the first time. Examples abound, but consider only the Dreaming adventures of many sacred beings in Australian tribal religions, where the seemingly trivial acts performed while traveling around the countryside actually create the landscape and populate it with sacred places gravid with meaning. Or consider the Shinto myths in which with nearly every gesture of the gods—whether by sexual contact, by breaking or cutting something, or by uttering special words—new deities came into existence, deities whose intimate relationship with nature and culture made them constitutive of the world.
More dramatic examples may be found in the Hebrew scriptures, in the Book of Job, for example, where frequent references are made to God's creative power in ordering the world and controlling the awesome forces of the cosmic ocean. As the text comes down to us, Job's response is one of terror and repentance without understanding. The Hindu classic Bhagavadgītā provides another forceful revelation of the sacred as power in Arjuna's trembling witness to Lord Kṛṣṇa's true nature: nothing less than the world process is portrayed in the deity's simultaneous destructive function as death and his creative function as the womb of all beings.
Power and Theories of the Origins of Religion
Although scattered speculations can be found in the classical civilizations of China, India, and Greece, theoretical reconstructions of the possible origins of religion stem in their modern forms from the European encounter with those cultures that, from about the time of the Enlightenment until a few decades ago, were known collectively as "the savages." Knowledge of these so-called primitive (or archaic, or nonliterate) peoples made a strong impression on the Western imagination. Among other things, it played an important role in the foundation during the nineteenth century of such academic disciplines as psychology, sociology, and anthropology. Perhaps because many of the more detailed accounts of such cultures came from religious professionals and perhaps also because it was an age in the West of great religious ferment, the discovery of primitive cultures was both a discovery of exotic social customs and of strange and disquieting systems of belief and ritual. The most significant systematic attempt of this period to reconstruct a "natural history" of religion was E. B. Tylor's Primitive Culture (1871). There the theory of animism was first propounded.
Tylor defined animism as belief, or a tendency toward belief, that all nature was endowed with a spiritual, animating essence, or soul. Thus, by anthropomorphizing analogy, every natural power or object was directed by a personality possessing intellect and will. According to this theory, all things were supposed by humans' primitive ancestors to be humanlike—if not in outward appearance, then in their inner being. Power was implied in this view in that the power of being of every thing, its uniqueness and its efficacy, was assumed to be potentially greater than what one would call its mere physical possibilities. Yet the experiences that lay behind this animistic worldview were not, in Tylor's view, fundamentally of power, with its exciting, often daunting emotional concomitants, but were instead of a different and more coolly logical kind. He reasoned that primitives must have been perplexed by their own dreams and thoughts, in which they themselves as well as other people, both living and dead, and not present in the usual sense, appeared. Adding this to their own natural experience of themselves as thinking, willing, self-moving beings, primitives must have concluded that a soul, or animating principle, must inhere in all things and that it could sometimes be separated from the body. In this way, Tylor sought not only to explain primitive beliefs but also to define a proto-religious stage of cultural evolution. Religion, or more strictly the prerequisite for religion, he went on to define as "a belief in spiritual beings." Animism, then, is but one type of religion, namely, the belief that all things have souls, or, as it were, both a material and a spiritual "body" or aspect.
Tylor's theory of animism, and indeed his view of religion as a phenomenon that properly encompasses both primitive and so-called higher forms in a unified theory, provided the locus classicus of most anthropological work, including the formation of new theories, until well past the turn of the century. The main thrust of theorizing in this period was to reconstruct the origins of religious behavior itself, that is, to isolate the most elementary impulse, feeling, or experience that constituted the sine qua non of religion, and to place all forms of religious behavior on an evolutionary scale of development from this point of origin. It should be noted here that a shift in emphasis in anthropological studies occurred in an early reaction to what was deemed by many to be Tylor's excessively intellectualist view of human nature, at least as it was displayed regarding primitives. Increasingly anthropologists viewed human beings primarily as active creatures whose thought processes are subordinated to action: thought "rationalizes" action to the degree that ideas are formed only in reaction to deeds and to provide a more or less emotionally satisfying intellectual justification for them. It is here that the idea of power, in a variety of forms, began to play its part in the great quest for origins.
Animatism is the name given to a theory, formulated by R. R. Marett, that sought to build upon the work of Tylor. Although he accepted animism as a higher stage in religious development, Marett rejected the "intellectualist fallacy" inherent in the theory of animism insofar as it claimed to represent the first stage of religion. He suggested instead that primitives experience the world as fundamentally divided into the familiar and the unfamiliar. The unfamiliar object is so because it exhibits some sort of strangeness suggestive of hidden power. This he called variously "occult power" and "the sacred." To the compound of unusual and hidden power he added the notion of life in much the same sense that Tylor had used animus, that is, life or soul, except that he believed that, at the stage of animatism, the primitive mind had not yet made the leap from life or life force to separable soul. This meant that animatism could also properly be understood as "preanimism."
The full articulation of this theory was published in 1909 in Marett's The Threshold of Religion, but as early as 1900 he had made the first steps toward it in his establishment of the Oceanic word mana as a general category of religious experience. He based his usage primarily upon the work of R. H. Codrington (see The Melanesians, 1891), who reported that for many South Pacific island cultures, the religious system was based upon a single concept, which they called mana. Among the Melanesians, mana, the power that inhered in all things, had special significance for their religious and social system, because it could be concentrated in some objects and because it inhered in a concentrated form in some people. Indeed, the hierarchical structure of their society was justified upon the basis of the aristocrats' inborn great mana. Everything possessed some mana, and, in this respect, the term might be translated "the power of being." Since so much was made of its concentratability, however, in many cases the term is better rendered as "sacred." But for many scholars, particularly in the nineteenth century, this usage permitted an unacceptable broadening of the meaning of sacred, since mana could be transferred from one object or person to another. Many tended to classify this notion not as religious but as pertaining to magic. The fluidity of mana made it a kind of physical energy, or at least analogous to such an energy: the transfer could be affected by touching one mana-charged object with another with less mana ; in particular, a person of high mana could infuse an object with some of his or her mana by handling it.
It was not long after the publication of Codrington's findings that similar discoveries began to be made in other parts of the world. American anthropologists were especially active at this time, and the Huron orenda, the Lakota wakan, and the Algonquin manitou were soon added to the list of mana -like concepts. Later the Arabic barakah and East Asian terms such as the Chinese ling-pao and the Japanese kami were suggested as counterparts to the Melanesian idea of mana. From such evidence, Marett then posited a general psychological tendency of human beings to experience the world as well as themselves under the guise of a controlling religious concept: sacred or occult power. This view has had great influence among scholars. However, contemporary anthropology does not generally accept Marett's insistence that even the most elementary religious experience engrafts to the notion of power the assumption of personality—or, to put it another way, that mana and animatism are necessarily combined. It may, of course, be true in certain cultures, as he argued, that because mana most powerfully manifested itself in certain types of persons, it was treated as if it were the willpower of a human being, but it is not true in all cultures. And the value of the term mana is just in its use as a general descriptive category denoting a sacred power that is not in itself personal. Thus, in fact, the modern usage implies a psychological, if not necessarily chronological, priority to the idea of mana over even Marett's animatism.
Power and the Nature of Religion
In 1909, with the publication of Les rites de passage, Arnold van Gennep applied the label dynamistic to the theories of the origin of religion put forth by Marett (1900) and by J. N. B. Hewitt (1902), based upon the experience of the sacred as power. But van Gennep drew a sharp line between what he called dynamism, or the conceptual framework that assumed impersonal sacred power, and animism, which assumed that sacred power was personal. Since his goal was to classify rituals, and to a large extent to understand by means of classification, he did not enter into the theoretical debate concerning the origins of religion. Yet, because of the obvious value of his way of discussing ritual activities, his work did influence the theoretical debate, if only by showing that it was possible to make significant contributions to the study of religion without choosing a position concerning the question of origins.
No less implicit in van Gennep's work was the assumption of the centrality of the idea of power in religion, not so much in its own theorizing or attempts at self-understanding, but in its actual behavior. Thus he coined the term magico-religious to emphasize the practical side of human interaction with sacred power. All ritual activity he labeled as magical because it was in the realm of technique; that is, it sought to implement a practical goal, namely, to influence or even to manipulate the sacred power for useful purposes. It was, therefore, the efficacy of the sacred, its potentiality to effect change or to prevent change—in short, its power—that van Gennep emphasized in his basic insight that ritual, or, at any rate, many rituals, seek to effect transitions from one state or situation to another.
At about the same time that Marett and van Gennep were formulating their views of religion, other theories about the nature and, to some extent, the origin of religion were being formulated outside the conceptual circle of the new discipline of anthropology. Influenced by anthropological and ethnological studies, but operating in a very different intellectual framework, was Rudolf Otto, a theologian who took as his spiritual mentor Friedrich Schleiermacher. In Das Heilige (1917), Otto presented what might be called a phenomenological psychology of religion, in that he sought to describe the structure of human reaction to what is experienced as "the holy." Otto's work as a religious theorist, because of his attitude toward human nature and in his introspective approach to religion, may be considered a late flowering of the Romantic movement. He exhibits a qualified anti-intellectualist stance toward religious psychology: religion is an ineradicable part of human nature, present from the beginning, but, while religion itself admits of historical development, the psychological makeup of human beings, which makes religion possible, does not. Therefore, any religious experience, however far removed in time and space, can be understood by the modern student, because it shares a fundamental unity with all religion. Further, Otto appeals in a famous passage to the reader's own experience, rather than to his rational faculties, as the guarantor of the accuracy and usefulness of his descriptions:
The reader is invited to direct his mind to a moment of deeply-felt religious experience, as little as possible qualified by other forms of consciousness. Whoever cannot do this, whoever knows no such moments in his experience, is requested to read no farther; for it is not easy to discuss questions of religious psychology with one who can recollect the emotions of his adolescence, the discomforts of indigestion, or, say, social feelings, but cannot recall any intrinsically religious feelings. (Otto,  1923, p. 8)
The fundamental religious experience Otto termed as the feeling of the presence of "the numinous." In this, his theory closely approximates that of Marett's "occult power" (or mysterious power or the sacred). But Otto sought in a systematic way to show that this feeling existed psychologically prior to any conceptualization of a god or spirit or soul and, at the same time, was the religious sine qua non behind these concepts. As he put it, the "ideogram" of the numinous must be present in the "concept" of god, since the former is the nonrational, feeling component of the rational concept. The mental process by which ideograms become concepts he called "schematization."
Implicit in his argument is a tension between experience or feeling, on the one hand, and a priori ideas, on the other, since he wished to affirm both the priority of religious experience and the truth of certain religious concepts. Indeed, it is his strong allegiance to a belief in the superiority of Christian theological formulations that has been largely responsible for Otto's lack of influence in anthropology and in related disciplines concerned with the study of religion. Added to this was his insistence upon the sui generis character of religious experience, which tended to isolate religion from other psychological realms, such as the experience of beauty, sexual pleasure, or terror.
The heart of Otto's system is his description of the feelings that, to a greater or lesser extent and in varying mixtures, all religious experiences evoke. These are mysterium tremendum and mysterium fascinans. The ambivalence in the human response to the object of religion that has already been encountered in the dichotomy of mana and taboo, the positive and negative aspects of sacred power. In Otto's schema, van Gennep's work focused primarily upon the fascinans aspect, since the efficacy of sacred power is necessary for ritual goals to be realized, although of course van Gennep also discussed rituals of avoidance. It is particularly in the analysis of the negative side of the dichotomy that Otto's unique contribution to the understanding of religious experience can be seen. Choosing as his illustrative data primarily the canonical literature of Christianity, but supplementing it with references to such famous Christian virtuosi as Martin Luther as well as to Islamic and Hindu mystics, he documents minutely the daunting presence of the numinous in the more complex or "higher" religions. For purposes of exposition, he divides his first category into two. The first is mysterium, which he explains as having its closest analogy in the feeling of uncanniness that irrationally can seize one when, for example, one is listening to ghost stories or passing graveyards. This feeling emphasizes the radical otherness (das ganz Andere ) of the numinous and results in a uniquely religious dread. If, according to Otto, this feeling is allowed to predominate in the religious experience, aberrations such as demon worship can result. To this is inextricably joined the element of tremendum, the overpoweringness of the numinous, whose ideogram in Christianity is God's wrath. Moving from experience (der Moment ) to ideogram to developed theological concept, tremendum becomes divine omnipotence.
Tremendum, therefore, is the place in Otto's schema where the experience of sacred power has its proper location. He further elaborates its effects by the ideogram of "creature consciousness," the elementary feeling articulated by the thought of having been created, assembled, as it were, as a kind of contingent and therefore somewhat arbitrary and temporary configuration with no intrinsic merit or value or power. To sense this is to feel that one is nothing over against the infinite power and presence of the Other. Out of it come the relatively sophisticated ideas of creation and of sin. Notice that sin is now partly derived not only from the memory of having contravened a law or broken a taboo; it is also intrinsic to the religious encounter itself, particularly from the encounter with power in its overwhelming immensity. Of course it is here referred to the joining or schematization of tremendum and the doctrine of sin, especially of original sin, which Otto argues finally makes the Christian concept of sin credible and intellectually satisfying.
It could be argued that the element of fascinans, or attraction, in the numinous experience also implies a tacit recognition of kratophany, but in Otto's own handling of it, fascinans is expressed in such terms as love, duty, and the motivation to pursue the religious life. It is an elementary recognition or experience of value rather than a perception of utility or status, which seem to predominate in the idea of mana.
Mircea Eliade and the History of Religions
Mircea Eliade linked his own work in the phenomenology of religion with that of Otto when in The Sacred and the Profane (1957) he expressed admiration for Otto's descriptions of religious experience. Yet he sought to establish, at the same time, a different perspective, one that took as its starting point the categories of the sacred/profane dichotomy first given prominence by the French sociologist Émile Durkheim. Eliade was concerned with what might be called collective psychology, rather than a psychology of individual, particular experiences. His work has sought to catalog and explain (as in Patterns in Comparative Religion, 1958) the great collective representations, that is, symbols, by which religious meaning is mediated in a variety of cultural contexts. In accepting Otto's description of the "irrational" aspect of encounters with the sacred, Eliade infuses his use of the term sacred with specific meaning that includes power as a central element. Thus the encounter with sacred power is seen in the structure of the symbols of the sacred, while power is one of the necessary attributes of the sacred.
Eliade is perhaps most like Otto when he discusses archaic techniques of ecstasy, as he does at length in his Shamanism (1951). Here he shows that the shaman often unwillingly encounters, and is possessed by, sacred power in an unequal test of strength that leaves the human personality transformed. The result is the ability ritually to achieve ecstasis, or a projection of self out of self, in order to tap the power of sacred realities as a religious specialist serving the community. But the interpretation of shamanism is not restricted to psychological aspects: the symbols, for example, of drum and "flying costume," by which shamanic rituals are accomplished, are also presented, as well as myths that both buttress and explain the worldview of shamanism.
Throughout his works dealing with archaic religion, Eliade has emphasized the creative power of myth and of the sacred beings whose stories myths are (see Myth and Reality, 1963). Of course, this power is understood by those for whom myths still live as the power of the sacred itself, made knowable and thus usable through myth. For Eliade, cosmogonic myth is perforce the most important type, since it taps into the ubiquitous psychological tendency that he has termed the assumption of the "prestige of origins." Here, knowledge of the origin of a thing is equivalent to having power over that thing. Thus knowledge of the origin of the world as contained in the cosmogonic myth gives human beings power over their entire environment. Rituals that celebrate this knowledge by reiterating the myth, or, more dramatically, by reenacting it, are at least very useful to the scholar in attempting to grasp the meaning of a religious worldview. Eliade has also noted that the prestige of origins and the supposed power of origins continue to function psychologically, often unconsciously, in modern secular contexts.
The sacred has power, in Eliade's view, both to make the world meaningful by providing a religious worldview and to provide a means of escape from a desacralized and therefore meaningless world (Cosmos and History, 1949). His work on yoga (Yoga: Immortality and Freedom, 1954) details this latter function of sacred power in Hinduism and Buddhism. In samadhi, the yogin achieves the final stage in the personal journey by which the true self realizes its identity with the sacred. This state brings with it not only the bliss of a superconsciousness but also a number of sacred powers: knowledge and sensitivity beyond the ordinary as well as psychophysical powers (siddhis ) that mark the accomplished practitioner of yoga.
In his discussion of yoga, Eliade also touches upon an especially revealing concept of Hinduism, namely, tapas. This idea, which is very old in the Indian subcontinent, can be rendered as "the power of asceticism," or "the sacred power by which the world was created." Sometimes, indeed, in later popular folk tales and myths, tapas becomes the power of desire and of sexual potency, which both creates all beings and threatens all with dissolution. Yoga as an ascetic discipline is thought to tap the power of tapas, for it is sometimes understood that tapas is the power by which the extraordinary accomplishment of final liberation is won. Among the devotional cults of modern Hinduism, the Śaivas honor Śiva, the phallic creator god who is also the prototype of all yogins.
Belief in the power of sacred models to raise individuals to new states of being (see Rites and Symbols of Initiation, 1958), especially as this power is brought to bear in rituals, is documented in Eliade's work on "initiation scenarios," which are so widespread even in secular literature and fantasy. These survivals of living symbol systems continue to haunt modern people's dreams and imaginative creations. In archaic societies, these symbols of death and rebirth—of being swallowed by a monster, for example—are especially significant ways by which the power of the sacred can bring about the transition from childhood to adulthood, from ordinary living human being to powerful ancestor, from ordinary human to powerful shaman. In salvation religions, these same techniques and symbols are employed in the crucial transition from a state of damnation to that of salvation and beatitude.
The amazing ability of symbols to endure through the ages and despite profound cultural changes, as Eliade has documented in the historical portions of his work, testifies to the power that symbols wield in human life. These powerful symbols appear to possess almost a life of their own, inasmuch as they are constitutive of the human personality. To possess sacred power is at the same time to be possessed by it, a view that Rudolf Otto would heartily support and one that the psychologist C. G. Jung emphasized with his theory of archetypes.
Phenomenologically, it is impossible to determine the source of symbols either within or without the self that experiences them. Indeed, Jung regarded religion as a traditional response to especially powerful symbols that arose from the hidden energy- and meaning-centers of the psyche, that is, the archetypes. What a symbol in a dream of myth masked or partially revealed of an archetype could be determined from the human reaction to it. Archetypal symbols engender great fear, awe, and longing: they are the mainsprings of the deepest and strongest emotions, and are experienced as numinous centers of power:
When an archetype appears in a dream, in a fantasy, or in life, it always brings with it a certain influence or power by virtue of which it either exercises a numinous or fascinating effect, or impels to action.… Owing to their specific energy—for they behave like highly charged autonomous centres of power—they exert a fascinating and possessive influence upon the conscious mind and can thus produce extensive alterations in the subject. (Jung, 1953, p. 80)
The very process of maturation, both culturally and individually, which Jung believed to be the main focus of religious behavior, is a process of the ever-deepening experience of archetypal images and of the progressive transformation of archetypally generated symbols.
Thus, in Jung's thought the ideas of power and of religious experience were strongly associated. Religion was one way of dealing with these internal structures although by no means the only way. On the other hand, religious behavior was derived from these structures as the driving force of both thought and action.
Van der Leeuw and the Phenomenology of Religion
One major work on the nature of religion requires special mention, because it uses the idea of power as its central organizing principle. This is Gerardus van der Leeuw's Phänomenologie der Religion (1933), translated into English as Religion in Essence and Manifestation (1938). Van der Leeuw begins his ambitious work with a discussion of the experience of power as the founding impetus of religion:
The religious man perceives that with which his religion deals as primal, as originative or causal; and only to reflective thought does this become the Object of the experience that is contemplated.… Theory, and even the slightest degree of generalization, are still far remote; man remains quite content with the purely practical recognition that this Object is a departure from all that is usual and familiar; and this again is the consequence of the Power it generates. (van der Leeuw, 1938, p. 23)
He thus describes a pretheoretical mode of perception in which the experience of power and otherness are combined, and in which the notion of efficacy dominates. This power originates and causes events; it is thus fundamentally creative.
Van der Leeuw quickly finds the traditional language of scholarship to be misleading, since it improperly distinguishes religion and magic at this elemental level:
It is precisely a characteristic of the earliest thinking that it does not exactly distinguish the magical, and all that borders on the supernatural, from the powerful; to the primitive mind, in fact, all marked "efficiency" is per se magical, and "sorcery" eo ipso mighty.… Magic is certainly manifested by power; to employ power, however, is not in itself to act magically, although every extraordinary action of primitive man possesses a tinge of the magical. (ibid., pp. 24-25)
Although he often calls this elemental level of religiosity "primitive," he rejects the hypothesis that it exists as a stage in religious evolution. For him, the term designates a level of thought and experience that is found, to a degree, in all religions at all times. Further, van der Leeuw considers the notion of an ordering power, or sacred order, as in the Sanskrit ṛta or the Chinese dao, to be theories about power as advanced as the notion of an individual soul as a personal center of power.
Van der Leeuw interprets taboo as perhaps the most elemental reaction to the experience of sacred power: one is characteristically fearful in the face of the disparity of power, and taboo is an attempt to mount some defense against it. Indeed, he derives the Roman religio from an experience of dread. Thus religion for the Romans was a system of taboos set up in response to the awesome appearance of sacred power. "Observance," he writes, "is just benumbed awe which, at any moment, can be revived" (ibid., p. 50).
The entire first part of Religion in Essence and Manifestation is a long essay demonstrating that the notion of power is the key to understanding a wide variety of religious phenomena. For example, celestial symbols are an important part of many religions because they manifest cosmic power in such a way that humans can model their behavior upon the orderly motions of heavenly bodies, thus tapping their great power. Again, animal cults and totemism van der Leeuw explains as an attempt by humans to obtain for themselves the powers that animals control by virtue of their superior strength and skills, such as the ability to fly. The totem animal is especially significant in this regard because it "is a sort of reservoir for the potency of the tribe or clan" (ibid., p. 79). Angels represent a projection or emanation (they are "messengers") of specific powers of gods; sacred kingship is a recognition that the power of the most powerful person is, in part, sacred power, while belief in salvation implies faith in an extraordinary power of transformation.
Part 2 of this work takes up the reaction to sacred power as apprehended within: that is, the effect of the experience of power on human lives. Here religious functionaries, such as priest or shaman, are discussed, as well as the transformed life of the saint. Finally, religious organization, the social reaction to power, is sketched out.
Further description of van der Leeuw's work must founder because of his own interpretation of the phenomenological task: he eschewed any conscious hermeneutic or theory of religion as false to the data. Thus his work cannot be neatly summed up by reference to a relatively simple theoretical model. But in much of his work, the basic experience of power functions as much as a heuristic device as a basic insight into the nature of religion.
Another scholar who has influenced the notion of religious power held by students of religion in recent years is Georges Dumézil, who sought to develop some structural tools for dealing not with all religions but with that large class of religions known to have been derived from Indo-European cultures. His fundamental thesis is that the gods of Indo-European peoples reflect, and in turn are reflected in, the social structure of a given culture. This structure, in three main divisions, can be described in terms of the functions, or typical activities, performed by the gods or social classes in question. Although this thesis has far-reaching implications, most important for present purposes is the fact that in many cultures, most clearly in ancient India in the Vedic literature, these functions, in turn, seem to be based upon different concepts of power. Thus, because the concerns of the third-function gods are fecundity and productivity in the terrestrial sphere, they possess a special power or energy that controls and thus either promotes or inhibits the growth of herds or the abundance of harvests. This power was often thought of as sexual in nature.
But it is in the second and first functions, as Dumézil defined them, that differences in the basic nature of power become most apparent. Here he distinguishes sharply between the mysterious, hidden, even magical, power of the first-function gods and the merely physical power wielded by the gods of the second function. The second function belongs to the warrior, in India especially to Indra, who slew the cosmic demon Vrtra, and who was the protector of the Aryan tribes and the leader of the human warriors. Indeed, so important did this physical power become that there is evidence in the Ṛgveda that Indra to some extent replaced Varuna, the primary first-function god. Varuna and Mitra together are the representative of the function of sovereignty, whose position at the apex of the hierarchy of gods and humans was, originally at least, assured by the power they wielded. The first function Dumézil characterizes in general as celestial, priestly, and concerned with the exercise of magical and juridical sovereignty. Varuna especially is "a great sorcerer, disposed more than any other on the level of sovereignty to maya, magic which creates forms either temporary or permanent, disposed also to the knots in which he binds the guilty, a capture both immediate and irresistible" (Dumézil, 1968–1973, vol. 1, p. 148).
Coupled both to the characteristic celestial symbolism and to the idea of mysterious power is the association of Varuna and Mitra with the cosmic order, ṛta. Increasingly subservient to this impersonal order, the first-function gods nonetheless reflect and to a degree wield the very power by which the cosmos moves. This dynamism was especially impressive because the means of its motion was unseen: just as the stars or the sun followed their preordained courses; just as the seasons followed their patterns and other events such as disease occurred as punishments whose agents or mechanism, so to speak, could not be discovered by means of the ordinary senses; just so did the sovereign gods control the very power by which the world was ordered and by which its order was maintained. Physical power, the power of Indra and of war, could be understood, if not always defended against. Even the enormous physical power of a god was still physical and palpable, and therefore of a fundamentally different nature than was maya, the unseen and all the more frightening power of Varuna.
In the human realm, according to Dumézil's thesis, the social structure also reflected these different types of power. Of course it is the brahman caste, the hereditary priests, who wield Varuna's power, to some degree, because of their knowledge of the rites of sacrifice. In the cult, the priests function as mediators of sovereign sacred power: the words and actions of the rituals place in the priestly hands this same mysterious power, which is the power to influence cosmic forces for the benefit of humans.
Although Dumézil's point of departure is the Vedic texts of India, he applies this schema also to later Indian epics as well as to Persian, Greek, and other European religious literature. Beyond this, other scholars have sought to extend the three function theory to non-Indo-European cultures as well. Most notable of these, perhaps, is Atsuhiko Yoshida, whose "La mythologie japonaise: Essai d'interpretation structurale" (1961–1963) is the most thorough attempt to apply these categories not so much in order to show Indo-European influences upon Japanese mythology but as a useful interpretive tool.
Power, Magic, and Charisma
The use of the term magic has had a checkered career, both within Christian theological circles and within the realm of comparative religion or history of religions (Religionswissenschaft ). On the one hand, it has shared the pejorative connotations of such terms as superstition and idolatry in its emic or confessional evaluation; on the other hand, as evidenced by such compounds as magico-religious, from the etic viewpoint the term has been used in a purely descriptive way, as, for example, in the work of Arnold van Gennep, noted above. From this latter perspective, magic denotes simply sacred power experienced as impersonal and, to a degree, manipulatable: it is power in its most useful mode, since it can be turned to one's advantage with what might be called a minimum of harmful side effects. Providing only that the formulae and rituals are properly followed, results are predictable, even automatic. For many theologically inclined thinkers, this notion, and even more the attitude toward the sacred that it implies, must necessarily be a "lower" form of religion, or degenerate religion—or perhaps not religion at all. This is because it is felt to be incompatible with the proper sense of reverence and dependence due to a personal god as in Christianity or Judaism. From this perspective, to treat God as an object of magic is to blaspheme since this tends to reduce the majesty and freedom of the deity.
The lack of consensus among scholars as to the proper definition and use of the term magic reflects not so much differences in perception as differences in the purposes to which the data are put. From the purely descriptive point of view, a distinction between magic and religion, or between magical religion and pure religion, has proved practically impossible to make. But from the normative, theological point of view, the term magic has proved too useful a term to be easily given up, since it delineates what is felt to be a theologically unacceptable attitude toward the power of God. Thus, even when a pejorative sense is not intended in descriptive works, it is often improperly assumed by many readers.
Examples of the difficulties that lie in wait for those who would distinguish between a manipulative approach to the sacred and a properly humble and propitiatory approach are easily produced. Subtle psychological distinctions must be made, since the existential concern of all religious people for their own welfare makes a totally unselfserving approach to sacred power improbable, if not impossible, for ordinary human beings. Put another way, one may ask how often Christians pray for forgiveness of sins out of nothing more than a pure and unselfish love of their god? Or again, rites of passage, which are ubiquitous, seek always a more or less definite personal or communal gain—but who can assess with complete certainty the motivation of the participants? Discounting "manipulativeness" can lead to a restriction of the term religion to such an extent that it is lost as a useful descriptive term.
Another conceptual tool relating to religious power is charisma, a term made popular by the sociologist of religion Max Weber (in Religionssoziologie, 1922), who defined it as the authority by which individuals were accorded status and power over others or, related to that, by which the functions or offices themselves—regardless of the officeholder—were felt to be worthy of respect. Indeed, Weber expressly linked charisma both to mana and to the Iranian maga (Skt., maya ), from which the word magic is derived. Looked at closely, it may be seen that the notion of charisma, at least from the limited horizon of sociology, is rather mysterious. That is to say, the reason or means whereby one person is accorded this respect, or is seen as having a special inner power of attraction, is not explained or well understood. Certainly such things as character, unusual skills, great stature or strength, or force of mien or manner all seem to contribute, but, finally, charisma remains a relational term that classifies the reaction of others to the person whom scholars then label as charismatic.
The Chinese religious tradition offers a concrete example of belief in charisma, and even of theorizing about it within two ancient systems of thought, namely, Confucianism and Daoism. These two religions, although often antagonistic, nonetheless share a common origin and a number of common ideas. Two are especially relevant here: dao, or cosmic order, and de, variously translated as "virtue," "character," "power," or "charisma." It is possible to view these two concepts not only as closely associated in Chinese thought but as two aspects of a single reality: sacred power. Dao is in many ways similar to the Sanskrit ṛta, in that it is not only order but also the power that drives a dynamic universe. All things ultimately derive from dao (Laozu appropriately calls it "the mother of all things"), and all things move and change according to its "laws." To be sure, it is not entirely knowable, although Confucianism is more optimistic on this point, with its emphasis on study of the way of the ancients and its belief that dao is perfectly embodied in li (ritual or decorum).
When dao is perfectly embodied in a person, then he is called a sage. Such a one is as perfect an exemplar of the universal dao as a human being can be. To be a sage is to be perfectly in harmony with dao. But taken from the point of view of the individual, such a one has great de or personal power. This power, like dao, although it may be embodied in a person, is not in itself personal: it is without consciousness, or will, or emotion; it has no purpose. The intrinsic power of a sage is expressed, both in Confucianism and in Daoism, in the image of the sage-king Shun, who "acted without action"—yet all things were accomplished, and the empire was at peace. Shun is also likened to the pole star, which merely sits facing south, while all things revolve around it in a kind of cosmic ballet.
This de or charisma is brought down to earth, as it were, in the Confucian ideal of the junzi the "superior man" or "true gentleman," who also brings about by example, by ritual, and by the power of his presence the longed-for proper ordering of human society. It is not, of course, that he does nothing; rather, he is so well attuned to dao (or to "heaven," tian ) that whatever he chooses to do will be the correct thing in the circumstances. When such a person is a ruler, or, one might say after Weber, when charisma of person is combined with charisma of office, one has an especially powerful force for harmony. Interestingly, however, even here, at least in the more mystical Daoist writings, a sage does not will the right, does not arrive by careful thought or logical deduction at the right course of action; rather, because he is a sage, such action will spontaneously occur, sometimes with the sage as direct agent, but sometimes at the hands of others mysteriously influenced by him.
This mysteriously acting power, action at a distance and without conscious will, sounds in many ways like the Vedic maya. It is sacred power, at work in the human world, that reflects and ultimately is one with the sacred power that underlies all activity in the world of nature.
Is such a belief crude magic, or perhaps mere superstition? Some would answer in the affirmative. Certainly it insists upon the impersonal nature of the sacred and of the workings of sacred power. And the will to manipulate this power to benefit self, or the society as a whole, is strong, especially in Confucianism. Yet there is also awe and reverence for the power: it is difficult to gain, and it has its own ways. Others would claim that this example shows the impossibility of separating magic and religion, that they are inextricably merged into the idea of sacred power and into the active responses of human beings as they have perceived that power over the millennia of religious history.
In religions having a central concern for extra-worldly salvation, the way in which the power by which such a transformation can be effected has been understood has resulted in unresolved and perhaps unresolvable controversies. The early fifth century struggle between the Christian thinkers Pelagius and Augustine of Hippo concerned two very different assessments of that most elusive form of power known as human freedom. Did human beings, as Pelagius argued, have the power within themselves to live sinless lives and thus achieve salvation through their own efforts? Or, as Augustine argued, did the Fall that occurred in the Garden of Eden taint all humans descended from Adam and Eve to such an extent that they were incapable of perfection without divine aid, or grace? Roman Catholic orthodoxy eventually declared for Augustine's position, although the issue has continued to this day to exercise Christian theologians, and it played a major role in the Protestant Reformation. Strikingly similar was the controversy that raged in the thirteenth century in Japan with the rise of the Pure Land movement in Buddhism. Shinran witnessed to the all-encompassing "other-power" (ta-riki ) offered by the Buddha Amitabha that was available through faith to ordinary believers; this he juxtaposed to the use of "self-power" (ji-riki ) by the traditional monastic forms of Buddhism, by which nirvana itself could be achieved. By contrast Augustine saw the fall into sin, and thus also relative powerlessness, to be positioned at the beginning of time with its effects pervading all human existence until the end of history; whereas Shinran's "fall" occurred within history. Shinran believed that the degenerate age (mappo ), characterized by humans' decreased powers, had arrived, but it had come long after the founding of the path to salvation by the historical Buddha. Such parallel developments so far separated in time and space strongly suggest that there exists a universal deep structure of meaning to the conceptualization of power as it relates to salvation.
To be sure, many religions know of powers associated with the persons of religious elites, especially perhaps the power of healing. This power may be understood to be ultimately from God or the gods, the practitioner merely acting as a channel for it, as in western monotheism, or it may be seen as the product of the practitioner's own spiritual accomplishments. In the latter category may be placed the many forms of Buddhism that know the powers (siddhi ) that often accompany yogic meditative accomplishments, a tradition that is also reflected in the pan-Indian idea of tapas, depicted in many Hindu tales of the puranic period as a power generated by yoga that is potentially so strong as to threaten the very sovereignty of the gods. The relationship between such forms of power and the ultimate goal of salvation is sometimes ambiguous.
Codrington, R. H. The Melanesians: Studies in Their Anthropology and Folklore (1891). New Haven, 1957.
Dumézil, Georges. L'ideologie tripartie des Indo-Européens. Brussels, 1958.
Dumézil, Georges. Mythe et epopée. 3 vols. Paris, 1968–1973.
Eliade, Mircea. Cosmos and History: The Myth of the Eternal Return (1949). New York, 1959.
Eliade, Mircea. Shamanism: Archaic Techniques of Ecstasy (1951). Rev. & enl. ed. New York, 1964.
Eliade, Mircea. Yoga: Immortality and Freedom (1954). 2d ed., rev. & enl. Princeton, 1969.
Eliade, Mircea. The Sacred and the Profane (1957). New York, 1959.
Eliade, Mircea. Patterns in Comparative Religion. New York, 1958.
Eliade, Mircea. Rites and Symbols of Initiation: The Mysteries of Birth and Rebirth (1958). New York, 1975.
Gennep, Arnold van. The Rites of Passage (1909). Chicago, 1960.
Hewitt, J. N. B. "Orenda and a Definition of Religion." American Anthropologist 4 (1902): 33–46.
Jensen, Adolf E. Myth and Cult among Primitive Peoples (1951). Chicago, 1963.
Jung, C. G. "The Psychology of the Unconscious" (1943). In The Collected Works of C. G. Jung, vol. 7. Princeton, 1953.
Leeuw, Gerardus van der. Phänomenologie der Religion. Tübing-en, 1933. Translated by J. E. Turner as Religion in Essence and Manifestation, 2 vols. (1938; reprint, Gloucester, Mass., 1967).
Marett, R. R. "Pre-animistic Religion." Folklore 11 (1900): 162–182.
Marett, R. R. The Threshold of Religion. London, 1909.
Otto, Rudolf. Das Heilige. Breslau, 1917. Translated by John W. Harvey as The Idea of the Holy (1923; 2nd ed., London, 1960).
Tylor, E. B. Primitive Culture (1871). 2 vols. New York, 1970.
Vries, Jan de. The Study of Religion. New York, 1967.
Wach, Joachim. Sociology of Religion (1944). Chicago, 1962.
Weber, Max. Religionssoziologie. Tübingen, 1922. Translated by Ephraim Fischoff as The Sociology of Religion (Boston, 1963).
Yoshida, Atsuhiko. "La mythologie japonaise: Essai d'inter-pretation structural." Revue de l'histoire des religions 160 (1961): 47–66; 161 (1962): 25–44; and 163 (1963): 225–248.
Yoshido, Atsuhiko. "Nihon shinwa to In-o shinwa." In Nihon shinwa no hikaku kenkyu, edited by Obayashi Taryo. Tokyo, 1974.
Alan L. Miller (1987 and 2005)
In approaching the study of politics through the analysis of power, one assumes, at a minimum, that relations of power are among the significant aspects of a political system. This assumption, and therefore the analysis of power, can be applied to any kind of political system, international, national, or local, to associations and groups of various kinds, such as the family, the hospital, and the business firm, and to historical developments.
At one extreme, an analysis of power may simply postulate that power relations are one feature of politics among a number of others–but nonetheless a sufficiently important feature to need emphasis and description. At the other extreme, an analyst may hold that power distinguishes “politics” from other human activity; to analysts of this view “political science, as an empirical discipline, is the study of the shaping and sharing of power” (Lasswell & Kaplan 1950, p. xiv).
In either case, the analyst takes it for granted that differences between political systems, or profound changes in the same society, can often be interpreted as differences in the way power is distributed among individuals, groups, or other units. Power may be relatively concentrated or diffused; and the share of power held by different individuals, strata, classes, professional groups, ethnic, racial, or religious groups, etc., may be relatively great or small. The analysis of power is often concerned, therefore, with the identification of elites and leadership, the discovery of the ways in which power is allocated to different strata, relations among leaders and between leaders and nonleaders, and so forth.
Although the approach to politics through the study of power relations is sometimes thought to postulate that everyone seeks power as the highest value, analysts of power generally reject this assumption as psychologically untenable; the analysis of power does not logically imply any particular psychological assumptions. Sometimes critics also regard the analysis of power as implying that the pursuit of power is morally good or at any rate that it should not be condemned. But an analysis of power may be neutral as to values; or the analyst may be concerned with power, not to glorify it, but in order to modify the place it holds in human relations and to increase the opportunities for dignity, respect, freedom, or other values (Jouvenel 1945; Lasswell & Kaplan 1950; Oppenheim 1961, chapters 8, 9).
Indeed, it would be difficult to explain the extent to which political theorists for the past 25 centuries have been concerned with relations of power and authority were it not for the moral and practical significance of power to any person interested in political life, whether as observer or activist. Some understanding of power is usually thought to be indispensable for moral or ethical appraisals of political systems. From a very early time–certainly since Socrates, and probably before–men have been inclined to judge the relative desirability of different types of political systems by, among other characteristics, the relations of power and authority in these systems. In addition, intelligent action to bring about a result of some kind in a political system, such as a change in a law or a policy, a revolution, or a settlement of an international dispute, requires knowledge of how to produce or “cause” these results. In political action, as in other spheres of life, we try to produce the results we want by acting appropriately on the relevant causes. As we shall see, power relations can be viewed as causal relations of a particular kind.
It therefore seems most unlikely that the analysis of power will disappear as an approach to the study of politics. However, the fact that this approach is important and relevant does not shield it from some serious difficulties. These have become particularly manifest as the approach has been more earnestly and systematically employed.
The attempt to study and explain politics by analyzing relations of power is, in a loose sense, ancient. To Aristotle, differences in the location of power, authority, or rule among the citizens of a political society served as one criterion for differentiating among actual constitutions, and it entered into his distinction between good constitutions and bad ones [seeAristotle]. With few exceptions (most notably Thomas Hobbes) political theorists did not press their investigations very far into certain aspects of power that have seemed important to social scientists in the twentieth century [seeHobbes]. For example, most political theorists took it for granted, as did Aristotle, that key terms like power, influence, authority, and rule (let us call them “power terms”) needed no great elaboration, presumably because the meaning of these words was clear to men of common sense. Even Machiavelli, who marks a decisive turning point from classical–normative to modern–empirical theory, did not consider political terms in general as particularly technical. Moreover, he strongly preferred the concrete to the abstract. In his treatment of power relations Machiavelli frequently described a specific event as an example of a general principle; but often the general principle was only implied or barely alluded to; and he used a variety of undefined terms such as imperio, forza, potente, and autorita É [seeMachiavelli].
From Aristotle to Hobbes political theorists were mainly concerned with power relations within a given community. But external relations even more than internal ones force attention to questions of relative power. The rise of the modern nation-state therefore compelled political theorists to recognize the saliency of power in politics, and particularly, of course, in international politics (Meinecke 1924).
Thus political “realists” found it useful to define, distinguish, and interpret the state in terms of its power. Max Weber both reflected this tradition of “realism” and opened the way for new developments in the analysis of power [seeWeber, Max]. “Power’ (Macht) 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” (Weber  1957, p. 152). This definition permitted Weber to conclude that “the concept of power is highly comprehensive from the point of view of sociology. All conceivable & combinations of circumstances may put him [the actor] in a position to impose his will in a given situation” (p. 153). It follows that the state is not distinguishable from other associations merely because it employs a special and peculiarly important kind of power– force. In a famous and highly influential definition, Weber characterized the state as follows: “A compulsory political association with continuous organization (politischer Anstaltsbetrieb) will be called a ‘state’ if and in so far as its administrative staff successfully upholds a claim to the monopoly of the legitimate use of physical force in the enforcement of its order” (p. 154).
In his well-known typologies and his analyses of political systems, however, Weber was less concerned with power in general than with a special kind that he held to be unusually important– legitimate power, or authority.
Later theorists, practically all of whom were directly or indirectly influenced by Weber, expanded their objectives to include a fuller range of power relations. In the United States attempts to suggest or develop systematic and comprehensive theories of politics centering about power relations appeared in books by Catlin (1927; 1930), an important essay by Goldhamer and Shils (1939), and numerous works of the Chicago school–principally Merriam (1934), Lasswell (1936), and, in international politics, Morgenthau (1948). In the decade after World War ii the ideas of the Chicago school were rapidly diffused throughout American political science. [SeeMerriam.]
Power terms evidently cover a very broad category of human relations. Considerable effort and ingenuity have gone into schemes for classifying these relations into various types, labeled power, influence, authority, persuasion, dissuasion, inducement, coercion, compulsion, force, and so on, all of which we shall subsume under the collective label power terms. The great variety and heterogeneity of these relations may, in fact, make it impossible–or at any rate not very fruitful–to develop general theories of power intended to cover them all.
At the most general level, power terms in modern social science refer to subsets of relations among social units such that the behaviors of one or more units (the responsive units, Jr) depend in some circumstances on the behavior of other units (the controlling units, C). (In the following discussion, R will always symbolize the responsive or dependent unit, C the controlling unit. These symbols will be used throughout and will be substituted even in direct quotations where the authors themselves have used different letters.) By this broad definition, then, power terms in the social sciences exclude relations with inanimate or even nonhuman objects; the control of a dog by his master or the power of a scientist over “nature” provided by a nuclear reactor would fall, by definition, in a different realm of discourse. On the other hand, the definition could include the power of one nation to affect the actions of another by threatening to use a nuclear reactor as a bomb or by offering to transfer it by gift or sale.
If power-terms include all relations of the kind just defined, then they spread very widely over the whole domain of human relations. In practice, analysts of power usually confine their attention to smaller subsets. One such subset consists, for example, of relations in which “severe sanctions …are expected to be used or are in fact applied to sustain a policy against opposition”-a subset that Lasswell and Kaplan call power (1950, pp. 7475). However, there is no agreement on the common characteristics of the various subsets covered by power terms, nor are different labels applied with the same meaning by different analysts.
Despite disagreement on how the general concept is to be defined and limited, the variety of smaller subsets that different writers find interesting or important, and the total lack of a standardized classification scheme and nomenclature, there is nonetheless some underlying unity in the various approaches to the analysis of power. In describing and explaining patterns of power, different writers employ rather similar elements (compare Cartwright 1965). What follows is an attempt to clarify these common elements by ignoring many differences in terminology, treatment, and emphasis.
Some descriptive characteristics
For purposes of exposition it is convenient to think of the analysis of power in terms of the familiar distinction between dependent and independent variables. The attempt to understand a political system may then be conceived of as an effort to describe certain characteristics of the system: the dependent variables; and to explain why the system takes on these particular characteristics, by showing the effects on these characteristics of certain other factors: the independent variables. Some of the characteristics of a political system that analysts seek to explain are the magnitude of the power of the C’s with respect to the R’s, how this power is distributed in the system, and the scope, and domain, of control that different individuals or actors have, exercise, or are subject to.
Magnitude. Political systems are often characterized explicitly or implicitly by the differences in the “amounts” of power (over the actions of the government or state) exercised by different individuals, groups, or strata. The magnitude of C’s power with respect to R is thought of as measurable, in some sense, by at least an ordinal scale; frequently, indeed, a literal reading would imply that power is subject to measurement by an interval scale. How to compare and measure different magnitudes of power poses a major unsolved problem; we shall return to it briefly later on. Meanwhile, we shall accept the assumption of practically every political theorist for several thousand years, that it is possible to speak meaningfully of different amounts of power. Thus a typical question in the analysis of a political system would be: Is control over government highly concentrated or relatively diffused?
Distribution. An ancient and conventional way of distinguishing among political systems is according to the way control over the government or the state is distributed to individuals or groups in the systems. Aristotle, for example, stated: “The proper application of the term ‘democracy’ is to a constitution in which the free-born and poor control the government–being at the same time a majority; and similarly the term ‘oligarchy’ is properly applied to a constitution in which the rich and betterborn control the government–being at the same time a minority” (Politics, Barker ed., p. 164). Control over government may be conceived as analogous to income, wealth, or property; and in the same way that income or wealth may be distributed in different patterns, so too the distribution of power over government may vary from one society or historical period to another. One task of analysis, then, is to classify and describe the most common distributions and to account for the different patterns. Typical questions would be: What are the characteristics of the C’s and of the R’s? How do the C’s and R’s compare in numbers? Do C’s and R’s typically come from different classes, strata, regions, or other groups? What historical changes have occurred in the characteristics of C and R?
Scope. What if C’s are sometimes not C’s, or C’s sometimes R’s, or R’s sometimes C’s? The possibility cannot be ruled out that individuals or groups who are relatively powerful with respect to one kind of activity may be relatively weak with respect to other activities. Power need not be general; it may be specialized. In fact, in the absence of a single world ruler, some specialization is inevitable; in any case, it is so commonplace that analysts of power have frequently insisted that a statement about the power of an individual, group, state, or other actor is practically meaningless unless it specifies the power of actor C with respect to some class of R’s activities. Such a class of activities is sometimes called the range (Cartwright 1965) or the scope of C’s power (Lasswell & Kaplan 1950, p. 73). There is no generally accepted way of defining and classifying different scopes. However, a typical question about a political system would be: Is power generalized over many scopes, or is it specialized? If it is specialized, what are the characteristics of the C’s, the elites, in the different scopes? Is power specialized by individuals in the sense that Ca and Cb exercise power over different scopes, or is it also specialized by classes, social strata, skills, professions, or other categories?
Domain. C’s power will be limited to certain individuals; the R’s over whom C has or exercises control constitute what is sometimes called the “domain,” or “extension,” of C’s power (Lasswell & Kaplan 1950, p. 73; Harsanyi 1962a, p. 67). Typical questions thus might be: Who are the R’s over whom C has control? What are their characteristics? How numerous are they? How do they differ in numbers or characteristics from the R’s not under C’s control?
Given the absence of any standard unit of measure for amounts, distributions, scopes, domains, and other aspects of power, and the variety of ways of describing these characteristics, it is not at all surprising that there is an abundance of schemes for classifying political systems according to some characteristic of power. Most such schemes use, implicitly or explicitly, the idea of a distribution of power over the behavior of government. The oldest, most famous, and most enduring of these is the distinction made by the Greeks between rule by one, the few, and the many (see Aristotle, Politics, Barker ed., pp. 110 ff.). Some variant of this scheme frequently reappears in modern analyses of power (e.g., Lasswell & Kaplan 1950, p. 218). Often, as with Aristotle himself, the distribution of power is combined with one or more other dimensions (e.g., Dahl 1963, p. 38). Rough dichotomous schemes are common. One based on “the degree of autonomy and interdependence of the several power holders” distinguishes two polar types, called autocracy and constitutionalism (Loewenstein 1957, p. 29). American community studies have in recent years called attention to differences between “pluralistic” systems and unified or highly stratified “power structures” [seeCommunity, article Onthe study of community power]. In One study that compares four communities the authors developed a more complex typology of power structures by combining a dimension of “distribution of political power among citizens” with the degree of convergence or divergence in the ideology of leaders; the four types of power structures produced by dichotomizing these two dimensions are in turn distinguished from regimes (Agger et al. 1964, pp. 73 ff.).
Some explanatory characteristics
Given the different types of political systems, how are the differences among them to be explained? If, for example, control over government is sometimes distributed to the many, often to the few, and occasionally to one dominant leader, how can we account for the differences? Obviously these are ancient, enduring, and highly complex problems; and there is slight agreement on the answers. However, some factors that are often emphasized in modern analysis can be distinguished.
Resources. Differences in patterns or structures of power may be attributed primarily, mainly, or partly to the way in which “resources/” or “base values/” are distributed among the individuals, strata, classes, and groups in different communities, countries, societies, and historical periods. This is an ancient, distinguished, widespread, and persuasive mode of explanation, used by Aristotle in Greece in the fourth century B.C., by James Harrington in seventeenth-century England, by the fathers of the American constitution in the late eighteenth century, by Marx and Engels in the nineteenth century, and by a great many social scientists in the twentieth century. A central hypothesis in most of these theories is that the greater one’s resources, the greater one’s power. Although explanations of this kind do not always go beyond tautology (by defining power in terms of resources), logical circularity is certainly not inherent in this mode of explanation. However, there is no accepted way of classifying resources or bases. Harold Lasswell has constructed a comprehensive scheme of eight base values which, although not necessarily exhaustive, are certainly inclusive; these are power (which can serve as a base for more power), respect, rectitude or moral standing, affection, well-being, wealth, skill, and enlightenment (Lasswell & Kaplan 1950, p. 87). Other writers choose more familiar categories to classify resources: for example, in trying to account for the patterns of influence in one community, the author described the patterns of social standing; the distribution of cash, credit, and wealth; access to legality, popularity, and control over jobs; and control over sources of information (Dahl 1961, pp. 229 ff.).
Skill. Two individuals with access to approximately the same resources may not exercise the same degree of power (over, let us say, government decisions). Indeed, it is a common observation that individuals of approximately equal wealth or social status may differ greatly in power. To be sure, this might be accounted for by differences in access to other resources, such as the greater legality, bureaucratic knowledge, and public affection that fall to any individual who is chosen, say, to be prime minister of Britain or president of the United States. Another factor, however, one given particular prominence by Machiavelli, is political skill. Formally, skill could be treated as another resource. Nonetheless, it is generally thought to be of critical importance in explaining differences in the power of different leaders–different presidents, for example, as in Neustadt’s comparison of presidents Roosevelt, Truman, and Eisenhower (1960, pp. 152 ff.). However, despite many attempts at analysis, from Machiavelli to the present day, political skill has remained among the more elusive aspects in the analysis of power.
Motivations. Two individuals with access to the same resources may exercise different degrees of power (with respect to some scope) because of different motivations: the one may use his resources to increase his power; the other may not. Moreover, since power is a relationship between C’s and R’s, the motivations not only of the C’s but also of the R’s are important. One person may worship authority, while another may defy it. A number of writers have explored various aspects of motivations involved in power relations (e.g., Lasswell 1930; Rogow & Lasswell 1963; Cartwright 1959).
Costs. Motivations can be related to resources by way of the economists’ language of cost–a factor introduced into the analysis of power by a mathematical economist (Harsanyi 1962a; 1962b). In order to control R, C may have to use some of his resources. Thus C’s supply of resources is likely to have a bearing on how far he is willing to go in trying to control R. And variations in C’s resources are likely to produce variations in C’s power. C’s opportunity costs in controlling R–that is, what C must forgo or give up in other opportunities as a result of using some of his resources to control R–are less (other things being equal) if he is rich in resources than if he is poor in resources. In concrete terms, to a rich man the sacrifice involved in a campaign contribution of $100 is negligible; to a poor man the sacrifice entailed in a contribution of $100 is heavy. C’s willingness to use his resources to control R will also depend on the value to C of R’s response; the value of R’s response is, in turn, dependent in part on C’s motivations. The relationship may also be examined from R’s point of view. R’s opportunity costs consist of what he is then unable to do if he complies with C. In R’s case, as in C’s, his supply of resources and his motivations help determine his opportunity costs. Thus a power relation can be interpreted as a sort of transaction between C and ft.
Like all other approaches to an understanding of complex social phenomena, the analysis of power is beset with problems. At a very general level, attempts to analyze power share with many –perhaps most–other strategies of inquiry in the social sciences the familiar dilemma of rigor versus relevance, and the dilemma has led to familiar results. Attempts to meet high standards of logical rigor or empirical verification have produced some intriguing experiments and a good deal of effort to clarify concepts and logical relationships but not rounded and well-verified explanations of complex political systems in the real world. Conversely, attempts to arrive at a better understanding of the more concrete phenomena of political life and institutions often sacrifice a good deal in rigor of logic and verification in order to provide more useful and reliable guides to the real world.
There are, however, a number of more specific problems in the analysis of power, many of which have only been identified in the last few decades. Relevant work is quite recent and seeks (1) to clarify the central concepts, partly by expanding on the analogy between power relations and causal relations, (2) to specify particular subsets that are most interesting for social analysis, (3) to develop methods of measurement, and (4) to undertake empirical investigations of concrete political phenomena.
Power and cause
The closest equivalent to the power relation is the causal relation. For the assertion “C has power over R” one can substitute the assertion, “C’s behavior causes R’s behavior.” If one can define the causal relation, one can define influence, power, or authority, and vice versa (Simon [1947–1956] 1957, p. 5).
Since the language of cause is no longer common in the formal theoretical language of the natural sciences, it might be argued that social scientists should also dispense with that language and that insofar as power is merely a term for a causal relation involving human beings, power-terms should simultaneously be dispensed with. But it seems rather unlikely that social scientists will, in fact, reject causal language. For the language of cause, like the language of power, is used to interpret situations in which there is the possibility that some event will intervene to change the order of other events. In medical research it is natural and meaningful to ask, Does cigarette smoking cause lung cancer and heart disease? In social situations the notion of cause is equally or even more appropriate. What makes causal analysis important to us is our desire to act on causes in the real world in order to bring about effects–reducing death rates from lung cancer, passing a civil-rights bill through Congress, or preventing the outbreak of war.
To interpret the terms power, influence, authority, etc., as instances of causal relations means, however, that the attempt to detect true rather than spurious power relations must run into the same difficulties that have beset efforts to distinguish true from spurious causal relations. Some analysts have confronted the problem; others have noted it only to put it aside; most have ignored it entirely, perhaps on the assumption that if social scientists tried to solve the unsolved problems of philosophy they would never get around to the problems of the social sciences. Yet if power is analogous to cause –or if power relations are logically a subset of causal relations–then recent analyses of causality must have relevance to the analysis of power.
In the first place, properties used to distinguish causation also serve to define power relations: covariation, temporal sequence, and asymmetry, for example. The appropriateness of these criteria has in fact been debated, not always conclusively, by various students of power (e.g., Simon [1947–1956] 1957, pp. 5, 11, 12, 66; Dahl 1957, p. 204; Cartwright 1959, p. 197; Oppenheim 1961, p. 104).
Thus, the problem whether A can be said to cause B if A is a necessary condition for B, or a sufficient condition, or both necessary and sufficient, has also plagued the definition of powerterms. Some writers have explicitly stated or at least implied that relations of power mean that some action by C is a necessary condition for JR’s response (Simon 1953, p. 504; March 1955, p. 435; Dahl 1957, p. 203). Oppenheim has argued, however, that such definitions permit statements that run flatly counter to common sense; he holds that it would be more appropriate to require only that C’s action be sufficient to produce R’s response (1961, p. 41). Riker has suggested in turn that “the customary definition of power be revised… to reflect the necessary-and-sufficient condition theory of causality” (1964, p. 348). However, Blalock in his Causal Inferences in Non-experimental Research has shown that defining cause in terms of necessary and sufficient conditions leads to great practical difficulties in research. “In real-life situations we seldom encounter instances where B is present if and only if A is also present” (1964, p. 30); moreover, specifying necessary and sufficient conditions requires the researcher “to think always in terms of attributes and dichotomies,” whereas “there are most certainly a number of variables which are best conceived as continuously distributed, even though we may find it difficult to measure them operationally in terms of a specified unit of some kind” (p. 32). “The use of ‘necessary and sufficient’ terminology …may work well for the logician but not [for] the social scientist” (p. 34). Blalock’s criticism, and indeed his whole effort to explore problems of causal inference in nonexperimental research, are highly relevant to the analysis of power.
Aside from these somewhat rarefied philosophical and definitional questions, which many social scientists are prepared to abandon to metaphysicians or philosophers of science, the analogy between power and cause argues that the problem of distinguishing cause from correlation, or true from spurious causation, is bound to carry over into the analysis of power. And indeed it does. The difficulty of distinguishing true from spurious power relations has proved to be quite formidable.
The most rigorous method of distinguishing true from spurious causation is, of course, experimentation, and this would be the most rigorous method for distinguishing true from spurious power relations, provided the proper experimental conditions were present. Unfortunately, however, as in many areas of the social sciences, so too in the analysis of power, experimental methods have so far been of limited value, and for similar reasons. In nonexperimental situations the optimal requirements for identifying causal relations seem to be the existence of satisfactory interval measures, a large supply of good data employing these measures, and an exhaustive analysis of alternative ways of accounting for the observations (Blalock 1964). Unfortunately, in the analysis of power, existing methods of measurement are rather inadequate, the data are often inescapably crude and limited, a variety of simple alternative explanations seem to fit the data about equally well, and in any case the complexity of the relations requires extraordinarily complex models.
The shortage of relevant models of power may disappear in time. In fact, the causal analogue suggests that the development of a great array of carefully described alternative models to compare with observations is probably a prerequisite for further development in the analysis of power. Again, the analogy between power and cause readily reveals why this would seem to be the case. In trying to determine the cause of a phenomenon it is of course impossible to know whether all the relevant factors in the real world are actually controlled during an investigation. Consequently, it is never possible to demonstrate causality.
It is possible to make causal inferences concerning the adequacy of causal models, at least in the sense that we can proceed by eliminating inadequate models that make predictions that are not consistent with the data.… [Such] causal models involve (1) a finite set of explicitly defined variables, (2) certain assumptions about how these variables are interrelated causally, and (3) assumptions to the effect that outside variables, while operating, do not have confounding influences that disturb the causal patterning among the variables explicitly being considered, (ibid., p. 62)
If power relations are a subset of causal relations, these requirements would also be applicable in the analysis of power.
In analyzing power, why have analysts so rarely attempted to describe, in rigorous language at any rate, the alternative causal models relevant to their inquiry? There seem to be several reasons. First, students of power have not always been wholly aware that distinguishing true from spurious power relations requires intellectual strategies at a rather high level of sophistication. Second, the crude quality of the observations usually available in studying power may discourage efforts to construct elegant theoretical models. Third, until recent times the whole approach to power analysis was somewhat speculative: there were a good many impressionistic works but few systematic empirical studies of power relations. Of the empirical studies now available most are investigations of power relations in American communities undertaken since 1950. These community studies have provoked a good deal of dispute over what are, in effect, alternative models of causation. So far, however, investigators have usually not described clearly the array of alternative models that might be proposed to explain their data, nor have they clearly specified the criteria they use for rejecting all the alternatives except the one they accept as their preferred explanation.
Theories about power relations in various political systems are of course scattered through the writings of a number of analysts (e.g., Pareto 1916, volume 4; Mosca 1896, passim; Lasswell & Kaplan 1950, chapters 9, 10; Mills 1956; Dahl 1961; Rossi 1960; Polsby 1963; Parsons 1963a; 1963b). But a straightforward presentation of an empirical theory of power relations in political systems is a rarity. A notable exception is offered by March’s formulation of six models of social choice that involve, in some sense, relationships of power.
The analogy between cause and power calls attention to one further point: any attempt to develop an empirical theory of power will run headlong into the fact that a causal chain has many links; that the links one specifies depend on what one wishes to explain; and that what one wishes to explain depends, in part, on the theory with which one begins. In causal analysis, it is usually
… possible to insert a very large number of additional variables between any two supposedly directly related factors. We must stop somewhere and consider the theoretical system closed. Practically, we may choose to stop at the point where the additional variables are either difficult or expensive to measure, or where they have not been associated with any operations at all.… A relationship that is direct in one theoretical system may be indirect in another, or it may even be taken as spurious. (Blalock 1964, p. 18)
Some of the links that a power analyst may take as “effects” to be explained by searching for causes are the outcomes of specific decisions; the current values, attitudes, and expectations of decision makers; their earlier or more fundamental attitudes and values; the attitudes and values of other participants—or nonparticipants—whose participation is in some way significant; the processes of selection, self-selection, recruitment, or entry by which decision makers arrive at their locations in the political system; the rules of decision making, the structures, the constitutions. No doubt a “complete” explanation of power relations in a political system would try to account for all of these effects, and others. Yet this is an enormously ambitious task. Meanwhile, it is important to specify which effects are at the focus of an explanatory theory and which are not. A good deal of confusion, and no little controversy, are produced when different analysts focus on different links in the chain of power and causation without specifying clearly what effects they wish to explain; and a good deal of criticism of dubious relevance is produced by critics who hold that an investigator has focused on the “wrong” links or did not provide a “complete” explanation.
Classifying types of power
Even though the analysis of power has not produced many rigorous causal models, it has spawned a profusion of schemes for classifying types of power relations (e.g., Parsons 1963a; 1963b; Oppenheim 1961; French & Raven 1959; Cartwright 1965).
Among the characteristics most often singled out for attention are (1) legitimacy: the extent to which R feels normatively obliged to comply with C; (2) the nature of the sanctions: whether C uses rewards or deprivations, positive or negative sanctions; (3) the magnitude of the sanctions: extending from severe coercion to no sanctions at all; (4) the means or channels employed: whether C controls R only by means of information that changes R’s intentions or by actually changing R’s situation or his environment of rewards and deprivations. These and other characteristics can be combined to yield many different types of power relations.
As we have already indicated, no single classification system prevails, and the names for the various categories are so completely unstandardized that what is labeled power in one scheme may be called coercion or influence in another. Detached from empirical theories, these schemes are of doubtful value. In the abstract it is impossible to say why one classification system should be preferred over another.
Nonetheless, there are some subsets of power relations—types of power, as they are often called —that call attention to interesting problems of analysis and research. One of these is the distinction between having and exercising power or influence (Lasswell & Kaplan 1950, p. 71; Oppenheim 1961, chapters 2, 3). This distinction is also involved in the way anticipated reactions function as a basis for influence and power (Friedrich 1963, chapter 11).
To illustrate the problem by example, let us suppose that even in the absence of any previous communication from the president to Senator R, or indeed any previous action of any kind by the president, Senator R regularly votes now in a way he thinks will insure the president’s favor later. The senator calculates that if he loses the next election, he may, as a result of the president’s favorable attitude, be in line to receive a presidential appointment to a federal court. Thus, while Senator R’s voting behavior is oriented toward future rewards, expected or hoped for, his votes are not the result of any specific action by the president.
If one holds that C cannot be a cause of R if C follows R in time, then no act of the incumbent president need be a cause of Senator R’S favorable vote. Obviously this does not mean that Senator R’s actions are “uncaused.” The immediate determinant of his vote is his expectations. If we ask what “caused” his expectations, there are many possible answers. For example, he might have concluded that in American society if favors are extended to C, this makes it more likely that C will be indulgent later on. Or he may have acquired from political lore the understanding that the general rule applies specifically to relations of senators and presidents. Thus, the causal chain recedes into the senator’s previous learning—but not necessarily to any specific past act of the incumbent president or any other president.
This kind of phenomenon is commonplace, important, and obviously relevant to the analysis of power. Yet some studies, critics have said, concentrate on the exercise of power and fail to account for individuals or groups in the community who, though they do not exercise power, nonetheless have power, in the sense that many people try assiduously to anticipate their reactions (Bachrach & Baratz 1962). This failure may be a result of certain paradoxical aspects of having power that can make it an exceedingly difficult phenomenon to study.
For in the limiting case of anticipated reactions, it appears, paradoxically, that it is not the president who controls the senator, but the senator who controls the president—i.e., it is the senator who, by his loyal behavior, induces the president to appoint him to a federal court. Thus, it is not C who controls or even attempts to control R, but R who attempts to control C—and to the extent that R anticipates C’s reactions correctly, R does in fact control C. It is, then, not the king who controls the courtier but the courtier who controls the king.
Now if we examine this paradox closely we quickly discover that it arises simply because we have tried to describe the relationship between king and courtier, president and senator, C and R by distinguishing only one aspect, namely, the exercise of power. The courtier does indeed exercise power over the king by successfully anticipating the reactions of the monarch and thereby gaining a duchy. But it was not this that we set out to explain. For it is the king who has, holds, or possesses the capacity to confer that dukedom, and even though he does not exercise his power, he gains the willing compliance of the courtier.
What is it, then, that distinguishes having power from exercising power? The distinction could hinge upon the presence or absence of a manifest intention. We could define the exercise of power in such a way as to require C to manifest an intention to act in some way in the future, his action to be contingent on R’s behavior. By contrast, C might be said to have power when, though he does not manifest an intention, R imputes an intention to him and shapes his behavior to meet the imputed intention. If one were to accept this distinction, then in studying the exercise of power, one would have to examine not only R’s perceptions and responses but also C’s intentions and actions. In studying relationships in which C is thought to have power, even though he does not exercise it, one would in principle need only to study R’s perceptions, the intentions R imputes to C, and the bearing of these on R’s behavior. Carried to the extreme, then, this kind of analysis could lead to the discovery of as many different power structures in a political system as there are individuals who impute different intentions to other individuals, groups, or strata in the system.
The distinction between having and exercising power could also turn on the directness involved in the relation between C and R and on the specificity of the actions. In the most direct relationship R’s response would be tripped off by a signal directly from C. In this case, C is exercising power. But some relationships are highly indirect; for example, C may modify R’s environment in a more or less lasting way, so that R continues to respond as C had intended, even though C makes no effort to control R. In these cases, one might say that although C does not exercise control over R, he does have control over R. There are a variety of these indirect, or “roundabout,” controls (Dahl & Lindblom 1953, pp. 110 ff.).
Even more than with power terms themselves, notions of “more” or “less” power were in classical theory left to the realm of common sense and intuition. Efforts to develop systematic measures of power date almost wholly from the 1950s. Of those, some are stated partly in mathematical formulas, some entirely in nonmathematical language. Since the essential features can be suggested without mathematics, we shall describe these measures in ordinary language. (The reader should consult the sources cited for the precise formulations. Most of the best-known measures are presented and discussed in Riker 1964.)
In a rough way, the various criteria for measuring power can be classified into three types: gametheoretical, Newtonian, and economic.
Game-theoretical criteria. Shapley, a mathematician, and Shubik, an econometrician, have jointly formulated a “method for evaluating the distribution of power in a committee system” (1954). This is intended to measure the power accruing to a voter where the outcome or decision is determined exclusively by voting. In these cases the rules prescribe what proportion of votes constitutes a winning proportion (e.g., a simple majority of all committee members). Thus each member has a certain abstract probability of casting the last vote that would be needed to complete a winning coalition, in other words to occupy a pivotal position with respect to the outcome. By adding his vote at this crucial juncture, a voter may be conceived of as having made a particularly decisive contribution to the outcome; thus, gaining his vote might have considerable value to the other members of a coalition that would lose without his vote. Shapley and Shubik proposed measuring the power of a voter by the probability that he would be the pivotal voter in a winning coalition. Because their measure is entirely limited to voting situations and excludes all outcomes other than the act of voting itself, the utility of the measure is limited to cases where most of the other familiar elements of political life—various forms of persuasion, inducement, and coercion–are lacking. [SeeCoalitions]
Newtonian criteria. On the analogy of the measurement of force in classical mechanics, a number of analysts propose to measure power by the amount of change in R attributable to C. The greater the change in R, the greater the power of C; thus Ca is said to exert more power than Cb if Ca induces more change in Ra than Cb induces in Ra (or in some other R). Measures of this kind have been more frequently proposed than any other (Simon 1947—1956; March 1957; Dahl 1957; 1963, chapter 5; Cartwright 1959; Oppenheim 1961, chapter 8 ).
“Change in R” is not, however, a single dimension, since many different changes in R may be relevant. Some of the important dimensions of the “change in R” brought about by C that have been suggested for measuring the amount of C’s power are (1) the probability that R will comply; (2) the number of persons in R; (3) the number of distinct items, subjects, or values in R; (4) the amount of change in R’s position, attitudes, or psychological state; (5) the speed with which R changes; (6) the reduction in the size of the set of outcomes or behaviors available to R; and (7 the degree of R’s threatened or expected deprivation.
Economic criteria. Where the game-theoretical measure focuses on the pivotal position of C, and Newtonian measures on changes in R, a third proposal would include “costs” to both C and R in measuring C’s power. Harsanyi has argued that a complete measure of power should include (1) the opportunity costs to C of attempting to influence R, which Harsanyi calls the costs of C’s power, and (2) the opportunity costs to R of refusing to comply with C, which Harsanyi calls the strength of C’s power over R (1962a, pp. 68 ff.). The measure Harsanyi proposes is not inherently limited to the kinds of cost most familiar to economists but could be extended—at least in principle—to include psychological costs of all kinds.
Designing operational definitions
Empirical studies discussed by Cartwright (1965), March (1965), and others, and particularly community studies, have called attention to the neglected problem of designing acceptable operational definitions.
The concepts and measures discussed in this article have not been clothed in operational language. It is not yet clear how many of them can be. Yet the researcher who seeks to observe, report, compare, and analyze power in the real world, in order to test a particular hypothesis or a broader theory, quickly discovers urgent need for operationally denned terms. Research so far has called attention to three kinds of problems. First, the gap between concept and operational definition is generally very great, so great, indeed, that it is no always possible to see what relation there is between the operations and the abstract definition. Thus a critic is likely to conclude that the studies are, no doubt, reporting something in the real world, but he might question whether they are reporting the phenomena we mean when we speak of power. Second, different operational measures do not seem to correlate with one another (March 1956), which suggests that they may tap different aspects of power relations. Third, almost every measure proposed has engendered controversy over its validity.
None of these results should be altogether surprising or even discouraging. For despite the fact that the attempt to understand political systems by analyzing power relations is ancient, the systematic empirical study of power relations is remarkably new.
Robert A. Dahl
[See alsoCommunity, article onthe study of community power; Political Science; Political theory. Directly related are the entriesAuthority; Balance of power; Government; Military power potential; Power transition. Other relevant material may be found inCausation; Coercion; Decision making; International Relations; Monopoly; Oligopoly; Political process; Social control.]
Agger, Robert E.; Goldrich, Daniel; and Swanson, Bert 1964 The Rulers and the Ruled: Political Power and Impotence in American Communities. New York: Wiley.
AristotleThe Politics of Aristotle. Translated and edited by Ernest Barker. New York: Oxford Univ. Press, 1962.
Bachrach, Peter; and Baratz, Morton 1962 Two Faces of Power. American Political Science Review 56:947–952.
Blalock, Hubert M. Jr. 1964 Causal Inferences in Nonexperimental Research. Chapel Hill: Univ. of North Carolina Press.
Cartwright, Dorwin 1965 Influence, Leadership, Control. Pages 1—47 in James G. March (editor), Handbook of Organizations. Chicago: Rand McNally.
Catlin, George E. G. 1927 The Science and Method of Politics. New York: Knopf; London: Routledge.
Catlin, George E. G. 1930 A Study of the Principles of Politics, Being an Essay Towards Political Rationalization. New York: Macmillan.
Dahl, Robert A. 1957 The Concept of Power. Behavioral Science 2:201–215.
Dahl, Robert A. (1961) 1963 Who Governs? Democracy and Power in an American City. New Haven: Yale Univ. Press.
Dahl, Robert A. 1963 Modern Political Analysis. Englewood Cliffs, N.J.: Prentice-Hall.
Dahl, Robert A.; and Lindblom, Charles E. 1953 Politics, Economics, and Welfare: Planning and Politico-economic Systems Resolved Into Basic Social Processes. New York: Harper. → A paperback edition was published in 1963.
French, John R. P.; and Raven, Bertram 1959 The Bases of Social Power. Pages 150–167 in Dorwin Cartwright (editor), Studies in Social Power. Research Center for Group Dynamics, Publication No. 6. Ann Arbor: Univ. of Michigan, Institute for Social Research.
Friedrich, Carl J. 1963 Man and His Government: An Empirical Theory of Politics. New York: McGraw-Hill.
Goldhamer, Herbert; and Shils, Edward 1939 Types of Power and Status. American Journal of Sociology 45:171–182.
Harsanyi, John C. 1962a Measurement of Social Power,
Opportunity Costs, and the Theory of Two-person Bargaining Games. Behavioral Science 7:67–80
Harsanyi, John C. 1962b Measurement of Social Power in w-Person Reciprocal Power Situations. Behavioral Science 7:81–91.
Jouvenel, Bertrand De (1945) 1952 Power: The Natural History of Its Growth. Rev. ed. London: Batchworth. → First published in French.
Lasswell, Harold D. (1930) 1960 Psychopathology and Politics. New ed., with afterthoughts by the author. New York: Viking.
Lasswell, Harold D. 1936 Politics: Who Gets What, When, How? New York: McGraw-Hill.
Lasswell, Harold D.; and Kaplan, Abraham 1950 Power and Society: A Framework for Political Inquiry. Yale Law School Studies, Vol. 2. New Haven: Yale Univ. Press.→ A paperback edition was pulished in 1963.
Loewenstein, Karl 1957 Political Power and the Governmental Process. Univ. of Chicago Press.
March, James G. 1955 An Introduction to the Theory and Measurement of Influence. American Political Science Review 49:431–451.
March, James G. 1956 Influence Measurement in Experimental and Semiexperimental Groups. Sociometry 19:260–271.
March, James G. 1957 Measurement Concepts in the
Theory of Influence. Journal of Politics 19:202–226.
March, James G. (editor) 1965 Handbook of Organizations. Chicago: Rand McNally.
Meinecke, Friedrich (1924) 1957 Machiavellism: The Doctrine of Raison d’ttat and Its Place in Modern History. New Haven: Yale Univ. Press. → First published as Die Idee der Staatsrdson in der neueren Geschichte.
Merriam, Charles E. 1934 Political Power: Its Composition and Incidence. New York: McGraw-Hill. → A paperback edition was published in 1964 by Collier.
Mills, C. Wright 1956 The Power Elite. New York: Oxford Univ. Press.
Morgenthau, Hans J. (1948) 1967 Politics Among Nations: The Struggle for Power and Peace. 4th ed. New York: Knopf.
Mosca, Gaetano (1896) 1939 The Ruling Class (Elementi di scienza politica). New York: McGraw-Hill.
Neustadt, Richard E. 1960 Presidential Power: The Politics of Leadership. New York: Wiley. → A paperback edition was published in 1962.
Oppenheim, Felix E. 1961 Dimensions of Freedom: An Analysis. New York: St. Martins; London: Macmillan.
Pareto, Vilfredo (1916) 1963 The Mind and Society: A Treatise on General Sociology. 4 vols. New York: Dover. → First published as Trattato di sociologia generale. Volume 1: Non-logical Conduct. Volume 2: Theory of Residues. Volume 3: Theory of Derivations. Volume 4: The General Form of Society.
Parsons, Talcott 1963a On the Concept of Influence. Public Opinion Quarterly 27:37–62. → A comment by J. S. Coleman appears on pages 63–82; a communication by R. A. Bauer, on pages 83–86; and a rejoinder by Talcott Parsons, on pages 87–92.
Parsons, Talcott 1963b On the Concept of Political Power American Philosophical Society, Proceedings 107:232–262.
Polsby, Nelson W. 1963 Community Power and Political Theory. Yale Studies in Political Science, Vol. 7. New Haven: Yale Univ. Press.
Riker, William H. 1959 A Test of the Adequacy of the Power Index. Behavioral Science 4:120–131.
Riker, William H. 1964 Some Ambiguities in the Notion of Power. American Political Science Review 58:341–349.
Rogow, Arnold A.; and Lasswell, Harold D. 1963 Power, Corruption and Rectitude. Englewood Cliffs, N.J.: Prentice-Hall.
Rossi, Peter H. 1960 Power and Community Structure. Midwest Journal of Political Science 4:390—401.
Shapley, L. S.; and Shubik, Martin 1954 A Method for Evaluating the Distribution of Power in a Committee System. American Political Science Review 48:787–792.
Simon, Herbert A. (1947–1956) 1957 Models of Man: Social and Rational; Mathematical Essays on Rational Human Behavior in a Social Setting. New York: Wiley.
Simon, Herbert A. 1953 Notes on the Observation and Measurement of Political Power. Journal of Politics 15:500–516.
Weber, Max (1922) 1957 The Theory of Social and Economic Organization. Edited by Talcott Parsons. Glencoe, III.: Free Press. → First published as Part 1 of Wirtschaft und Gesellschaft.
family relationshipsbrian jory
marital relationshipscarrie l. yodanis
Family power is important to those who want to understand how families function as a unit to make decisions about how to manage money, about where to live, about occupational and educational choices, about parenting practices, about where to go on a vacation, and so on. Family scientists define power in terms of who is able to influence others to get their way in the family, and who is able to block others from getting their way. In most cases, family power is a property of the family system, not of a single individual, because it is almost impossible for one individual to have their way all of the time. Although the rules that govern power in a particular family may evolve as children are born, grow up, and move out, as the marital relationship changes or dissolves, or as the circumstances of the family changes, power is deemed to be fairly predictable within these stages. This predictability can be a comfort to those family members who are happy with the power arrangements or a matter of disdain, perhaps even a matter of personal health and safety, for those who find themselves dominated by others.
Ronald Cromwell and David Olson (1975) classified family power into three areas: power bases, power processes, and power outcomes.
J. R. P. French and Bertran Raven (1959) took a microsystemic view of family power. That is, they examined power strictly from inside the family and suggested that there are six bases of family power. Legitimate power is sanctioned by the belief system within the family, such as the belief that the husband should be the head of the household, that parents should have control over raising small children, or that adolescents should have control over what they wear. In the United States, an uncle who tries to impose his will on his nieces and nephews might be viewed as a meddler who is trying to exercise illegitimate power. In other cultures uncles are accorded legitimate power over nieces and nephews and might be respected for this kind of guidance.
Informational power has its foundation in specific knowledge that is not available or is unknown to others in the family and in one's ability to verbally present the pertinent information in a persuasive way. For example, if the man in the household is the only one who knows his income, or if he is viewed as knowledgeable about money, then he is likely to make decisions about how money is spent in the family. Alternatively, if a wife can assemble pertinent information about the benefits of purchasing a new car, she may be able to convince her reluctant husband.
Referential power is based on affection, mutual attraction, friendship, and likeability within the family. Positive feelings can be a powerful force in making alliances with others, if others want to make those they care about happy and, conversely, not to disappoint them. A parent's desire to please a favored child, a husband's desire to please his wife, a child's desire to please a grandparent are examples of referential power.
Coercive power involves the use of physical or psychological force in imposing one's way on others in the family, assuming that others are resistant or opposed. Parental discipline, threats, aggression, conflict, and competition are inherent in the use of coercive power because getting one's way is usually realized at the expense of others getting theirs. An example of coercive power: a parent forces a child to attend a school or college he or she does not wish to attend by threatening to withdraw the child's support.
Expert power is based on education, training, or experience that is relevant to the issue at hand. For example, if the woman of the household is a licensed real estate agent, she may have the most influence on where the family lives. If a child has studied the attractions of Florida, he or she may use the expert power accumulated to wield influence on decisions about a Florida vacation. Expert power can also be derived from the specific knowledge and experience of one individual in dealing with a specific issue. For example, if the husband was raised in Mexico, he is likely to be considered the expert about what relatives to visit in Mexico and where to stay on a visit there. Although he may not be considered an expert on Mexico outside the family, within the family he is.
Reward power is the ability to influence others by providing physical and psychological benefits to those who comply with one's wishes. With small children, parents often influence behavior with candy or sweets. With older children and adolescents, the price of power might be more expensive—a new outfit or bicycle. Adults in families often strike bargains, exchange pleasing behaviors, and "sweet talk" others to get their way.
The power bases articulated by French and Raven are often unclear in actual families. For example, if one family member has used coercion in the past, others may have learned that it is best to give in and keep their opinions to themselves. Although it may not be apparent to outsiders, those inside the family may feel coerced even though they do not signal their resistance in visible ways.
Robert Blood and Donald Wolfe (1960) took a macrosystemic view when they presented their resource theory of family power. That is, they looked for associations between power inside the family and power outside the family, and argued that power was apportioned between husbands and wives based on the relative resources that each contributed to the family. Blood and Wolfe specifically focused on the resources of income, occupational prestige, and educational attainment and, based on interviews with hundreds of white, middle-class wives in Detroit, Michigan, demonstrated that the greater the men's resources in these three areas, the greater the men's perceived power within the family.
The resource theory of family power was influential because the idea suggested that men do not become heads of households by divine right or natural biological processes, but because they have more and easier access to educational, financial, and occupational resources in society. The idea suggested that opening up women's access to resources outside the family could result in a more evenly balanced distribution of power within the family.
There has been considerable research support for resource theory in the United States and in Third World countries. Philip Blumstein and Pepper Schwartz (1983) conducted a study in the United States and found that when men made substantially more income than their wives, they were more likely to exert greater power in financial decision-making when compared with husbands that made about the same income as their wives. A study conducted in Mexico by R. S. Oropesa (1997) found that wives with higher education were equal to their husbands in family power, felt more satisfaction with their influence in the family, and were less likely to be a victim of domestic violence. A study of 113 nonindustrialized nations conducted by Gary Lee and Larry Petersen (1983) found that the more wives contributed to food production, the more power they exerted in marriage.
There has also been substantial criticism of resource theory. It has been pointed out that income, occupation, and education are only three among many resources that influence family power. Edna Foa and U. G. Foa (1980) suggested that in addition to tangible resources such as money, education, and occupation, intangible resources such as intelligence, physical attractiveness, likeability, love, and comfort impact family power. Actually, any trait or behavior that is valued by others in the family can be a resource that is exchanged for influence and power. For example, in immigrant families it has been observed that the ability to speak the host language can increase one's power if other family members depend on that ability to translate and interpret messages (Alvarez 1995). Among the Fulani tribes of West Africa, who primarily practice the religion of Islam, family members, especially women, can increase their power in the family by practicing traditional Fulani customs of conjuring the spirits of dead ancestors and others who have passed on to the other world ( Johnson 2000).
Most family scientists take a macrosystemic view, first articulated by Constantina Safilios-Rothschild (1967), that the bases of family power are a reflection of culturally defined gender ideologies and gender-segregated resources in the wider society in which a family is embedded. In practically all societies, this means that males have more power in families because of patriarchal beliefs about male authority. For example, a 1996 Gallup Poll conducted in twenty-two countries found that women are almost universally perceived as more emotional, talkative, and patient than men, whereas men are perceived as more aggressive, ambitious, and courageous than women. Even though there may be little scientific justification for these perceptions, they exert a strong influence in favor of male dominance in families that might be diminished through women's resources, but not completely muted.
An examination of power processes reveals that getting one's way in the dynamic interaction of families entails an ongoing set of complex and subtle maneuvers involving communication, commitment, bargaining and negotiation, coalition formation, conflict and conflict resolution, and parenting styles. Moreover, an examination of power processes reveals that in virtually all cultures, variables like the number of children and where the family lives make family power processes more complex.
Willard Waller (1938) is credited with first articulating the idea that family power is sometimes affected by commitment: The principle of least interest states that in disputes involving power, the individual who is least interested in continuing the relationship usually has more power than the one who is more interested in continuing the relationship. In dating relationships, the threat to break up can level the playing field of relative power. In some cases, an individual who feels "one-down" can make the threat and gain an equal footing if the other wants to stay together. In worse cases, an individual who is already "one-up" can threaten to break up and gain an even stronger hand in future disputes. In marriage, the principle of least interest can involve threatening to divorce, or in parent-child relationships, by parents threatening to send a child to foster care, to boarding school, or to live with a relative. Children and adolescents sometimes invoke the principle of least interest by threatening to run away or, in cases where parents are divorced, by threatening to go live with the noncustodial parent. In order to increase power, however, threats to leave must be feared by those one is threatening. Otherwise, they may say, "Go ahead and leave." If this happens, the tables of power could be turned against the one making the threat.
The principle of least interest applies mostly in societies where marriage is a free choice rather than arranged, and where it is possible for men and women to dissolve marriage through divorce. In many cultures, divorce is restricted by social and religious tyranny that makes personal selectivity in one's partner irrelevant to the establishment or continuation of marriage (Swidler 1990). For example, in societies that are ruled by intolerant legalists or religionists, the courts might allow a husband to obtain a divorce simply because he has lost emotional interest in his wife or because she has done something of which he disapproves. In the same society, a wife might not be granted a divorce even if she has legitimate reasons, such as her husband's abuse, desertion, criminal behavior, or, in polygamous societies, if he were to take another wife without the permission of the wife or wives he already has. In these societies, family power processes are so structured along gender and generational lines that selectivity has little to do with the establishment and maintenance of marital and family relationships. Alternatively, selectivity may be applied unfairly, allowing men to make choices that are not accorded to women or children. As previously discussed, family power processes reflect power bases in society: Without power in society, it is difficult to get power in the family.
Anthropologist Janice Stockard (2002) analyzed the power processes of married couples in four cultures and found that parent-child alliances had a strong impact on family power. For example, girls of the !Kung San tribe of South Africa were traditionally married around age 10, usually to men who were much older. Marriages were arranged by the girls' parents, who expected the bridegroom to live with them for a few years following the marriage and help out by hunting for food. Although one might think that these young girls would be powerless in relation to their older husbands, the fact that brides and grooms lived with the girls' parents permitted the girls to maintain strong alliances with their parents. These strong alliances tended to equalize power between husbands and wives, to the degree that !Kung San girls had strong veto powers over the marriage arrangement, which they often exercised.
In sharp contrast, girls in traditional Chinese societies were required to abandon alliances with their parents, grandparents, and siblings following marriage. On her wedding day, a traditional Chinese girl would be transported to live with her husband's family, where her mother-in-law would hold authority over her. The restriction of Chinese girls, who seemingly were not permitted to make many personal choices about their lives, was rationalized with the understanding that they would be compensated in later years by gaining rule over their own daughters-in-law. Because young girls were temporary participants in their families as they were growing up, it was difficult for Chinese girls to form deep, lasting alliances with their parents, grandparents, and siblings.
In Western culture, Theodore Caplow (1968) hypothesized that powerful male heads of households might find themselves at a power disadvantage in families with older children and adolescents because mothers and children might form coalitions to neutralize and override the fathers' power. A study conducted by Brian Jory and his colleagues (1996) found substantial support for coalition theory by observing the power processes of middle class families in the midwestern United States in moderately stressful problem-solving situations. In these families mothers were five times more likely to form power alliances with adolescent sons and daughters than with their husbands. These fathers, who were mostly in high power occupations, were at a clear disadvantage in family power negotiations. The importance of gender in family power processes was evident in another way: The study found that adolescent boys were more active in communication and bargaining than adolescent girls, and mothers offered more supportive communication to adolescent sons than daughters.
Diana Baumrind (1971) studied the balance between power and support in the childrearing behavior of parents in the United States and identified three parenting styles. The authoritarian style of parenting emphasizes obedience, giving orders, and discipline. Parents who exercise this style relate to their children with little emotional warmth because they view the child as a subordinate whose primary need is discipline. Children raised by authoritarian parents often feel rejected because their ideas are not welcomed, and these children may have trouble in tasks that demand autonomy, creativity, and reflection.
The permissive parenting style de-emphasizes parental control of children in favor of absolute acceptance and approval of the child. Permissive parents encourage children to make decisions on their own and to exercise creativity and independence in whatever they do. In the absence of parental guidance and limits, children raised by permissive parents may feel neglected and may struggle with tasks where focus, self-control, and perseverance are required.
The authoritative style of parenting combines a balance of parental control and parental warmth and support. Authoritative parents set limits on acceptable behavior in children, yet do so in an affectionate environment that encourages autonomy, values expression of opinions, and encourages participation in family decision-making. In reviewing a number of studies, Lawrence Steinberg and his colleagues (1991) demonstrated that children raised by authoritative parents—whatever their race, social class, or family type—develop better moral reasoning, do better academically, have less anxiety and depression, feel that their families are happier, are more self-confident, and are less likely to become delinquent.
A study by Brian Jory and his colleagues (1997) discovered that, in families with adolescents, power is not limited strictly to parental behavior, but is a property that affects the family system as a whole in terms of communication, bargaining, how affect is expressed, and how solutions to problems are generated. The study found four types of family locus of control. In families with individualistic locus of control, power resided in individuals who looked out for themselves. In these families, communication was egocentric and calculated, affect could turn negative or aggressive, and individuals sought solutions that benefited themselves at the expense of others.
In families with authoritarian locus of control, power was located in the parents, particularly the father whose role as head of household was pronounced. Communication in these families was directed one-way from fathers to mothers and mothers to children, affect was stilted, and bargaining was nonexistent as solutions to problems took the form of parental pronouncements, exclusively by fathers.
In families with external locus of control, nobody in the family was viewed as having power, and control seemed to be located in circumstances, fate, or the control of others. Communication in these families was chaotic, affect was directed towards others outside the family, and solutions to problems were sought from authority figures and others who were viewed as having control.
In families with collaborative locus of control, communication was systematically elicited from each family member, ideas were valued, affection was warm, supportive, and caring, and great effort was dedicated to find solutions to problems that had the least negative impact on individuals and would benefit the group as a whole.
As each of these studies shows, power processes in families involve a large number of complex cultural and family-related variables, many of which are yet to be discovered by family scientists. Making matters more complex, those variables that have been discovered are subtle and difficult to measure. For example, keeping secrets—an intentional withholding of information—is a form of communication that affects power in families. Withholding information takes away the power of others to make reasonable decisions because they lack pertinent information. How does a scientist measure secrets? This reveals the scientific challenge of studying power in families, but also the importance.
Power is an underlying dimension of every family relationship and virtually every family activity, and its importance lies in the fact that having a sense of control over one's life is necessary for the health and happiness of humans, including children, adults, and the elderly. In the studies already discussed, it is evident that power should be fairly apportioned to every family member, from the youngest infant to the most elderly person. If every member of a family has a sense of personal control, balanced with family control, the family can be a source of power and strength through its guidance, support, and care. When someone in the family abuses power, however, the damage to trust, loyalty, and freedom can have long-term negative effects for everyone in the family.
In the last quarter of the twentieth century, Western society began paying attention to the dark side of family power. A new set of concepts developed that are common in the language of the twenty-first century: child abuse and neglect, child sexual abuse, elder abuse, marital rape, date rape, psychological abuse, wife abuse, and domestic violence. In a volume entitled, The Public Nature of Private Violence (1994), editors Martha Fineman and Roxanne Mykitiuk assembled a number of articles by scholars who suggest that our discovery of family abuse has created a new conception of the nature of family life for the twenty-first century. The old conception that families are guided by a higher moral law or a natural order of compassion has been replaced by a more realistic conception that, for many, the family is a place of anguish, worry, pain, and trauma. These scholars argue that the abuse of family power is not simply a private matter, but is a public matter that needs to be part of the public agenda to be addressed by policy-makers, police officers, judges, social workers, clergy, teachers, physicians, and counselors.
The abuse of power in families is not strictly a Western idea. Judy DeLoache and Alma Gottlieb (2000) compiled imagined childcare guides for seven societies. The variation in parental practices—the do's and don'ts of raising children—from society to society is astounding. For example, it may be difficult to understand why Turkish mothers keep their babies restrictively swaddled for several months following birth (to show that the baby is covered with care). It may seem odd, if enticing, that Beng mothers paint pretty designs on the faces of their infants every day (to protect the baby against sickness). Should parents clean and bathe children? That, according to the childcare guides, depends on what society the child is born into. Although parenting practices vary around the world, one principle underlies all cultural variations. In no extant culture are mothers or fathers legitimately granted absolute power to mistreat their children. There is a general ethical principle that is universal: the abuse of power in families is not socially condoned.
Building on the idea that family power should be subjected to the same ethical principles as other forms of social power, Brian Jory and his colleagues have conducted a number of studies exploring how the abuse of power in families is rooted in ethical beliefs about power ( Jory, Anderson, and Greer 1997; Jory and Anderson 1999; Jory and Anderson 2000). In studies of abusive men (and their women partners) conducted in the United States, Jory concluded that interventions that change the ethical beliefs of those who abuse power in their families can result in a positive transformation of their values and behavior. The abuse of power in families is a challenge for those who shape all societies to transcend the bounds of culture and custom and work towards balancing the scales of intimate justice in all societies by fostering ethical beliefs about equality, freedom, respect, fairness, and caring in families, and by showing compassion for those who are suffering the anguish of victimization, whatever their cultural heritage.
See also:Child Abuse: Physical Abuse and Neglect; Child Abuse: Psychological Maltreatment; Child Abuse: Sexual Abuse; Communication: Family Relationships; Decision-Making; Family and Relational Rules; Gender; Parenting Styles; Power: Marital Relationships; Problem Solving; Spouse Abuse: Prevalence; Spouse Abuse: Theoretical Explanations
alvarez, l. (1995). "pint-size interpreters of world for parents." new york times, october 1: a16.
baumrind, d. (1971). "current patterns of parental authority." developmental psychology monographs 4:1–102.
blumstein, p., and schwartz, p. (1983). american couples:money, work, and sex. new york: morrow.
blood, r., and wolfe, d. (1960). husbands and wives: thedynamics of married living. new york: free press.
caplow, t. (1968). two against one: coalitions in triads. englewood cliffs, nj: prentice hall.
cromwell, r., and olson, d., eds. (1975). power in families. newbury park, ca: sage.
deloache, j., and gottlieb, a., eds. (2000). a world of babies: imagined childcare guides for seven societies. cambridge, uk: cambridge university press.
fineman, m., and mykitiuk, r. (1994). the public nature of private violence: the discovery of domestic abuse. new york: routledge.
french, j., and raven, b. (1959). "the basis of power." in studies in social power, ed. d. cartwright. ann arbor: university of michigan press.
foa, e., and foa, u. (1980). "resource theory: interpersonal behavior as exchange." in social exchange: advances in theory and research, ed. k. gergen, m. greenberg, and r. willis. new york: plenum press.
gallup poll. (1996). gender and society: status andstereotypes. princeton, nj: gallup organization.
johnson, m. (2000). "the view from the wuro: a guide to child rearing for fulani parents." in a world of babies: imagined childcare guides for seven societies, ed. j. deloache and a. gottlieb. cambridge, uk: cambridge university press.
jory, b., and anderson, d. (1999). "intimate justice ii: fostering mutuality, reciprocity, and accommodation in therapy for psychological abuse." journal of marital and family therapy 25:349–363.
jory, b., and anderson, d. (2000). "intimate justice iii: healing the anguish of abuse and embracing the anguish of accountability." journal of marital and family therapy 26:329–340.
jory, b.; anderson, d.; and greer, c. (1997). "intimate justice: confronting issues of accountability, respect, and freedom in therapy for abuse and violence." journal of marital and family therapy 23:399–420.
jory, b.; rainbolt, e.; xia, y.; karns, j.; freeborn, a.; and greer, c. (1996). "communication patterns and alliances between parents and adolescents during a structured problem solving task." journal of adolescence 19:339–346.
jory, b.; xia, y.; freeborn, a.; and greer, c. (1997). "locus of control and problem solving interaction in families with adolescents." journal of adolescence 20:489–504.
lee, g., and petersen, l. (1983). "conjugal power and spousal resources in patriarchal cultures." journal of comparative family studies 14:23–28.
oropesa, r. s. (1997). "development and marital power in mexico." social forces 75:1291–1317.
safilios-rothschild, c. (1967). "a comparison of power structure in marital satisfaction in urban greek and french families." journal of marriage and the family 29:345–352.
steinberg, l.; mounts, n.; lamborn, s.; and dornbusch, s. (1991). "authoritative parenting and adolescent adjustment across varied ecological niches." journal of research on adolescence 1:19–36.
stockard, j. (2002). marriage in culture: practice andmeaning across diverse societies. new york: harcourt.
swidler, a. (1990). marriage among the religions of theworld. lewiston, ny: mellen press.
waller, w. (1938, revised 1951). the family: a dynamicinterpretation. new york: dryden.
Power is a fundamental aspect of all human relationships, including family and marital relationships. Since 1960, there has been a continuing dialogue among social scientists seeking to define, measure, explain, and understand the consequences of power differentials in marriage relationships.
Definitions and Measurement
Power in marriage has been defined and measured in various ways. The first and most common definition of power is the ability of one person to get another to do what she or he wants even in the face of resistance. Based on this definition, Robert Blood and Donald Wolfe (1960) developed the Decision Power Index. To measure power, respondents are asked to report whether wives, husbands, or both have the final-say on a number of decisions within the marriage, including selecting a car, home or apartment, vacation, doctor, husband's job, and whether or not the wife should work. Who has power in the relationship is measured based on who has the final-say.
This index has remained at the core of the dialogue on marital power. Being a short and easily administered instrument, this index continues to be included in surveys worldwide, although occasionally with adaptations. It has also been critiqued, developed, and improved. Nearly once a decade since its development, a review is written that raises methodological questions and concerns about the final-say decision-making measures (Mizan 1994). Many of these problems have been tested empirically.
One problem cited is the discrepancy between the answers given by husbands and wives. However, data from such countries as the United States, India, and Panama have tested this issue and found that wife and husband answers tend to be parallel (Allen and Straus 1984; Danes, Oswald, and De Esnaola 1998).
Another set of problems involves the types of decisions and assumptions about decisions that are included in measures. Merlin Brinkerhoff and Eugen Lupri (1978) and Vanaja Dhruvarajan (1992) present data from Canada to show that decisions are of varying importance and frequency and are made according to gender roles. Women tend to have final-say in some areas, particularly those decisions relating to care work—children, food, entertaining friends, and calling the doctor—which tend to be defined by both men and women as not very important. Thus, it is argued that a measure that gives each decision equal weight results in a flawed power score.
Furthermore, measures of power tend to be outcomes or consequences of power. The outcome serves as a proxy measure of power. For example, the individual with the most power in a relationship may or may not be most likely to make the decisions. Similarly, the division of household labor is an outcome of power differentials but is used in some studies as a measure of power.
It has been argued that it is important to define and measure power as a dynamic process, examining such issues as influence strategies and attempts (Aida and Fablo 1991; Zvonkovic, Schmiege, and Hall 1994). Using a multidimensional definition of power, Aafke Komter (1989) defines power as "the ability to affect consciously or unconsciously the emotions, attitudes, cognitions, or behaviors of someone else" (p. 192) and distinguishes between manifest power, latent power, and invisible power. As the usual conceptualization of marital power, manifest power refers to decision-making and associated conflict and influence strategies. Latent power refers to a lack of decision-making, conflict, or influence strategies as a result of one partner anticipating and deferring to the position of the other. This can result from the less powerful partner believing that they are unable to have influence or fearing negative reprisal. Finally, invisible power refers to an unconscious process in which social and psychological systems of inequality result in one partner being unable to even conceive of the possibility of having input in decision making, engaging in conflict, or using power strategies. In her study of Dutch couples, Komter found that although the couples share equally in decision making, there were uncovered hidden power mechanisms and strategies that result in women wanting more change in the relationship but being less successful in gaining it. As a result, an ideology of husbands' power over their wives was confirmed and justified.
Resources. Like their measure of power, Blood and Wolfe's (1960) resource theory has had a prominent role in explanations of marital power. According to their theory, power in marriage results from the contribution of resources—particularly education, income, and occupational status—to the relationship. The spouse who contributes the most will have the greater decisionmaking power. As with the measurement of power, theoretical and empirical work on explanations of marital power has often emanated from a critique and extension of Blood and Wolfe's theory.
Considerable work has been done within the realm of resources. Some researchers have added additional dimensions to the concept of resources. In a study of marital power in Israel, Liat Kulik (1999) found that not only material resources but also health and energy resources, psychological resources (problem-solving and social skills), and social resources (access to social networks) are directly or indirectly related to power in marriage. Particular attention has been paid to the impact of extended families and kin support resources on power in marriage. In Turkey and Mexico, a wife's ties to her family of origin can translate into power in marriage (Bolak 1995; Oropesa 1997). On the other hand, living in a joint residence with the family of the husband, as in India, has been found to be associated with higher levels of power for husbands (Conklin 1988). In a study of over a hundred nonindustrial societies, in general women have somewhat more power in nuclear than extended families. Nevertheless, in societies where extended families are the norm, women have substantially more power when residence practices are matrilocal and descent is matrilineal rather than patrilocal and patrilineal (Warner, Lee, and Lee 1986).
Greater attention has been paid to the meaning tied to the resources, and not just the amount of resources contributed. A spouse may contribute resources but if this contribution is not recognized as significant and valuable within the couple, the contribution is not likely to result in greater power (Bolak 1995; Blaisure and Allen 1995). From this perspective, unpaid family work can also be a valued contribution to the relationship and not working for pay may reflect women's power rather than lack of power (Pyke 1994).
Resources can also be thought of as alternatives to the relationship. Adding to Blood and Wolfe's theory, David Heer (1963) developed an exchange theory of marital power, arguing that the individual who has the greatest access to alternative resources outside of the marriage relationship will have the most power. In a related argument, Willard Waller's (1951) principle of least interest theory proposes that the spouse who is least interested in maintaining the relationship will have the greater power. Karen Pyke (1994) found that women's reluctance to marry after divorce is associated with their greater power in remarriage. Based on a study in the United Kingdom, Pat O'Conner (1991) argues for the need to also consider a principle of high mutual interest. Results show that women are powerful in relationships where dependence is high and balanced for both women and men. Using data from Israel, Liat Kulik (1999) developed the concept of anticipated dependence, defined as the extent to which one spouse expects to need the other at later points in life, and found greater anticipated dependence to be related to reduced power in the current relationship.
Culture. One of the most significant developments of Blood and Wolfe's resource theory came from Hyman Rodman (1967, 1972). Trying to understand cross-cultural inconsistency in the relationship between resources and marital power in Germany, United States, France, Denmark, Belgium, Greece, and Yugoslavia, Rodman developed the theory of resources in cultural context. This theory explains that the distribution of marital power results not only from an unequal contribution of resources, but also from the larger cultural context within which the marital relationship exists. Cultural gender norms affect the impact that resource contribution has on the distribution of power. In particular, he predicted that in patriarchal and egalitarian societies, the dominant norms would outweigh the influence of resources on marital power. So regardless of wife and husband's contribution, marriages will be male-dominated in patriarchal societies and equal in egalitarian societies. He predicted that the contribution of resources has the most significant impact on the balance of marital power in transitional egalitarianism, societies that are moving from patriarchal toward egalitarian norms, and among the upper classes in modified patriarchal societies, societies in which egalitarian values are new and common only among the upper strata.
The theory of resources in cultural context has been applied and tested in countries throughout the world. Some studies have focused on Scandinavian countries, which are considered egalitarian societies. A comparison of Danish and U.S. couples revealed that although couples in both countries often report equality in decision making, Danish marriages were even more likely to be described as equal. Within this egalitarian society, the resource contribution of spouses still has an impact on the balance of power (Kandel and Lesser 1972). Studies in Sweden and Norway show that even within these relatively egalitarian societies, male power may not be blatant but is widespread in marital relations (Calasanti and Bailey 1991; Thagaard 1997).
Latin American countries are often assumed to be characterized by machismo cultures and families. However, data from Chile, Mexico, and Panama show that although husbands may have somewhat greater power in marriage, many marriages tend toward egalitarianism and the contribution of resources is related to power (Alvarez 1979; Cromwell, Corrales, and Torsiello 1973; Oropesa 1997; Danes, Oswald, and De Esnaola 1998). This tendency toward egalitarianism has been discussed in terms of wider societal change, resulting from social movements, including the women's movement, and economic development.
The connection between culture and resources in the balance of marital power has been examined in a wide range of societies, including Turkey, India, Israel, Romania, Russia, and China. As Turkey and other Muslim countries combine modernity and traditionalism and become modified patriarchies, women are better able to negotiate power, with their contribution of resources having an impact on their success (Fox 1973; Bolak 1995). In Eastern Europe under communist regimes, both spouses worked outside of the home and tended to be egalitarian in decision making, but women remained primarily responsible for household work (Lapidus 1978; Elliot and Moskoff 1983). Today in some Eastern European countries, such as Russia, the woman's role as head of household and breadwinner does not necessarily result in greater power but is a result of necessity and lack of alternatives (Kiblitskaya 2000). In China, where the influence of Western ideology has increased the acceptance of egalitarian relationships, education and occupation are related to the distribution of marital power (Tang 1999).
Structure. The patriarchal and egalitarian differences between societies described above can be considered as not only cultural but also structural differences. Gender inequality is not merely found in norms and ideologies but characterizes the structure and practices of a society's political, legal, religious, educational, and economic institutions. Discrimination and male domination of these institutions result in women's lower access to resources, including income, occupational status, and education; condone and reinforce patriarchal ideology; and thereby contribute to the maintenance of gender inequality in marriage. As Dair Gillespie (1971) argued, marital power can be described as a caste/class system because husbands as a class have power over wives as a class as a result of male-dominated societal structures rather than any specific resources they contribute to the marriage.
Interaction. Researchers have looked beyond resources and culture to examine how marital power is part of the unconscious construction of gender categories and identity during interaction. Veronica Jaris Tichenor (1999) studied couples in which wives had higher income and occupational status than their husbands and found that wives did not tend to have greater marital power. Rather, women and men in these couples act to ignore or minimize the status and income differences and construct the husband in a powerful position, through such acts as upholding male veto power and redefining the provider role to fit the activities of the husband. Examining conversations of women and men, Caroline Dryden (1999) found that wives act to construct their marriage as equal and blamed themselves for aspects of the relationship that were not equal. Husbands, on the other hand, act to construct the marriage as unequal and also blamed their wives for the existing inequality. The outcome of these constructions is the reinforcement of the genders as unequal, with the man in the position of power. Tove Thagaard (1997) also found that power differences displayed during Norwegian couples' interactions served to confirm male and female identity, including gender inequality.
Another aspect of the growing interactionist perspective includes an examination of how couples publicly present marital power. The presentation may or may not correspond with the actual power dynamics in the relationship. In Turkey, couples who are equal in power may present their relationship as male-dominated to outsiders as a way to appear to be in compliance with dominant cultural norms (Bolak 1995). Another study found that couples who define themselves as feminist act to publicly present themselves as equal. Strategies used include maintaining different last names and putting the wife's name first on tax returns or car registration (Blaisure and Allen 1995). In a similar study, couples who claim to be egalitarian were found not to be in practice. However, they used language to create a "myth of equality" (Knudson-Martin and Mahony 1998).
Multivariate models. Attempts have been made to integrate many of these theories into comprehensive multivariate models. Rae Blumberg and Marion Tolbert Coleman's (1989) is one of the most complete. Starting with women's resource contribution, the model factors in societal and individual characteristics that can enhance or diminish women's ultimate economic power and then outlines the path through which women's net economic power translates into multiple dimensions of marital power.
Inequality in marriage is related to a number of consequences, many of which are interrelated and, in turn, reinforce gender inequality. This section focuses on two related categories of consequences: health and happiness.
Health. A lack of power in marriage is a threat to women's mental and physical health. Wives who have power in marriage report lower rates of depression (Mirowsky 1985) and less stress (Kaufman 1988). In India and Kenya, women's power in marriage is related to lower fertility rates and greater use of methods of family planning (Sud 1991; Gwako 1997).
Violence against women is also related to unequal power between women and men. Studies show that women are less likely to be victims of physical and verbal abuse in egalitarian relationships (Coleman and Straus 1986; Tang 1999). However, a loss of men's power relative to women's may also result in a greater likelihood for violence. According to Craig Allen and Murray Straus's (1980) ultimate resource theory, when husbands lack economic or interpersonal skill resources to maintain a dominant position in marriage, they may fall back on physical size and strength—resources that, on average, husbands tend to have more of than their wives.
Happiness. In addition to individual mental health and happiness, the distribution of power in marriage is also related to marital quality and satisfaction. Some older studies found that wives' satisfaction was highest in egalitarian marriages (Alvarez 1979), whereas others found that quality and satisfaction were highest in male-dominated marriages (Buric and Zecevic 1967). These findings may well reflect pressure to correspond with previously dominant gender ideologies. More recent studies in countries such as Norway and China are quite consistent in finding that marital satisfaction is highest in egalitarian marriages (Thagaard 1997; Tang 1999; Pimental 2000). Not only is equality directly related to increased satisfaction but also indirectly related through the development of closer emotional ties and perceptions of a spouse as fair and sympathetic. In addition, marital satisfaction is related not only to the distribution of power between spouses but also the types of power strategies and attempts used. Although the use of any influence strategy has been found to be related to lower marital satisfaction, indirect and emotional strategies, including negative affect and withdrawal, seem to have particularly negative effects (Aida and Falbo 1991; Zvonkovic, Schmiedge, and Hall 1994).
Marital power has been the topic of a dialogue among social scientists from diverse perspectives, cultures, and methodological approaches who build on the past and add to the future. The dialogue has been international, using cultural and societal differences to test and advance theory. The result of this dialogue is rich theoretical and empirical work on marital power. Still, participation in the dialogue waxes and wanes over time. Although since 1990 new insights and approaches have been brought to the debate, as Jetse Sprey (1999) outlines, there is still a need to revisit and rethink longstanding approaches and assumptions to studying and understanding power in marital relationships.
See also:Conflict: Couple Relationships; Decision Making; Depression: Adults; Division of Labor; Dual-Earner Families; Equity; Family Life Education; Family Roles; Gender; Marital Quality; Marital Typologies; Power: Family Relationships; Problem Solving; Rape; Resource Management; Retirement; Rich/Wealthy Families; Self-Esteem; Spouse Abuse: Prevalence; Spouse Abuse: Theoretical Explanations; Stress; Therapy: Couple Relationships; Work and Family
aida, y., and falbo, t. (1991). "relationships between marital satisfaction, resources, and power strategies." sex roles 24:43–56.
allen, c. a., and straus, m. a. (1984). "'final say' measures of marital power: theoretical critique and empirical findings from five studies in the united states and india." journal of comparative family studies 15:329–344.
alvarez, m. d. l. (1979). "family power structure in chile: a survey of couples with children in primary schools." international journal of sociology of the family 9:123–131.
blaisure, k. r., and allen, k. r. (1995). "feminists and the ideology and practice of marital equality." journal of marriage and the family 57:5–19.
blood, r. o., and wolfe, d. m. (1960). husbands andwives: the dynamics of married living. glencoe, il: free press.
blumberg, r. l., and coleman, m. t. (1989). "a theoretical look at the gender balance of power in the american couple." journal of family issues 10:225–250.
bolak, h. c. (1995). "towards a conceptualization of marital power dynamics: women breadwinners and working-class households in turkey." in women in modern turkish society, ed. s. tekeli. london: zen books.
brinkerhoff, m., and lupri, e. (1978). "theoretical and methodological issues in the use of decision-making as an indicator of conjugal power: some canadian observations." canadian journal of sociology 3:1–20.
buric, o., and zecevic, a. (1967). "family authority, marital satisfaction, and the social network in yugoslavia." journal of marriage and the family 29:325–336.
calasanti, t. m., and baily, c. a. (1991). "gender inequality and the division of household labor in the united states and sweden: a socialist-feminist approach." social problems 38:34–53.
coleman, d. h., and straus, m. a (1986). "marital power, conflict, and violence in a nationally representative sample of american couples." violence and victims 1:141–157.
conklin, g. h. (1988). "the influence of economic development on patterns of conjugal power and extended family residence in india." journal of comparative family studies 19:187–205.
cromwell, r. e.; corrales, r.; and torsiello, p. m. (1973). "normative patterns of marital decision making power and influence in mexico and united states: a partial test of resource and ideology theory." journal of comparative family studies 4:177–196.
danes, s. m.; oswald, r. f.; and de esnaola, s. a. (1998). "perceptions of couple decision making in panama." journal of comparative family studies 29:570–583.
dhruvarajan, v. (1992). "conjugal power among first generation hindu asian indians in a canadian city." international journal of sociology of the family 22:1–34.
dryden, c. (1999). being married, doing gender: a critical analysis of gender relationships in marriage. london: routledge.
elliot, j. e., and moskoff, w. (1983). "decision-making power in romanian families." journal of comparative family studies 14:39–50.
fox, g. l. (1973). "another look at the comparative resource model: assessing the balance of power in turkish marriages." journal of marriage and the family 35:718–730.
gillespie, d. l. (1971). "who has the power? the marital struggle." journal of marriage and the family 33:445–458.
gwako, e. l. m. (1997). "conjugal power in rural kenya families: its influence on women's decisions about family size and family planning practices." sex roles 36:127–147.
heer, d. m. (1963). "the measurement and bases of family power: an overview." marriage and family living 25:133–139.
kandel, d. b., and lesser, g. s. (1972). "marital decision- making in american and danish urban families: a research note." journal of marriage and the family 34:134–138.
kaufman, g. m. (1988). "relationship between marital power and symptoms of stress among husbands and wives." wisconsin sociologist 25:35–44.
kiblitskaya, m. (2000). "russia's female breadwinners: the changing subjective experience." in gender, state and society in soviet and post-soviet russia, ed. s. ashwin. london: routledge.
knudson-martin, c., and mahony, a. r. (1998), "language and processes in the construction of equality in new marriages." family relations 47:81–91.
komter, a. (1989). "hidden power in marriage." gender and society 3:187–216.
kulik, l. (1999). "marital power relations, resources and gender role ideology: a multivariate model for assessing effect." journal of comparative family studies 30:189–206.
lapidus, g. w. (1978). women in soviet society: equality,development, and social change. berkeley: university of california press.
mirowsky, j. (1985). "depression and marital power: an equity model." american journal of sociology 91:557–592.
mizan, a. n. (1994). "family power studies: some major methodological issues." international journal of sociology of the family 24:85–91.
o'conner, p. (1991). "women's experience of power within marriage: an inexplicable phenomenon?" sociological review 39:823–42.
oropesa, r. s. (1997). "development and marital power in mexico." social forces 75:1291–1317.
pimentel, e. e. (2000). "just how do i love thee?: marital relations in urban china." journal of marriage and the family 62:32–47.
pyke, k. d. (1994). "women's employment as a gift or burden? marital power across marriage, divorce, and remarriage." gender & society 8:73–91.
rodman, h. (1967). "marital power in france, greece, yugoslavia, and the united states: a cross-national discussion." journal of marriage and the family 29:320–324.
rodman, h. (1972). "marital power and the theory of resources in cultural context." journal of comparative family studies 3:50–69.
sprey, j. (1999). "family dynamics: an essay on conflict and power." in handbook of marriage and the family, 2nd ed., ed. m. b. sussman, s. k. steinmetz, and g. w. peterson. new york: plenum.
sud, s. l. (1991). marital power structure, fertility, andfamily planning in india. new delhi, india: radiant.
tang, c. s-k. (1999). "marital power and aggression in a community sample of hong-kong chinese families." journal of interpersonal violence 14:586–602.
thagaard, t. (1997). "gender, power, and love: a study of interaction between spouses." acta sociologica 40:357–376.
tichenor, v. j. (1999). "status and income as gendered resources: the case of marital power." journal of marriage and the family 61:638–650.
waller, w. (1951). the family: a dynamic interpretation. new york: dryden.
warner, r. l; lee, g. r.; and lee, j. (1986). "social organization, spousal resources, and marital power: a cross-cultural study." journal of marriage and the family 48:121–128.
zvonkovic, a. m.; schmiege, c. j.; and hall, l. d. (1994). "influence strategies used when couples make work-family decisions and their importance for marital satisfaction." family relations 43:182–188.
carrie l. yodanis
In its most general sense, power refers to the capacity to have an effect. A powerful hurricane comes ashore uprooting trees and destroying houses; the power of music stirs our emotions; the power of love makes gentle the surly man or woman. We speak of electric or wind power, of powerful machines or human bodies. Many different phenomena with a very wide range of effects are said to be or to have power. The definition of the term therefore has led to a good deal of controversy.
Even more controversial has been the moral assessment of power: is power a good? Should one seek to acquire and to use power or is it the duty of morally conscientious persons to avoid power and the ability it implies of coercing others?
Much of the time, power is discussed only as it manifests itself in economic, social, and political life.
The power to get what one wants.
Most generally, "The power of man is his present means to obtain some future apparent good" (Hobbes, p. 56). The ability to get what one wants because one believes it to be good is what the English philosopher Thomas Hobbes (1588–1679) calls "power." Bertrand Russell (1872–1970), another English philosopher, defended a very similar conception of power. This is power to get what one wants.
The power to dominate.
Much more common, however, is a more limited conception according to which power consists of getting what one wants from another person, particularly if it is ceded unwillingly. Thus Max Weber (1864–1920) defines power as "the probability that one actor in 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" (Weber, p. 152). Power here concerns the relationship between individual actors or groups of individuals where some are able to do as they please in spite of the resistance by others. A similar definition was given by Robert A. Dahl (b. 1915): "A has power over B to the extent that he can get B to do something that B would not otherwise do" (1957, pp. 202–203). These formulations describe a form of power excessively familiar to children who are bullied by classmates or teachers; to their parents who feel oppressed by their employers; and to citizens who must unwillingly obey the dictates of the government, sometimes in the guise of the neighborhood policeman. This is the power to dominate.
The power to manipulate.
However familiar, the preceding definition has encountered many criticisms and has had to defend itself against other concepts of power. Some theorists have pointed out that we do indeed exercise power when we overcome the resistance of others. But power can also shape situations and persons so that they will never think to resist. Instead, they submit willingly to a yoke, which later, perhaps, they may find onerous. The power to manipulate so that resistance does not arise is more impressive than the power to overcome resistance. Telling a child about to be inoculated that "it won't hurt" sidesteps the child's resistance and struggle. Official lies that persuade citizens that they have been attacked by a foreign power, when in fact it is their government that is the aggressor, have persuaded many who would not ordinarily consider becoming soldiers to flock to the flag of their country. The number of casualties in a war and the extent of destruction and suffering imposed would produce strong criticism, but if no one knows about the damage done by the armies, opposition will not develop.
Hegemonic power, or ideology.
Manipulation occurs between individuals and groups. But preemptive shaping of the agendas for discussion and action may be much more insidious and less deliberately deceptive. There are organizational structures that, although not intended for that purpose, function to lessen dissent and opposition. Electoral systems often function in that way, especially in a situation of severe economic and social stratification. The dissatisfied have a ready mechanism at hand for expressing their discontent, but unless they belong to the ruling elites, their efforts will consume a great deal of energy but bear little fruit. Such tamping down of discontent may be the unintended consequences of certain political arrangements. There exists no ready term for power in that sense. Terms sometimes used in this context such as hegemonic power or ideology are often understood in an excessively intellectual sense. At issue here is the power, inherent in existing institutions, of privileged strata of a society to avoid popular criticisms and demands.
These various forms of power, whether intentional or the side effect of certain political institutions, still deal with antagonists who are latent, if not actual, resisters. Were they not misinformed, or sidetracked by the complexities of electoral politics, they would certainly express their discontent quite vocally. These are all different variants of power that some one or some group has over another, individual or group. They are still variants of the power to dominate.
Influence and authority.
But not all power is power to dominate. Influence is the most obvious example of a different kind of power. Frequently someone is only too glad to follow the wishes of another whom they love and who loves them. Their love is strong; there is no shadow of a threat to withdraw love if the other does not yield. But each is glad to do what is asked of them because it pleases them to please their lover. In similar ways parents and children often have influence over one another. Each person has power in relation to the other—a power freely given—which is power to influence the other. Influence does not dominate and neither does authority: one allows another to influence one's thought and action because the other is an expert, or a revered spiritual teacher. Their power to give us orders or directions is, in some way, legitimate. (The word authority also has a different meaning. "The authorities"—not only legitimate authorities—are only too frequently able and willing to dominate by making us do what we do not want to do.)
Neither influence nor authority is the power to dominate—power over. Instead it is power with that arises out of the cooperation of several individuals or perhaps an entire people. We experience this in times of crisis, in a family or in a nation, when everyone ignores their own immediate needs and desires and pitches in to solve a shared problem or protect the entire group against imminent danger. Hannah Arendt (1906–1975) pointed out that a people gains strength from cooperation, from a shared sense of a common goal or mission: "Power corresponds to the human ability not just to act, but to act in concert" (p. 44). The familiar slogan "United we stand; divided we fall" represents the popular understanding of this power that arises out of unity and cooperation. This power enhances the life of a nation and stiffens its resolve and ability to overcome difficult challenges. Only when it loses this power with does a government have to resort to violence. A nation subdued by violence may be tyrannized, but in order to govern it, one requires the support of the people. Only then can one have power.
Such power with exists not only in larger groups but is important in the relations between very few people. Couples find themselves better able to confront problems and to tackle difficult tasks because their shared values and understanding of the world increase their power. Power grows out of the mutual respect and recognition of two persons. Small groups, like families, grow better able to face demanding tasks when their mutual respect and understanding makes each stronger and wiser.
In the intellectual space of the West, power with, if recognized at all, is assigned a subsidiary role. But anthropologists report that among many of the indigenous inhabitants of the American continent, power was traditionally of this sort—power with. The chiefs of many tribes in North and South America did not have any power to dominate or to coerce. They provided leadership because they were respected. Many tribes had two chiefs; one led by authority and the voluntary compliance of the tribe's members in peace time. The other, the war-time chief, had the more familiar kind of power—the power to coerce.
Comparisons between Western philosophy and Chinese thought are fraught with difficulties. Nonetheless, some experts recognize in Confucian classics distinctions very similar to the ones we draw between power to dominate and power with. Confucianism sought an ordered society through the education of a morally superior elite to lead the nation. For the purpose of forming such an elite, classical Confucian texts reject laws as means for developing this elite and, instead, recommend a civil, humane practice within communities resting on cooperation and mutuality. Laws threaten and coerce. The power to dominate wielded by laws is not acceptable to the Confucian educator who, instead, puts his faith in the power with of communal and reciprocal action.
The different forms of power discussed so far, whether variants of power to dominate or power with, are all intimately connected to specific and identifiable persons. One knows who the bully is in a group. One can identify the government functionaries who put out misleading propaganda. One can identify the persons who, through their solidarity and unity, give strength to their group, or even their nation. The power of ideology is more diffuse; it is less easy to identify the prominent contributors to dominant beliefs that reduce the possibility of resistance. But the ruling ideology is promulgated by and in the interest of the ruling strata of the society, and one knows who the members of these ruling strata are. One may not know all of them, but one knows that there are specific individuals and associations of individuals that yield the powers that manipulate the majority of the population. All the variants of power discussed so far belong to specific human beings or groups who wield this power.
But the power individuals wield presupposes complex social arrangements. Thomas Wartenberg points out that a judge's power to punish a convicted criminal depends on an intricate network of other institutions and the different roles these institutions contain. Judges can condemn the criminal only while seated on the bench in the courtroom; they must be within a prescribed context in order to exercise their official role. They cannot do it in their home while lounging in the bathtub. The court is not only a place; it is an institution, complete with court officers, stenographers, lawyers, prosecutors, and so forth. The entire power of the judge presupposes the law and the many different institutions that give rise to the law and legitimate it. The judge's power does not belong to the judge unless there exists an extended set of institutions that function because everyone is doing his or her job and exercising the power that comes with that job.
It is important that the different participants in the drama of crime and punishment understand the significance of the places, the institutions, and the actions of functionaries. Everyone must understand what it means when the judge bangs her gavel and says "Twenty years." Various persons understand that these two words instruct them to lead the criminal out of the courtroom and remand him to the prison authorities who, in their turn, must follow established bureaucratic procedures. In these two words the judge manifests her power and everyone must understand that. For power to function there must be a wealth of shared understanding of the significance of words and actions.
For this reason, Niklas Luhmann says that power is a "code"—a quasi language. This conception was first introduced by the American sociologist Talcott Parsons, who criticized the conception of power as belonging to individual persons, urging us, instead, to understand power as a "medium" analogous to money as the medium of the economy. An economic system, Parsons pointed out, consists of a large number of offers and demands for goods and services. These offers, and their acceptance or rejection, must be communicated, and money is the medium of this communication. ("Money" here does not refer to the dollar bills in your wallet but to the institution of the market and the banking institutions and other financial contrivances for communicating offers and demands within the economy.) Political institutions correspondingly are the medium in which in politics we pass on messages. Power is a form of "social knowledge."
The importance of personal power.
Power as medium or as code or as social knowledge is, indeed, social; it no longer belongs to specific individuals but inheres in the society. It is interesting to notice that, from the beginning of Western philosophy, the power inherent in societies to shape each of us has been familiar. In Plato's dialogue Crito, Socrates, on the eve of his execution, receives his friend Crito in prison. Crito urges Socrates to escape with him; everything is prepared for him to flee. But Socrates refuses. The laws of Athens, he says, have made him who he is. He owes them obedience even if they are, in his case, applied unjustly. Our culture shapes and constrains us and makes us who we are. While not a modern discovery, this social power has attracted a great deal of attention in modern times. Jean-Jacques Rousseau (1712–1778) complained about the conformity modern societies demand of their members, forcing them to pretend to be persons they are not rather than be themselves. Friedrich Nietzsche (1844–1900) echoes that understanding of the power of social structures. Karl Marx (1818–1883) is very clear that the injuries capitalism inflicts on workers are not to be laid at the door of individual capitalists because they, too, are under the compulsion of the laws of the capitalist marketplace. The power to exploit inheres in the capitalist system; it is not the fault of the capitalists.
Discussions of the impersonal power of social institutions, the "regime without a master" in the words of Joan Cocks (p. 187), have become even more frequent in the twentieth century. John Dewey complained about the standardization of human beings in modern society. Members of the Frankfurt School—Theodor Adorno, Max Horkheimer, Herbert Marcuse, and others—wrote about the power of mass entertainment to regiment members of modern societies and to make it impossible for critical voices to be raised, let alone be heard. The concept of a power not owned by anyone, implicit in these critiques, was then made explicit by the theorists of power, Talcott Parsons, Niklas Luhmann, and Barry Barnes.
Popular culture took up the worry of these theorists; the subservience of the individuals to their culture that seemed to Socrates a perfectly obvious and uncontroversial fact about each of us, has in our world become a matter of unease. In many cases this unease takes the form of a polemic against what the author labels "conformism" in such works as The Organization Man by William H. Whyte (1956), or Arthur Miller's Death of a Salesman (1949), and in David Riesman's The Lonely Crowd (1950) as well as the novel by Richard Yates, Revolutionary Road (1961). After World War II, talk about authenticity became very common—being oneself in the face of great social pressures to conform became a frequently heard moral demand. Films like 2001: A Space Odyssey, and more recently The Matrix, had enormous appeal because, among other things, they expressed the anxiety felt by many that impersonal forces, unknown to us, hold us in their thrall.
It was left for the French philosopher Michel Foucault (1926–1984) to attempt an explanation of why we moderns are much more concerned about being subject to impersonal social power. The eighteenth century, he believes, brought many changes: the beginning of capitalism with its much more rational, carefully planned methods of production, the acceleration of urbanization, which brings many persons together in a small space, and the beginnings of the study of humans and their societies. These changes gave rise to many new techniques of surveillance, of controlling and keeping track of persons. The new social sciences developed a typology of persons; each of us exemplifies many different types, being men or women, parents or not, educated or not, belonging to a particular income bracket or not, practicing a particular religion or not, having children in the approved way ("in wedlock"), and so forth. The typologies serve to "normalize," to set standards by which we are measured from birth as being within normal range of variation or needing to be supervised, cured, corrected, counseled, tutored, or otherwise brought back to the norm.
All of these are forms of domination, Foucault argues. But the origins of the acts of domination are obscure. We are dominated by impersonal systems, much more often than by specific persons. "Power is no longer substantially identified with an individual … it becomes a machinery that no one owns" (Foucault, p. 156).
The Value of Power
There exists a range of different ideas in the different world traditions about the value of pursuing, having, and using power. Among the Old Testament Hebrews and in Islam, where temporal government and religion are very closely associated, the use and pursuit of power are accepted without question. Both the Old Testament Jews and the followers of Mohammed were quite content to use the temporal power in their possession for the purpose of extending the realm of the one true God. At the same time, the close association of governance and the deity imposed moral limitations on the uses of power. They were permitted only in the service of extending God's realm.
Both Judaism and Islam, in their classical formations, were religions of this world. While Islam speaks of an afterlife, the work of the religious person was in this world to extend and fortify the rule of the divine. Some Eastern religions, by contrast, such as Hinduism and Buddhism, saw everyday life as full of suffering and a realm of mere appearance. The goal of the religious individual was to overcome suffering by detaching him-or herself from ordinary desires, from emotions, from attachment to the things of this world.
Nevertheless, dominant strands of Buddhism and Hinduism were willing to use secular and military power to extend the sway of their religions. But at the same time, Buddhist and Hindu thought made room for a radical rejection of the pursuit and use of power. Doing no harm to any human or animal is an important principle of Buddhism. The thought of Mahatma Gandhi illustrates one version of Hinduism that rejects all uses of power, other than the power of persuasion. It resolutely refuses to coerce anyone or to use power to dominate. In this context, the use of violence is definitely an inferior alternative to the use of nonviolence.
The Christian tradition is equally conflicted. Jesus's saying that we should "render unto Caesar the things that are Caesar's" (Matthew 22:21, Mark 12:17, and Luke 20:25) has encouraged Christian monarchs and popes to use violence and power to dominate in the Crusades, in the Inquisition, and in the conquest and devastation of the Americas. But there is another Christian tradition of nonviolence exemplified by the Christian injunction that we turn the other cheek to those who strike us (Matthew 5:39 and Luke 6:29). Martin Luther King Jr. (1929–1968) elaborated this reading of Christianity in his doctrine and practice of nonviolence, which substituted persuasion for coercion.
See also Authority ; Autonomy ; Liberty .
Arendt, Hannah. On Violence. New York: Harcourt Brace and World, 1970.
Barnes, Barry. The Nature of Power. Urbana: University of Illinois Press, 1988.
Clastres, Pierre. Society against the State: The Leader as Servant and the Humane Uses of Power among the Indians of the Americas. Translated by Robert Hurley, in collaboration with Abe Stein. New York: Urizen, 1977.
Cocks, Joan. The Oppositional Imagination—Feminism, Critique, and Political Theory. London: Routledge, 1989.
Dahl, Robert A. "On the Concept of Power." Behavioral Science 2 (1957): 201–215.
De Bary, Wm. Theodore. Asian Values and Human Rights: A Confucian Communitarian Perspective. Cambridge, Mass.: Harvard University Press, 1998.
Foucault, Michel. Power/Knowledge: Selected Interviews and Other Writings, 1972–1977. Edited and translated by Colin Gordon. New York: Pantheon, 1980.
Hobbes, Thomas. Leviathan. Edited by Michael Oakeshott. Oxford: Blackwell, n.d.
Luhmann, Niklas. Trust and Power. Translated by Howard Davis, John Raffran, and Kathryn Rooney. Chichester, U.K., and New York: John Wiley, 1979.
Parsons, Talcott. Sociological Theory and Modern Society. New York: Free Press, 1967.
Russell, Bertrand. Power: A New Social Analysis. New York: W.W. Norton, 1938.
Schmitt, Richard. Beyond Separateness: The Social Nature of Human Beings—Their Autonomy, Knowledge, and Power. Boulder, Colo.: Westview, 1995.
Wartenberg, Thomas E. The Forms of Power: From Domination to Transformation. Philadelphia: Temple University Press, 1990.
Weber, Max. The Theory of Social and Economic Organization. Translated by A. R. Henderson and Talcott Parsons. Edited with an introduction by Talcott Parsons. New York: Oxford University Press, 1947.
Power is a central concept in the social and political sciences. It is also commonplace in everyday discussions: We often refer to a political party getting into power, or to the power of governments or individuals to perform a particular action or achieve a certain result, or to someone having power over another, or to a country being a super-power. Power would appear to be self-evident. However, power is an extremely elusive concept, and there are numerous disagreements over its definition, foundation, function, and operation. Power remains, as Steven Lukes says, an “essentially contested” concept (1974, p. 26).
Power is usually associated with the bringing about of consequences. However, what these consequences are, whether or not they are intended, how they are actually brought about, who brings them about and in whose interests—are all a matter of unresolved debate across a number of different disciplines. It is not possible here to give an exhaustive list of these controversies, or to touch on all the many questions regarding the nature and exercise of power. There are, however, several major areas of contention that should be mentioned:
- Should power be seen in terms of the actions or capacities of individual agents, or should it be seen as deriving from broader social structures?
- Is power a resource or capacity that can lie dormant, or does it only exist when it is exercised?
- Does power refer to the ability to achieve certain desired outcomes, or is it a relationship between agents where one exercises power over another?
- Does power necessarily involve domination, coercion, or constraint, or can it be based on consent?
- Is power exercised only where the consequences of a certain action are intended, or do unintended or unforeseen consequences also count as evidence of the exercise of power?
The remainder of the entry will explore a number of key theories and debates about power, which refer to several of the questions outlined above. It will also trace a certain logical development in the theorization of power from pluralist-behavioralism to structuralism to poststructuralism.
The idea that power has three dimensions or “faces” comes from Lukes, who argues that the formula for power—A has power over B to the extent that he can get B to do something that he would not otherwise do —can be seen as operating in three distinguishable, yet interrelated, ways. The first face of power is usually associated with Robert Dahl who, along with Polsby (1963) and Wolfinger (1971), tried to show that power in the U.S. political system was distributed pluralistically. In doing so, they were opposing the “ruling elite” theorists such as Mills (1956), who believed that power was concentrated in the hands of a dominant group in society. In his study of local politics in the New Haven area of Connecticut—which he took as a microcosm of the broader distribution of power in American society—Dahl argued that there was no empirical evidence to support the idea of a ruling elite, and that, in fact, different groups were influential over different areas of policymaking (1961). Dahl’s analysis contained the implicit idea that there are plural centers of power in a democratic society. More importantly, its central focus was on decision-making behavior in cases where there is an observable conflict of interests. For Dahl, power is the ability to affect another’s decision-making: A exercises power over B when he can get B to make a decision that he would not have otherwise made.
However, this idea of power as decision-making was criticized for being too one-dimensional. Peter Bachrach and Morton Baratz argued that power also has a hidden or covert dimension—a “second face” (1962). Power involves not only decision-making, but also what they call nondecision-making. This refers to the ability of dominant elites to “set the political agenda” in such a way that certain issues are prevented from being aired, thus precluding the very possibility of a decision being made about them. In situations of conflict, there is often a “mobilization of bias” against certain interests. The mass media would be an example of this: Whether consciously or unconsciously, it reinforces dominant values and practices, thereby delegitimizing or marginalizing opposing viewpoints and preventing potential issues from becoming actual issues. In this paradigm, power operates not necessarily by directly influencing B’s decision-making, but by preventing B from raising concerns that might be detrimental to A’s preferences.
Lukes, however, contends that even this understanding of power was limited, because, like the pluralist-behavioralist view, it assumed that power is only exercised in situations of observable conflict between different interests. But what if it were the case that power functioned in such a way as to prevent conflict from arising in the first place (1974)? Here Lukes points to an even more insidious dimension of power—its “third face”—where power operates not simply by A getting B to do what he does not want to do, but by shaping B’s thoughts and desires in such a way that B does what A wants him to do as if it were a free and autonomous act. In other words, power may operate as a form of subtle thought-control or manipulation, and may cause someone to act, not according to his own interests, but in the interests of those who are exercising this power. What is being suggested here is that what we think we want and is in our best interests may not actually be so—our preferences may be shaped by external influences. Here we might think of the advertising industry, which sells us products that we do not necessarily need or even want, by manipulating our desires. This distinction Lukes draws between subjective interests and real interests is problematic, as discussed below, but his analysis of power is nonetheless interesting for the way it moves away from the domain of individual decision-making behavior, toward some notion of an overall structuring of the ideas and values that shape individual behavior.
The structuralist argument is that power derives not so much from individual or even collective agents, but from their place in a broader social structure. In other words, it is the structural position of agents that allows them to exercise power over others. Marxists like Nicos Poulantzas argue, for instance, that in a relationship of class conflict, the economic and political power of the bourgeoisie—and its capacity to realize its interests—derives from its structural location within the capitalist system. In his debate with Ralph Miliband—who suggested that the class bias of the state could be explained by the privileged background and class allegiances of those who manned the state apparatus (1969)—Poulantzas argued that Miliband’s view places too much emphasis on individual behavior, and neglects the effects of structural relations in the capitalist system (1973). In Poulantzas’s view, power derives from the ensemble of structures that make up capitalist society, which shapes relations between classes and allows one class to dominate others.
Michel Foucault further radicalizes the concept of power by taking it beyond questions of both individual behavior and structure. For Foucault, power is a non-derivative concept that cannot be reduced to the preferences of individual agents, economic classes, or even the structural requirements of the capitalist system. Rather, power must be studied in its own right. Here he introduces a number of important methodological innovations. Firstly, the focus must be on the “how” of power—because power only exists when it is being exercised. There is no mysterious substance called power that can lie dormant without being exercised: Indeed, Foucault goes so far as to suggest that power “as such” does not exist (1994, p. 336). Secondly, power is relational, rather than an individual or structural capacity: That is to say, power is a mutual relation between agents—both individual and collective. Power is a way of acting on the actions of others. This implies, thirdly, a certain freedom of action on the part of both agents in a power relationship. The power relationship presupposes that agents are able to act differently, that they have a range of actions open to them, and that power involves constraining or influencing these actions. Foucault argues, for instance, that slavery is not a power relationship because there is no possibility of the slave acting differently. In this sense, power is not a zero-sum game as many suggest: Rather, it involves a dynamic interplay between agents. Fourthly, then, while power is not the same as coercion, neither is it a matter of consent, as Arendt (1969) or Parsons (1969) would claim. While power constrains, there is always the possibility of there being resistance to it, even in situations of domination, where the normally free and reciprocal flow of power becomes congealed.
However, the question of resistance in Foucault’s theory of power is also ambiguous and problematic (see Newman 2001; 2004). This is because Foucault sees power as being not only repressive and prohibitive, but also productive: Power produces and incites (1978). Unlike Lukes, who sees power as distorting the subject’s “real interests,” Foucault believes that this notion of “real interests” is an essentialist illusion manufactured by power itself. Power intersects with discourses and “regimes of truth” to construct the very identity of the subject. While this avoids the dubious notion of “real interests,” it would seem, at the same time, to undermine the idea of a firm ontological and normative foundation for resistance to power: The subject who resists power is at the same time constructed by it. Foucault’s theory of power raises as many questions as it answers, and it should not be thought that he has revealed some elusive “fourth dimension” of power beyond which we cannot proceed any further. However, by focusing on the “how” of power, and by seeing power in terms of relationships rather than as a substance or capacity, Foucault has considerably advanced our understanding of the concept.
SEE ALSO Arendt, Hannah; Community Power Studies; Dictatorship; Foucault, Michel; Galbraith, John Kenneth; Mills, C. Wright; Political Science; Politics; Postmodernism; Poulantzas, Nicos; Power Elite; Power, Political; Repression; Resistance; Structuralism; Wealth; Weber, Max; Zero-sum Game
Arendt, Hannah. 1969. On Violence. New York: Harcourt Brace.
Bachrach, Peter, and Morton S. Baratz. 1962. Two Faces of Power. American Political Science Review 56: 947–952.
Foucault, Michel. 1978. Histoire de la sexualité [The History of Sexuality]. Vol. I: An Introduction. Trans. Robert Hurley. New York: Pantheon.
Foucault, Michel. 1994. The Subject and Power. In Power: Essential Works of Foucault, 1954–1984, Vol. 3, ed. James Faubion, 326–348. Trans. Robert Hurley. London: Penguin.
Lukes, Steven. 1974. Power: A Radical View. London: Macmillan. Miliband, Ralph. 1969. The State in Capitalist Society. New York: Basic Books.
Newman, Saul. 2001. From Bakunin to Lacan: Anti-Authoritarianism and the Dislocation of Power. Lanham, MD: Lexington.
Newman, Saul. 2004. New Reflections on the Theory of Power: A Lacanian Perspective. Contemporary Political Theory 3 (2): 148–167.
Parsons, Talcott. 1969. Politics and Social Structure. New York: Free Press.
Polsby, Nelson W. 1963. Community Power and Political Theory. New Haven, CT: Yale University Press.
Poulantzas, Nicos. 1973. Pouvoir politique et classes sociales del’état capitaliste [Political Power and Social Classes]. Trans. Timothy O’Hagan. London: New Left Books, 1973.
Wolfinger, Raymond E. 1971. Nondecisions and the Study of Local Politics. American Political Science Review 65: 1063–1080.
The meanings of power, influence, control, and domination are uncertain, shifting, and overlapping. Although two of these words may be interchangeable in one context, in another context one of the words may refer to a genus and another to a species, or one may refer to a cause and another to an effect. To substitute power for influence would not matter much in the sentence "The United States has very great influence in South American politics," but to interchange them would radically change the meaning of the sentence "Colonel House's power derived not from any constitutional authority but from his influence over President Wilson."
Shifts like this account for much of the intractability of problems associated with power. For instance, power is often said to be a relation (Lasswell 1950, Friedrich 1950, Partridge 1963), yet we talk about the distribution of power, about the power of speech, about seeking power as a means to future enjoyment (Hobbes 1946), or about power as "the production of intended effects" (Russell 1938). If power is a relation, between what kind of terms or things does it hold? Does power over men require a minimum of acquiescence, consent, or cooperation (Hume; Friedrich 1956–1957), or can it be analogous to a physical force acting on an otherwise inert object? Is to exercise power always to succeed in what one intends (Russell 1938, Lasswell 1950), or can a man exercise power in ignorance of what he is doing (Dahl 1961, Partridge 1963, Oppenheim 1961), like a ruling elite that neither knows nor cares about the effects of its actions on other classes?
Instead of seeking a single analysis of power, it is more helpful to think of diverse uses of power and of associated words like influence as instances of different members of a family of concepts that do not all share any one particular characteristic but have various relations and resemblances by which they are recognizably kin. One might construct a power paradigm combining as many of these family features as possible. Thus, "A, by his power over B, successfully achieved an intended result r ; he did so by making B do b, which B would not have done but for A 's wishing him to do so; moreover, although B was reluctant, A had a way of overcoming this."
There are five main features of this paradigm: (1) an intention manifest in the exercise of power; (2) the successful achievement of this intention; (3) a relationship between at least two people; (4) the intentional initiation by one of actions by the other; and (5) a conflict of interest or wishes engendering a resistance that the initiator overcomes. Not every feature would be present, of course, in every instance in which we properly speak of power; but we can examine how different instances are related to the paradigm and to one another, and thus throw some light on a few of the questions listed above.
Power and Conflict
Some instances of power do not involve overcoming resistance to an initiative. A charismatic leader's power over his followers consists in being able merely by suggestion to move them to do willingly what he wants, even though their interests might have led them to act differently. The family of power concepts might be arranged along a conflict scale (Partridge 1963): At the end at which conflict is least would lie instances of influence, while at the other end would lie instances of domination, and in between, instances of authority. In the extreme case, exercising influence would not involve overcoming resistance, for to manipulate a man's actions by shaping what he considers to be his interests is not to impose action upon him in the face of his interests. Yet this would still be an instance of power satisfying the first four features of the paradigm.
The limiting case at the end of the scale at which conflict is least would be rational persuasion, for to offer a man good reasons for doing something is not to exercise power over him, although it may influence his decision. One possible difference between influence and power, then, seems to be that power generally implies a difference of standing between the two parties: The one stimulates, the other reacts. Rational persuasion, on the contrary, to the extent that it criticizes and invites criticism, presupposes at least the possibility of a dialogue between equals. To the extent that persuasion is really rational, the influence is not so much that of the persuader as of his arguments; the same arguments from anyone else would do as well. (By contrast, a threat of violence is more effective coming from a strong man than from a weak man.) Of course, if A rationally persuades B to help him, A may get power—not over B, however, but over C or D, or even simply the power to do something he could not otherwise do.
Power, Injury, and Interest
In the case of the man who punishes another for disobedience, conditions (1), (3), and (5) of the paradigm would be satisfied, but not (2) and (4), for the initiative has been refused. Instead, it suffices for an instance of power if the power-holder successfully and intentionally makes the subject suffer for refusing the initiative. And by yet a further extension of meaning, one can exercise power over someone by deliberately making him suffer, whether or not he has refused an initiative. Just as in the limiting case of rational persuasion one could speak of influence but hardly of power, so at the other end of the scale one can talk of power but not of influence, for influence is manifest in what a man is, does, or believes, not in what is simply made to happen to him by another man.
A stoic would probably resist the extension of the concept of power to cover the mere infliction of suffering. By not caring about physical pain or external conditions, he might say, one can remove oneself from the power of another man. So too Martin Luther believed that a true Christian is free because no outer things can touch him at any significant point. It would seem that what characterizes a power situation of this kind is not just the ability to make someone suffer, which after all a dentist possesses, but rather to do him harm—that is, to attack his interests. Thus, by revising the notion of a man's interests, and therefore the notion of harm, the stoic or the Christian can deny the reality of one man's power over another, since nothing that another man can do to me can affect my real interests; I am always free, if once I see what those interests are. This argument is a little odd, because the concept of power generally implies a restriction on choice; but according to the stoic or Christian view, one can always choose to make the restriction insignificant, and therefore one can choose whether to be in the power of another. In that case, there could not be a real restriction, and all power would be illusory. But then, what would power be like if it were real?
The stoic argument demonstrates, however, that whether one man has power over another depends not merely on what he can do to the other but also on the importance to be attached to his action and on whether the subject can reasonably be expected to disregard it. One would not say that X was in Y 's power if one thought that what Y could do to X was trivial—something that X could or should readily ignore.
Again, although threats of real harm are an exercise of power, bribes or promises of reward are not, unless some special feature of character or situation makes them irresistible—that is, unless no one so placed could reasonably be expected to resist them (although some in fact might). This is not to say, of course, that a man cannot exercise power by bribery. However, it need not be power over his hirelings but power over others through them; or it may be power only in the still more general sense of an ability to bring about an intended result. Thus, we speak of power in situations in which a man could either successfully determine another's actions or do him harm. An ability to do him some good is not in itself power over him, although the threat of withholding a good that he has come to count on may well be.
Problems of Power as a Relation
Power may not be a relation between people but between a person and a thing. There is a nonsocial kind of power that is simply an ability to produce an intended result, like a tenor's power to smash a tumbler with a high C. And even in a social context, the financier's power to destroy a government comes very close to this, for in this instance too power is manifest merely in the active achievement of an intended result. Although the financier no doubt works by initiating actions on the part of others, the relation between him and his object (the government) is that which exists between agent and patient. This case can be distinguished both from that in which power is exercised by punishing a subject for noncompliance and from that in which power is used to inflict deliberate injury. For in the present case the object of the exercise may be only to remove an obstacle. The manifestation of power does not consist in the government's being made to suffer, for it would be just as much a manifestation of power if the financier had chosen instead to prop it up or if the government welcomed its downfall as a blessed release from responsibility. Power is manifest simply in that what happens is the result of the financier's intentional action, just as the tenor's power is manifest in his being able to break a glass whenever he likes.
Power is of course relational in a logical sense in that it requires more than one term for a complete statement; and if more than one of the terms is a person, and the relation presupposes institutions, rules, and so forth, power will certainly be a social relation. But writers who stress that power is a relation usually mean that it is an initiative-response relationship of the kind that C. J. Friedrich had in mind when he wrote, "The power seeker must find human beings who value the things [he controls] sufficiently to obey his orders in return" (Constitutional Government and Democracy, p. 12).
Now, Friedrich's point is substantially true in those instances in which power implies a successful initiative and even perhaps in those instances in which power tends to injure its subject. To set about hurting someone, one must know how to get the right kind of response: There is no point in depriving nonsmokers of tobacco. It is not so clear, however, that the financier's power is of this type, for he does not secure a response from the government; he merely makes something happen to it. Although his agents respond to his initiatives, one must distinguish the power he has over them from the power he has over the government. These powers would be of the same kind only if he were able not just to destroy the government, but to use it as he wished. But it is presumably because he cannot do this that he uses his power to destroy it.
This analysis further elucidates the relation between power and consent. We have seen that at one extreme a man may exercise power over another by influencing his desires, or a man may do whatever he is told by another because he believes that he ought to do so, which is an instance of authority. Both cases imply some measure of consent or acquiescence, if not to the particular initiative, then to the right of the initiator to issue it. But in cases in which power depends on threats or on physical coercion, the subject's acquiescence amounts to no more than that he continues to value whatever is being used as a lever against him—an acquiescence that only the stoic, perhaps, would seriously regard as a matter of choice. However, political power cannot be entirely coercive. The few can rule the many because the many believe either that the few are entitled to do so or that they could harm them if they disobeyed. But they would not think that coercion were possible if they did not also believe that most of the people were prepared to obey without coercion. A political power situation, therefore, must almost always contain some elements of acquiescence as well as coercion—almost always because it is at least theoretically possible that a reign of terror might enslave a whole people simply by sowing such mistrust that its opponents could never know their own potential strength.
Power and Intention
Still further from the paradigm is the case in which one says quite generally that a person is powerful, or that he seeks power, without specifying the range of possible intended action or the persons subject to the power. Usually it would not be difficult to supply terms to complete either one or both of these blanks. Political theorists commonly insist that comparisons of power, without reference to its "domain" and "scope," are meaningless (Lasswell and Kaplan 1950, Oppenheim 1961). However, some have tried to generalize the concept by disregarding intentionality. R. A. Dahl defines power as "the difference in probability of an event, given certain actions by A, and the probability of the event given no such action by A " ("The Concept of Power," p. 214). At this level of abstraction, power is freed not only from intentionality but also from achievement and conflict; what remains is a relation between a stimulus and a reaction. Elsewhere (Modern Political Analysis, p. 40), Dahl defines influence as a relation among actors in which one induces others to act in some way in which they would not otherwise act. Dahl would want to purge, if he could, the hint of intentionality in the word induce. Like a field of force in mechanics, power is a potential for creating disturbance, like the potential of a stone cast in a pond for creating ripples. But this has some odd results. Instead of suffering a loss of power, the crashing financier who brings down thousands with him in his fall would be exercising a power that is perhaps greater than ever before. Admittedly, it is a mark of power if a man's actions cause disturbances, even if he is careless or even ignorant of them. Nevertheless, if powerful men cause incidental and unintended disturbances, they do so in the course of getting what they want. (C. Wright Mills's conception of a "power elite" seems to be of this kind.) One would not call someone powerful who, like a careless smoker constantly causing fires, was forever causing disturbances but never achieving anything he intended; nor is it clear that any useful methodological purpose in political science would be served by a definition of power that permitted the production of unintended effects alone to serve as a criterion.
To possess power or to be powerful is, then, to have a generalized potentiality for getting one's own way or for bringing about changes (at least some of which are intended) in other people's actions or conditions. Influence, it is true, is used in a more general sense. If a parent has the unintended influence of stiffening his child's determination to be as different from him as possible, this would not be described as an instance of power: It is more like "the influence of climate on national character." The use of the term influence suggests that there is a causal relationship between the behavior of the parent and that of the child (cf. P. H. Partridge, "Some Notes on the Concept of Power," p. 114). "A writer's influence on succeeding generations" stands somewhere between this case and that of influence by rational persuasion. For a writer may have influence only to the extent that other writers recognize his merits and choose to imitate him. Although such influence may not be intended, still it is not a cause, at least in the sense that climate is a possible cause of national character. In any case, none of these is an instance of an influence in the sense that House had influence with Wilson. "To use one's influence" usually implies actively and intentionally working through or on other people, and one who can do this recurrently "has influence." Of course, people who have power (that is, who can do many things they want and induce many other people to accept their initiative) are likely on that account to influence (that is, to have effects on) other aspects of society in ways that neither they nor their social inferiors necessarily understand. Other classes, envying and admiring them, may imitate their tastes and practices, and in this sense they may be influenced by them. But this influence is not a manifestation of power; it is only one of its effects.
classic theories of political power
Aiken, H. D., ed. Hume's Moral and Political Philosophy. New York: Hafner, 1948.
Hobbes, Thomas. Leviathan, edited by M. Oakeshott. Oxford: Blackwell, 1946.
Hume, David. "Of the First Principles of Government." In Essays Literary, Moral and Political (1741). London, 1963.
Hume, David. Treatise of Human Nature (1740), edited by L. A. Selby-Bigge. Oxford, 1951. Book III.
Plamenatz, John. Man and Society. 2 vols. London: Longmans, 1963. A critical discussion of classic theories.
Spinoza, Benedict de. The Political Works. Translated and edited by A. G. Wernham. Oxford: Clarendon Press, 1958.
Spinoza, Benedict de. Tractatus Politicus (1677).
Spinoza, Benedict de. Tractatus Theologico-Politicus. Hamburg, 1670.
general philosophical discussions
Acton, H. B. "Logique et casuistique du pouvoir." In Annales de philosophie politique: Tomes I et II: Le Pouvoir, Vol. II, 69–86. Paris, 1956–1957.
Emmet, Dorothy. "The Concept of Power." PAS 54 (1953–1954): 1–26.
Friedrich, C. J. "Le probleme de pouvoir dans la theorie constitutionnaliste." In Annales de Philosophie Politique: Tomes I et II: Le Pouvoir, Vol. I, 33–51. Paris, 1956–1957.
Jouvenel, Bertrand de. Le pouvoir. Geneva, 1945. Translated by J. F. Huntington as On Power; Its Nature and the History of Its Growth. Boston, 1962.
Jouvenel, Bertrand de. Pure Theory of Politics. Cambridge, U.K.: Cambridge University Press, 1963.
Partridge, P. H. "Politics and Power." Philosophy 38 (1963): 117–136.
Partridge, P. H. "Some Notes on the Concept of Power." Political Studies 2 (1963): 107–125.
Polin, R. "Sens et fondement du pouvoir chez John Locke." In Annales de Philosophie Politique: Tomes I et II: Le Pouvoir, Vol. I, 53–90. Paris, 1956–1957.
Russell, Bertrand. Power. London: Allen and Unwin, 1938.
sociology and methodology of political science
Bierstedt, Robert. "An Analysis of Social Power." American Sociological Review 15 (1950): 730–738.
Dahl, R. A. "The Concept of Power." In Introductory Readings in Political Behavior, edited by Sidney S. Ulmer. Chicago: Rand McNally, 1961.
Friedrich, Carl J. Constitutional Government and Democracy. Boston: Ginn, 1950.
Friedrich, Carl J. Man and His Government. New York: McGraw-Hill, 1963.
Merriam, Charles E. "Political Power." In A Study of Power, by H. D. Lasswell, C. E. Merriam, and T. V. Smith. Glencoe, IL: Free Press, 1950.
Oppenheim, Felix E. Dimensions of Freedom. New York: St. Martin's Press, 1961.
Walter, E. V. "Power and Violence." American Political Science Review 58 (1964): 350–360.
Weber, Max. From Max Weber; Essays in Sociology. Translated and edited by H. H. Gerth and C. Wright Mills. London: Routledge and Kegan Paul, 1948.
Arendt, Hannah. On Violence. New York: Harcourt Brace, 1970.
Connolly, William E. The Terms of Political Discourse, 2nd ed. Princeton, NJ: Princeton University Press, 1983.
Dahl, R. A. Modern Political Analysis, 5th ed. Englewood Cliffs, NJ: Prentice Hall, 1984.
Foucault, Michel. Power, edited by James D. Faubion; translated by Robert Hurley. New York: New Press, 2000.
Lukes, Steven. Power: A Radical View. London: Macmillan, 1974.
Morriss, Peter. Power: A Philosophical Analysis. Manchester, U.K.: Manchester University Press, 1987.
Parsons, Talcott. Politics and Social Structure. New York: Free Press, 1969.
Scott, John. Power. Cambridge, U.K.: Blackwell, 2001.
Wrong, Dennis H. Power: Its Forms, Bases, and Uses. New York: Harper and Row, 1979.
Stanley I. Benn (1967)
Bibliography updated by Philip Reed (2005)
Perhaps the best known of all the definitions is that of Max Weber in his essay ‘The Distribution of Power within the Political Community: Class, Status, Party’ (in Economy and Society, 1922). Weber regarded power as the fundamental concept in stratification, of which class, status, and party were three separate (sometimes related) dimensions. Broadly speaking, classes were the outcome of the distribution of economic power (in Weber's terms, market relationships); status was a kind of normatively defined social power; and parties were groups active in the political sphere in pursuit of various goals. Power was then defined by Weber in general terms as the probability of persons or groups carrying out their will even when opposed by others. Note that power is therefore a social relationship. Hence, for Weber, the differential distribution of power leads to a situation where life-chances are also differentially distributed; that is, the ability to obtain economic, social, and political resources is unequally distributed. In Weber's famous phrase, ‘classes, status groups and parties are all phenomena of the distribution of power in a society’. Weber took this view in a fairly explicit attempt to counter the crude Marxism of his day, which made too easy an elision between economic control and political rule. He wished to make clear his view that power need not depend on the possession of economic resources—hence the importance of the concept of status, and his various observations concerning this in his general sociology.
Weber made further observations concerning the nature of power in his political sociology. Few groups in society base their power purely on force or military might. Instead, ruling groups attempt to legitimate their power, and convert it into what he termed domination (or, as Talcott Parsons translates it, ‘authority’). According to Weber there are three bases of domination: traditional, rational-legal, and charismatic.
Is Weber's conception of power adequate—or are there more satisfactory ways in which the phenomenon can be conceived? Steven Lukes (Power: A Radical View, 1974) argues that power is an essentially contested concept; that is, one whose definition and application will always be a matter for dispute between sociologists. How we define power and how we operationalize it will be dependent on our theoretical position and value-orientation. Accepting that, however, is it still possible to improve on the Weberian conception of power?.
If we examine Weber's definition, it obviously has built into it a notion of conflict and intention. The notion of intention can be seen in the view of someone or some group ‘carrying out their will’. This implies a quality of conscious, rational, and calculated action in pursuit of a specific goal. Now, this may well characterize some power relationships, but does it characterize them all? Can power be exercised unwittingly? Should we perhaps see power as involving the achievement of one's preferences—whether by intention or not—rather than as the pursuit of one's will? The other problem we can see in Weber's definition is the assumption of conflict or antagonism which it incorporates. As various critics have noted, the definition suggests that A has power over B to the extent that he or she overcomes the resistance of B if it is offered, implying that—at least some of the time—the interests of B are being sacrificed to those of A. Weber was certainly interested mainly in power in situations of conflicting interests. Many sociologists since Weber have assumed that power involves—even provokes—subordinate resistance which must be overcome by superordinates. Does this mean power can never be exercised in a consensual context; that is, where subordinates accept it as being used legitimately? This is where we have to be more specific about the nature of the power being used. Where power is used over subordinates who attribute genuine legitimacy to superordinates we could talk of authority of persuasion. These are clearly very different from power which rests on force or manipulation. Yet we have to remember that all four of these terms refer to types of power relationships.
The idea of power used in an apparently consensual context also leads to further problems. For example, where legitimacy is attributed in a power relationship, does this legitimacy flow from subordinate to superordinate, implying authority (which is what Parsons and many political scientists would say); or is legitimacy imposed from above, by ‘swinging’ of social norms, implying manipulation (a view which has firm roots in Marxism, especially the Gramscian notion of ideological hegemony)? As Alvin Gouldner (The Coming Crisis of Western Sociology, 1970) noted, ‘power is, among other things [the] ability to enforce one's moral claims. The powerful can thus conventionalize their moral defaults.’ And, of course, this is part of what Weber meant by the term social status.
For all these reasons one should therefore remember David Lockwood's dictum: commenting on the problems of studying power, especially when it is recognized that power is a latent force, he observed that ‘power must not only refer to the capacity to realise one's ends in a conflict situation against the will of others; it must also include the capacity to prevent opposition arising in the first place. We often hear that the study of power should concentrate on the making and taking of important decisions. But in one sense power is most powerful if the actor can, by manipulation, prevent issues from coming to the point of decision at all’ (‘The Distribution of Power in Industrial Society—a Comment’, in J. Urry and and J. Wakeford ( eds.) , Power in Britain, 1973)
. So, power involves not only decision-making but also non-decision-making, not only the overt but the covert.
Finally, we should consider power resources. Power is a dispositional concept: it refers to the possibility of a certain action occurring rather than to its actual occurrence. So, power is a potential quality of a social relationship, and as such rests on actors' access to power resources. Quite obviously, in an advanced capitalist society, economic resources such as wealth and control over jobs are vital, but many other power resources exist: for example, organizational capacity, numerical support, competence, expert knowledge, control of information, occupation of certain social positions, control of the instruments of force, and reputation for power itself. The last of these is a unique power resource: it depends not on the actual possession of power but the mere belief by others that it is possessed. Equally, one does not have to own a power resource, but only to control it: senior civil servants and managers provide examples. Between all these potentials for power, and their manifestation, lies one's willingness (and efficiency) to use it. Potential power depends upon certain attributes. Manifest power, however, is revealed not by attributes but through social relationships; and part of the definition of a social relationship is its reciprocal nature. Consequently, the exercise of power involves feedback: A acts, B reacts, A reacts to B's reaction, and so on. Subordinates must have some effect on superordinates for there to be any relationship at all—a point noted long ago by Georg Simmel.
We can begin to see, then, how complex and difficult a concept power is to handle. Once we try to operationalize it we quickly appreciate Lukes's point concerning its essentially contested nature. This, and most of the other issues raised in this entry, are discussed is Dennis Wrong's Power (1979). See also BUREAUCRACY; COMMUNITY POWER; COMPLIANCE; FOUCAULT, MICHEL; GATEKEEPING; MICHELS, ROBERT; ORGANIZATION THEORY; ORGANIZATIONAL REACH; POLITICAL PARTIES; REFERENT POWER; STATE.
pow·er / ˈpou(-ə)r/ • n. 1. the ability to do something or act in a particular way, esp. as a faculty or quality: the power of speech | the power to raise the dead | (powers) his powers of concentration. 2. the capacity or ability to direct or influence the behavior of others or the course of events: the idea that men should have power over women she had me under her power. ∎ political or social authority or control, esp. that exercised by a government: the party had been in power for eight years| [as adj.] a power struggle. ∎ a right or authority that is given or delegated to a person or body: police do not have the power to stop and search emergency powers. ∎ the military strength of a state: the sea power of Venice. ∎ a state or country, esp. one viewed in terms of its international influence and military strength: a great colonial power. ∎ a person or organization that is strong or influential within a particular context: he was a power in the university. ∎ a supernatural being, deity, or force: the powers of darkness. ∎ (powers) (in traditional Christian angelology) the sixth highest order of the ninefold celestial hierarchy. ∎ [as adj.] inf. denoting something associated with people who hold authority and influence, esp. in the context of business or politics: a red power tie. ∎ used in the names of movements aiming to enhance the status of a specified group: gay power.3. physical strength and force exerted by something or someone: the power of the storm. ∎ capacity or performance of an engine or other device: he applied full power. ∎ the capacity of something to affect the emotions or intellect strongly: the lyrical power of his prose. ∎ [as adj.] denoting a sports player, team, or style of play that makes use of power rather than finesse: a power pitcher. ∎ the magnifying capacity of a lens.4. energy that is produced by mechanical, electrical, or other means and used to operate a device: generating power from waste| [as adj.] power cables. ∎ electrical energy supplied to an area, building, etc.: the power went off. ∎ [as adj.] driven by such energy: a power drill. ∎ [as adj.] power-assisted: power brakes. ∎ Physics the time-rate of doing work, measured in watts or less frequently horsepower.5. Math. the number of times a certain number is to be multiplied by itself: 2 to the power of 4 equals 16. • v. 1. [tr.] supply (a device) with mechanical or electrical energy: the car is powered by a fuel-injected 3.0-liter engine | [as adj. in comb.] (-powered) a nuclear-powered submarine. ∎ (power something up/down) switch a device on or off: the officer powered up the fighter's radar.2. [intr.] move or travel with great speed or force: they powered past the dock toward the mouth of the creek. ∎ [tr.] direct (something, esp. a ball) with great force: Nicholas powered a header into the net.PHRASES: do someone/something a power of good inf. be very beneficial to someone or something.in the power of under the control of: a church ministering in the power of the Holy Spirit.power behind the throne a person or organization that exerts authority or influence without having formal status.the powers that be the authorities.
One's capacity to act or to influence the behavior of others.
Power may be defined in both personal and interpersonal terms. In the first sense, it refers to one's physical, intellectual, or moral capacity to act. In the second, it denotes the ability to influence the behavior of others. Philosophers have often described power as an integral facet of human existence. Psychologist Harry Stack Sullivan (1892-1949) has claimed that power is a more crucial motivation than hunger or thirst.
Rollo May has written about power in terms of individual human potential, referring to the roots of the word "power" in the Latin word posse, which means "to be able." May distinguishes among five levels of intrapsychic power. The most basic level, the power to be, is literally the power to exist, which is threatened if one is denied the basic conditions of human sustenance. The second level, self-affirmation, goes beyond mere survival and involves recognition and esteem by others, while the third, self-assertion, refers to the more strenuous affirmation of one's existence that is required in the face of opposition. The next level of power, aggression , develops when one's access to other forms of self-assertion is blocked. In contrast to self-assertion, which May views as essentially defensive, aggression involves the active pursuit of power or territory. The endpoint in May's continuum of power is violence , which, unlike the other levels, is divorced from reason and verbal persuasion.
Power in its other sense—that of power over others—is a fundamental feature of all relationships, whether each party has a certain degree of power over the other (which is usually the case) or all the power resides with one party. Power may be based on force, acknowledged expertise, the possession of specific information that people want, the ability to reward others, or legitimization (the perception that one has the right to exercise it).
Other bases for power include identification with those who wield it and reciprocity (indebtedness to the wielder of power for providing a prior benefit of some sort). May has described various types of interpersonal power, ranging from harmful to beneficial: exploitative (characterized solely by brute force); manipulative (various types of power over another person); competitive (power against another); nurturing (power for another person); and integrative (power with another person).
power corrupts proverbial saying, late 19th century; now commonly used in allusion to Lord Acton (1834–1902) ‘Power tends to corrupt and absolute power corrupts absolutely.’
power of attorney a document, or clause in a document, giving a person the authority to act for another person in specified or all legal or financial matters.
See also balance of power, knowledge is power, powers.