Power/Knowledge, Society, and Truth: Notes on the Work of Michel Foucault

Mathieu Deflem
www.mathieudeflem.net

Unpublished paper, originally posted April 1999. Revised June 2001, December 2012.
Also downloadable in print-friendly pdf format.

Please cite as: Deflem, Mathieu. 1999. "Power/Knowledge, Society and Truth: Notes on the work of Michel Foucault." Unpublished paper. Available online via www.mathieudeflem.net. 


Introduction

Michel Foucault (1926-1984): there are several developments in his work:
1) Philosophy of Knowledge: Works on various forms of knowledge, science and praxis. Social-constructionist arguments, also anti-humanism and structuralism (philosophy of language, cf. de Saussure). 
2) Social History of Discipline: Then he focused on the internal dynamics of power (regardless of state and market), especially the development of the prison. Foucault, Michel. (1975) 1977. Discipline and Punish. New York: Pantheon. 
3) Governmentality: Extension of discipline, emphasis on the production of power. 
4) Sex: Finally, his sex studies. The end result of all the previous. Interrupted.

The remainder of this text focuses on Foucault's study of discipline and power.


1. Discipline: Technologies, Programs, and Strategies

a) The Transformation of Punishment

The central concern of Discipline and Punish is the disappearance of punishment as a public and violent spectacle centered on the infliction of pain to the body, to the emergence of surveillance of the soul, particularly the development of the prison system. During the years 1760 to 1840, public executions gradually disappeared, and punishment instead became hidden, and concealed. The torture of the body was replaced by the surveillance of the soul (pp.32-47). To account for this transformation, Foucault undertakes to study the concrete systems of punishment, or power as the political technology of the body. Foucault's analysis is a micro-physics of power; it focuses on the transformations of the strategies, tactics, techniques, and concrete functionings of power, to investigate how it came about that nowadays "the soul is the prison of the body" (1977:30). The book is a history of the present.

Foucault first discusses the reform proposals that were suggested since the second half of the 18th century. This reveals that power technologies and discourses are related, but that a concrete analysis of the actual practices can nevertheless show discrepancies between what is thought and what is done. The "great reformers" proposed leniency in punishment, but only at the cost of greater intervention (pp. 82-103). A more efficient economy and technology of punishment was proposed that would allow for a discreet but calculable exercise of power over the soul. The rules of the new punishment included: 1) unarbitrary punishment: for each crime there would be one specific type of punishment; 2) emphasis on crime prevention: the desire for crime had to be eliminated; 3) temporal modulation: punishment can be weakened as it gradually shows its positive effects; 4) punishment should benefit both the offender and the collective: punishment is a lesson in public morality, an open book to read, a school rather than a spectacle; 5) crime is a misfortune: the criminal must be re-educated. Strikingly, the prison system does not fit in this model, but nevertheless detention did become the essential form of punishment. While the prison is also focused at the transformation of the individual, and while prison terms are individualized (in duration and intensity), there are important differences in technology. So why did the prison replace torture?

Foucault then goes on to analyze the nature of the new punishment, what he calls discipline. First, discipline produces docile bodies (pp. 135-169). With the transformation of punishment, the body is seen as something to be manipulated, shaped, trained, and made to obey. As a result of different, multiple processes, which should be described in detail, the body is seen to have entered a machinery of power. The techniques of discipline include: - the distribution of individuals in space (cf. infra); - the control of activity (e.g. by time-tables); the organization of geneses (tasks are divided in temporally regulated sub-tasks); - the composition of forces (individuals are caught up in a maximum-efficiency machine).

Second, discipline relies heavily on the means of correct training (pp. 170-194). These include the following techniques: 1) hierarchical observation: the individuals are clearly visible to permit detailed control and to transform them; it becomes possible for "a single gaze to see everything constantly... a perfect eye" (p.173); supervision is continuous and intense, and "it is the apparatus as a whole that produces 'power' and distributes individuals in this permanent and continuous field" (p.177); 2) normalizing judgment: the goal of punishment is correction, which is opposed to the way penality is judicially conceived; 3) the examination: this combines supervision and normalization; the economy of visibility becomes an exercise of power, a power which is invisible and renders everybody visible; it creates individuality in the form of documentation, the power of writing to analyze the individual as a case, "as effect and object of power, as effect and object of knowledge" (p.192). Power is thus more anonymous, more functional, and more individualizing.

In the panopticon, this type of power finds its supreme manifestation (cf. infra). Here, discipline is a type of power technology that gradually develops into a generalized form of panopticism. Society becomes not a society of spectacle, but a society of surveillance so that everybody comes to live in the panoptic machine.

Within this overall pattern of the changing nature of punishment, the development of the prison system becomes understandable (pp. 231-256). Prisons are intended to make individuals more docile and useful; they are deprived of liberty and subject to transformation. The corrective task of the prisons Foucault considers crucial from the very beginning of the prison system: prison reform is contemporary with its development (p. 234).

The prison is also more than the deprivation of liberty; it provides a structure that is omni-disciplinary, to take care of the whole of the individual. The central principles in this care-taking are: - isolation: the individual is self-regulated (enlightenment from within) within a microcosm of vertically hierarchicized isolated individuals; - work: the prisoner's labor activities are not useful for production but for the effect they have on the human mechanism; - modulation of the penalty: the length of the penalty is related to success of the transformation.

The organization of the prison system is in the hands of an autonomous penitentiary judgment, which is related to medical, economic and political models, while the role of the judge is limited and secondary. Different forms of knowledge sustain the technology of prison surveillance. Knowledge is used for the improvement of the individual, to create permanent documentation and moral accounting. The offender becomes someone to know; the object of knowledge is the delinquent, not his acts but his life. Delinquency, as the official category of deviant acts, is thus created by the prison system (pp. 257-292). The real functions served by the prison system, since it cannot eliminate what it creates, are the ability to distinguish, distribute, and use offences; prisons succeed in producing delinquency, to supervise it, to control this form of isolated illegality.

Foucault sees the prison as only one manifestation of the disciplinary society (pp. 293-308). The "carceral system" becomes complete when cloister, prison, school and regiment come together. Al these systems are characterized by training, observation, knowledge and perpetual assessment of the soul. As the prison spreads farther and farther outwards, the following patterns emerge: - a continuous gradation to pass from a slight departure of a rule, a demand, a norm; - major delinquents are produced (disciplinary careers); there is no outside, everything is saved in the system; - the power to punish is natural and legitimate, strictly a matter of economy without physical violence; - the system becomes universal so that the level from which it is acceptable to punish is lowered; - a new form of law emerges to create the norm (cf. infra); - the society is the prison; - the human sciences accompany and justify power to create an intimate constellation of power-knowledge; - the prison system is deeply rooted in society, even the need for reform, the prison's failure, is included: "the growth of great national or international illegalities directly linked to the political and economic apparatuses (financial illegalities, information services, arms and drugs trafficking, property speculation) makes it clear that the somewhat rustic and conspicuous work force of delinquency is proving ineffective" (p.306).

b) The Nature of Disciplinary Power

From Foucault's analysis in Discipline and Punish some general characteristics of his conception of discipline and punishment can be derived (see Foucault, 1980, 1984; Garland, 1990; Gordon, 1980).

First, there is a distinction between the programs, technologies, and strategies of power. A program of power refers to the general way in which punishment is conceived (Bentham's panopticon is an example). A program is discursive and it still has to be complemented by a technology. It is important to note that the failure of one program (e.g. prisons do not transform criminals) can lead to success of other programs (medical treatment of delinquency). The analysis of a program illuminates a type of discourse. Technology, not in the limited meaning usually given to it, concerns "the government of individuals, the government of the souls, the government of the self by the self, the government of families, the government of children, and so on" (1984:256). Technologies are the actual practices of power. Strategies, finally, are non-discursive and more artificial and improvisational. Strategies are hybrids with concrete effects. Strategic options are opened because technologies have a logic of their own which is always somewhat independent from the program.

Second, Foucault explicitly mentions how power today should be conceived. Several characteristics of contemporary power need to be examined (Foucault, 1980:142).

1) Power is co-extensive with society: discipline is generalized. Power is pervasive throughout the social body to regulate at all times. At the same time, it represents a political anatomy of detail supervising all individual movements and gestures. Therefore, power must make everything and everybody visible: "A fear haunted the latter half of the eighteenth century: the fear of darkened spaces, of the pall of gloom which prevents the full visibility of things, men and truths.... If Bentham's project aroused interest, this was because it provided a formula applicable to many domains, the formula of 'power through transparency', subjection by 'illumination'" (1980:153,154). Power, then, is manifested not only in the prison, but also in the school, the barracks, the hospital, and the asylums. "The asylum is a religious domain without religion, a domain of pure morality, of ethical uniformity. Formerly the house of confinement... was a foreign country. Now the asylum must represent the great continuity of social morality" (1984:148).

2) Relations of power are interwoven with other relations (production, politics, law, kinship) which condition them and are conditioned by them. There is, therefore, not a unilinear relationship between power and state or capitalism. The concreteness of power derives from the fact that it is "more dependent upon bodies and what they do than upon the Earth and its products" (1980:104). The power of the sovereign was still "linked to a form of power that [was] exercised over the Earth and its products, much more than over human bodies and their operations" (1980:104). But today the body of the sovereign is dead, the social body has taken over.

3) Power relations are multiple, of various kinds. The procedures of power today are more diverse than only the disciplinary type (and still include repressive power forms). The principles of visibility and discipline do not govern all technologies of power (Foucault, 1980:148). "Power is not discipline; discipline is a possible procedure of power" (1984:380), and even today there remains a trace of torture in criminal justice (1977:16).

4) The relations of power are multiform and cannot be captured in a dichotomy of dominators and dominated. Precisely because power is neither too concentrated nor too divided, it can go "right down into the depths of society" (1977:27), "down to the finest grain of the social body" (1977:80). Power is non-localized and indiscriminate: "It's a machine in which everyone is caught, those who exercise power just as much as those over whom it is exercised" (1980:156). Power has no single reference point, no one single source: "these tactics were invented and organised from the starting points of local conditions and particular needs. They took shape in piecemeal fashion, prior to any class strategy designed to weld them into vast, coherent ensembles" (1980:159).

Power is a system of "total and circulating mistrust" (1980:158) and absolute intrusiveness: "power reaches into the very grain of individuals, touches their bodies and inserts itself into their actions and attitudes, their discourses, learning processes and everyday live" (1980:39). For instance, "The asylum as a juridical instance recognized no other. It judged immediately, and without appeal. It possessed its own instruments of punishment, and used them as it saw fit". Power in the asylum "did not borrow its modes of repression from the other justice, but invented its own" (1984:155). "Everything was organized so that the madman would recognize himself in a world of judgment that enveloped him on all sides; he must know that he is watched, judged, and condemned; from transgression to punishment, the connection must be evident, as a guilt recognized by all" (1984:156). Therefore "the asylum furnished simultaneously police, magistrates, and torturers. The asylum... is a juridical space where one is accused, judged, and condemned, and from which one is never released except by the version of this trial in psychological depth" (1984:158).

Specialists personify and legitimize these workings of power: "It is not as a scientist that homo medicus has authority in the asylum, but as a wise man. If the medical profession is required, it is as a juridical and moral guarantee, not in the name of science" (1984:159). The medical expert acts not as a scientist but by relying on prestige "which envelops the secrets of the Family, of Authority, of Punishment, and of Love" (1984:161). "In the patient's eyes, the doctor becomes a thaumaturge; the authority he has borrowed from order, morality, and the family now seems to derive from himself; it is because he is a doctor that he is believed to possess these powers" (1984:163).

Thus, power can move "through progressively finer channels, gaining access to individuals themselves, to their bodies, their gestures and all their daily actions" (1980:152). Power produces and is useful, it does not exclude, is not negative (1977:24). Power creates individuals to operate through rather than against them: "Prison professionalised people." (1980:42). Therefore, the individual should be seen as "a reality fabricated by this specific technology of power... called 'discipline'. We must cease once and for all to describe the effects of power in negative terms;... In fact, power produces; it produces reality; it produces domains of objects and rituals of truth" (1977:194). Power is subjectification, and "individuals are the vehicles of power, not its points of application... The individual, that is, is not the vis-a-vis of power; it is, I believe, one of its prime effects" (1980:98).

5) Power is functional because it serves strategies, and power is always related to knowledge. Power is both discourse and practice. These forms of knowledge also justify power in terms of leniency, while in fact it is a matter of certainty and calculability. The discourses of power make sure that the existence of delinquents is socially accepted. This acceptability runs throughout society: in aesthetics, for instance, the criminal is portrayed as the enemy of the poor (1980:45-47). Here the human sciences find their origin. Foucault's evaluation of criminology is often quoted in this respect: "Have you ever read any criminological texts? They are staggering. And I say this out of astonishment, not aggressiveness, because I fail to comprehend how the discourse of criminology has been able to go on at this level. One has the impression that it is of such utility, is needed so urgently and rendered so vital for the working of the system, that it does not even seek a theoretical justification for itself, or even simply a coherent framework. It is entirely utilitarian." (1980:47). Criminological discourse provides the functional alibi that criminal justice is about transformation and truth not punishment. However, Foucault does not just refer to criminology, but to all human sciences, and to knowledge as such. Power is precisely so strong because "it produces effects at the level of desire - and also at the level of knowledge. Far from preventing knowledge, power produces it" (1980:59). Power and knowledge directly imply one another; these are power-knowledge relations (1977:27).

6) There is also always resistance against power, although often it becomes interwoven with power (cf. prison reform). Disciplinary power is omnipresent but not omnipotent. Foucault's work on discipline does not say that power functions automatic, rather it deals with the idea that total control is possible and desirable. Therefore, modern society is disciplinary but not disciplined: the technologies of power are not univocal, there are always points of confrontation and struggle (1977:27).

In conclusion, Foucault's methodological rules for the study of power are the following (1977:23-24): a) do not concentrate on the repressive but on the productive effects of power; b) punishment is not the consequence of legislation, but refers to "techniques possessing their own specificity", to an "apolitical technology of power" (1977:24); c) look for the common matrix between penal law and the human sciences, under the principle of the technologies of power; d) the entire social body is invested by power relations, therefore do not consider any dichotomy to explain away power. 

Therefore, Foucault argues against Marxism where power is conceived solely in terms of the maintenance of the relations of production and class domination. He is also opposed to Marxism's counterpart, liberalism, where power is either seen in terms of contract-oppression (juridical model) or domination-repression (political model). All these theories are inadequate because they do not consider the how of power (1980:88-92). Since the question of who has power is unanswerable, Foucault focuses on how power is embodied in local, regional and material institutions: "Let us ask, instead, how things work at the level of on-going subjugation" (1980:97).


2. The Geography of Power

Foucault's discussions of the panopticon and the spatial and temporal distribution of individuals in power relations clearly indicate his analysis is related to space and its relation to power. However, Foucault insists that space as such is not of too great concern to him: "People have often reproached me for these spatial obsessions, which have indeed been obsessions for me. But I think through them I did come to what I had basically been looking for: the relations that are possible between power and knowledge" (1980:69). Also, in Discipline and Punish he states that he wants to unveil discipline in France alone (1977:309). Nevertheless, occasionally he allows "the frontier to wander about, sometimes over the whole of the West" (1980:67), but only because and when he uses documentation from abroad. I will shortly review how space is nevertheless relevant in Foucault's writings.

a) The Power of Space

The use of space in power relations is for Foucault more than a matter of architecture: "A whole history remains to be written of spaces - which would at the same time be the history of powers (both these terms in the plural) - from the great strategies of geo-politics to the little tactics of the habitat, institutional architecture from the classroom to the design of hospitals, passing via economic and political installations... Anchorage in space is an economico-political form which needs to be studied in detail" (1980:149). Space then is relevant only inasmuch as it reflects a particular technology of power. Doctors, for instance, are specialists of space. They are the first managers of collective space, by analyzing local conditions (environmental conditions of diseases), co-existences (cemeteries), residences (urban problems), and displacements (migration) (1980:150-151).

In an interview Foucault indicates the relative and limited importance of spatial categories in his work: "Territory is no doubt a geographical notion, but it's first of all a juridico-political one: the area controlled by a certain kind of power. Field is an economico-juridical notion. Displacement: what displaces itself is an army, a squadron, a population. Domain is a juridico-political notion. Soil is a historico-geological notion. Region is a fiscal, administrative, military notion. Horizon is a pictorial, but also a strategic notion. There is only one notion here that is truly geographical, that of archipelago... the way in which a form of punitive system is physically dispersed yet at the same time covers the entirety of a society" (1980:68). When space refers to power, then, it is important to consider. Foucault says: "Tactics and strategies deployed through implantations, distributions, demarcations, control of territories and organizations of domains which could well make up a sort of geopolitics... Geography must indeed necessarily lie at the heart of my concerns" (1980:77).

As a follow-up to the previous interview, Foucault (1984:239-256) further explains how during the 18th century architecture was reflected upon as a function of the aims and techniques of the government of societies: how to conceive the organization of space in terms of politics. Every political treatise in those days included chapters on urbanism, hygiene, and private architecture. The territory of the nation was conceived as a city: "the cities, with the problems that they raised, and the particular forms that they took, served as the models for the governmental rationality that was to apply to the whole of the territory... A state will be well organized when a system of policing as tight and efficient as that of the cities extends over the entire territory;... the police become the very type of rationality for the government of the whole territory" (1984:241). Space, then, changed in importance: it was not about territory but about society. The house, for instance, remained until the 18th century an undifferentiated space of several rooms; then they become functionally specified, which reflects a form of morality (1980:148-149). Other problems that are very much bound to space emerge. For instance, the following spaces and their specific problems can be distinguished: 1) urban space (e.g. urban revolts, concentration of workers); 2) railroads: "What was going to happen, for example, if it was possible to get married between Bordeaux and Nantes? Something that was not possible before. What was going to happen when people in Germany and France might get to know one another?" (1984:243); 3) electricity (interrelated networks of energy and communication).

Space, then, can be important when it is used in and as power. Foucault does not use spatial metaphors, he analyzes concrete spatial techniques. Therefore, not the architects but the engineers should be considered because they thought out space, while territory, communication and speed escape the domain of architects. "Space is fundamental in any form of communal life; space is fundamental in any exercise of power" (1984:252).

b) Spaces of Power

Several instances can be mentioned where Foucault analyzes how space is used in power. First, the principle of distribution in discipline includes the following techniques (1977:141-149): a) enclosure: individuals are locked up in closed places which are heterogeneous to all other places (confinement, workshop, factory) to allow for flexible and detailed control; b) partitioning: within these spaces there are further divisions so that each individual has its own space; space is divided into as many sections as there are individuals, to organize an analytical space (e.g. monastic cell); c) functional sites: useful space are created to eliminate confusion (space follows the production process); d) individuals are functionally interchangeable: "The unit is, therefore, neither the territory (unit of domination), nor the place (unit of residence), but the rank: the place one occupies in a classification, the point at which a line and a column intersect, the interval in a series of intervals that one may traverse one after the other" (1977:146). The created spaces are at once architectural, functional and hierarchical; they carve out individual segments and establish operational links. These mixed spaces are at once real and ideal.

One of the most clear examples of power-space is the panopticon (1977:195-228). The panopticon represents a system of centralized policing and exhaustive surveillance that makes all visible and is itself invisible. Bentham thought the panopticon to be both supervising and correcting the delinquent who "is seen, but he does not see; he is the object of information, never a subject in communication" (1977:200). The inmate is visible, the surveillance is unverifiable (not knowing when someone is watching): the principle of the dungeon is reversed, daylight is more effective than darkness (1980:147). The technique itself ensures that it does not matter who exercises power, or when since it functions at all times. Also, this system can be integrated into any function, so that the panopticon can spread all throughout the social body, to become a generalized function (1977:107), to form a disciplinary society (1977:209). Then, the disciplinary mechanisms circulate freely, and the centers of observation are disseminated throughout society.

Another manifestation of the discipline through space is the hospital, where "it was necessary to avoid undue contact, contagion, physical proximity and overcrowding, while... at once dividing space up and keeping it open, ensuring a surveillance which would both be global and individualising while at the same time carefully separating the individuals under observation" (1980:146). The hospital is a "fragment of space closed in on itself, a place of internment and diseases,... the seat of death for the cities" (1980:177).

Another example is the asylum (see Madness and Civilization, 1984:124-167). In the 17th century, Foucault writes, France saw the birth of the "country of confinement", for the poor, the unemployed, the prisoners and the insane, where the directors had "all power of authority, of direction, of administration, of commerce, of police, of jurisdiction, of correction and punishment over all the poor of Paris, both within and without the H“pital G‚n‚ral" (1984:125). This spatial seclusion does not just refer to a place; it is "a semijudicial structure, an administrative entity which, along with the already constituted powers, and outside of the courts, decides, judges, and executes;... [it] is a strange power that the king establishes between the police and the courts, at the limits of the law: a third order of repression (1984:125-126)". Throughout Europe, confinement, as a use of space, was intended to deal with poverty and economic crises. In England, for instance, the beggars had to be concentrated in space because "it was feared that they would overrun the country, and since they could not, as on the Continent, cross the border into another nation, it was proposed that they be 'banished and conveyed to the New-found land, the East and West Indies'" (1984:131). "The community acquired an ethical power of segregation, which permitted it to eject, as into another world, all forms of social uselessness" (1984:136).


3. The Contexts of Power

I have already indicated how Foucault considers it crucial to analyze the concrete technologies of power, and that he refuses to explain away power in terms of politics, law, or economy. However, Foucault does not deny the relevance of the broader contexts of power. His view on these issues is quite complicated. Power, he says, is "a whole complex mechanism, embracing the development of production, the increase in wealth, a higher juridical and moral value placed on property relations, stricter methods of surveillance, a tighter partitioning of the population, more efficient techniques of locating and obtaining information" (1977:77).

a) State and Capital

State and capital, according to Foucault, are not crucial for an analysis of power, yet they should not be ignored. Basically, Foucault's analysis moves from the institutional details of power to the broader patterns: one must conduct "an ascending analysis of power, starting, that is, from its infinitesimal mechanisms, which each have their own history, their own trajectory, their own techniques and tactics, and then see how these mechanisms of power have been - and continue to be - invested, colonised, utilised, involuted, transformed, displaced, extended, etc. by ever more general mechanisms and by forms of global domination" (1980:99). All-encompassing political and economic supra-analyses are both true and false, they can prove anything. Therefore, one must study power historically, beginning from the lowest level, and identify the real agents to see how mechanisms of power became economically advantageous and politically useful. The Marxist conception of the state neglects the technologies of power: "power isn't localised in the State apparatus and nothing will be changed if the mechanisms of power that function outside, below and alongside the State apparatuses, on a much more minute and everyday level, are not also changed" (1980:60). "The new punitive rationality must be relocated in the context of this technology, itself linked to the demographic, economic, and political changes which accompany the development of industrial states" (1984:338).

In his historical analysis of power Foucault precisely argues that power is no longer tied up to the "central spirit" (1980:98). Torture as a spectacle was still linked to the right of the superior man, the will of the sovereign. The execution had a juridico-political function: it concerned the reconstitution of the injured sovereign, and the exercise of terror was the triumph of the law (1977:47-57). Legal rights (and crimes) in Medieval times referred to the rights of the King. Rights still meant rights of sovereignty and the obligation to obey it; the right of death and life was still dependent on the King (see History of Sexuality, 1984:258).

The new technologies of power "cannot be localized in a particular type of institution or state apparatus. For they have resource to it; they use, select or impose certain of its methods. But, in its mechanisms and its effects, it is situated at a quite different level" (1977:26). The body of the King has been replaced by the body of society; the social body is healed by removing the sick, the exclusion of delinquents. The social body is constituted by "the materiality of power operating on the very bodies of individuals" (1980:55). Thus, after the King has died, domination comes from the subjects themselves and is revealed in their mutual relations (1980:94-96), because power "is never localised here or there, never in anybody's hands, never appropriated as a commodity or piece of wealth (1980:98). "To pose the problem in terms of the State means to continue posing it in terms of sovereign and sovereignty, that is to say in terms of law. If one describes all these phenomena of power as dependent on the State apparatus, this means grasping them as essentially repressive: the Army as a power of death, police and justice as punitive instance, etc. I don't want to say that the state isn't important; what I want to say is that relations of power, and hence the analysis that must be made of them, necessarily extends beyond the limits of the State" (1980:122).

The new forms of punishment form a "synoptic regime of power, a regime of its exercise within the social body, rather than from above it" (1980:39). This lead to the abolition of the King; the mythology of the sovereign has ended. Then, discipline spreads through society and concerns the human body of all living subjects and the regulatory control of the entire population. Life is an object of calculation. Yet, "the system of power takes a pyramidical form. Hence there is an apex. But even so, even in such a simple case [in the army or the factory], this summit doesn't form the 'source' or 'principle' from which all power derives" (1980:159). Thus, there is "a certain correlation between the two processes, global and local, but not an absolute one" (1980:39). The technologies of power are crucial in themselves, although Foucault realizes that these mechanisms can be annexed by more global phenomena, more general, political powers and economic interests (see Rabinow, 1984). But the State did not simply confiscate these power mechanisms; the apparatuses rested on small, regional, dispersed panopticisms. Therefore, "one cannot confine oneself to analysing the State apparatus alone if one wants to grasp the mechanisms of power in their detail and complexity... I do not mean in any way to minimise the importance and effectiveness of State power. I simply feel that excessive insistence on its playing an exclusive role leads to the risk of overlooking all the mechanisms and effects of power which don't pass directly via the State apparatus, yet often sustain the State more effectively than its own institutions, enlarging and maximising its effectiveness" (1980:72-73). (note that Foucault uses the word 'global' throughout his work to indicate broader forces and patterns, e.g. political, economic, legal).

Foucault applies the same logic to the relationship between power and economy or Capital. This may seem somewhat surprising since Foucault often discusses the economies of power. The panoptic system of surveillance, for instance, involves very little expense; it only needs an inspecting gaze, present or not (1980:155). In addition, sometimes Foucault quite explicitly refers to "market-mechanisms" in the explanation of power. For instance, he states that "the economic changes of the eighteenth century made it necessary to ensure the circulation of effects of power through progressively finer channels" (1980:151). In Discipline and Punish he discusses the rise of "economic" crime, and states that the illegality of property "was intolerable in commercial and industrial ownership: the developments of the ports, the appearance of great warehouses in which merchandise was stored, the organization of huge workshops ... also necessitated a severe repression of illegality" (1977:85). Delinquents are useful "in the economic domain as much as the political" (1980:40). Also, punishment by discipline emerged with "the new forms of capital accumulation, new relations of production and the new legal status of property;... the economy of illegalities was restructured with the developments of capitalist societies " (1977:86, 87). The confinement of the mad, too, was related to economic motives: the beggars, the unemployed were no longer driven away but taken in charge. Confinement was an answer to economic crises, so that there was cheap labor in periods of full employment, and protection against agitation in periods of unemployment. Moreover, private entrepreneurs could utilize the manpower in the asylums. Yet, the asylums did not play this double role effectively, and economic conditions alone therefore cannot account for the rise of madness. What was needed was a new moral perception and institution.

In other instances, Foucault indeed limits the economic motives of power, because "economic reasons could become determinant only with a technical transformation" (p.163). The spread of discipline throughout society could only occur because "the technological mutations of the apparatus of production, the division of labour and the elaboration of the disciplinary techniques sustained an ensemble of very close relations... Each makes the other possible and necessary; each provides a model for the other" (1977:221). The relation of the localities of control with the global structures of State and capital then is mutual, co-determinant, aiding one another: "The growth of the capitalist economy gave rise to the specific modality of disciplinary power, whose general formulas, techniques of submitting forces and bodies, in short, 'political anatomy', could be operated in the most diverse political r‚gimes, apparatuses or institutions" (1977:221).

b) Law and Police

Foucault's assessment of law is similar to his evaluation of the State: legal processes matter given specific technologies of power, which are captured by them or may conflict with them. On the one hand, Foucault rejects any preoccupation with law to account for power: "One impoverishes the question of power if one poses it solely in terms of legislation and constitution, in terms solely of the state and the state apparatus" (1980:158). In Discipline and Punish, for instance, his intent was to study punishment as a social phenomena "that cannot be accounted for by the juridical structure of society alone" (1977:24). "Instead of seeking the explanation in a general conception of the law, or in the evolving modes of industrial production (as Rusche and Kirchheimer did), it seemed to me far wiser to look at the workings of Power. I was concerned not with some omnipresent power,... but with refinement, the elaboration and installation since the seventeenth century, of techniques for 'governing' individuals" (1984:337-338). Thus, "the prison is not the daughter of law, codes or the judicial apparatus" (1977:307). The disciplinary mechanisms of power are not applied to a central law but to the apparatus of punishment. Actually, the rise of disciplinary power is even opposed to judicial penality which refers to a corpus of laws and texts, not to observable phenomena; it is a penality of the norm "which is irreducible in its principles and functioning to the traditional penality of the law" (1977:183). This contradiction is crucial; the conflicts between them become greater to reveal a complete incompatibility. Therefore, power should not be confused with law, should not be seen as a prohibition, something negative. Instead, power is "a productive network which runs through the whole social body" (1980:119). "What we need today... is a political philosophy that isn't erected around the problem of sovereignty, nor therefore around the problems of law and prohibition. We need to cut off the Kings' head" (1980:121), because "power is always already there" (1980:141).

On the other hand, Faucault does not deny the importance of legal phenomena. While discipline is not an extension of the juridico-political structures of society, it is not independent from it, since the system of rights was supported by the micro-powers of discipline. But the panoptic techniques generalize the power to punish, and do not represent a universal consciousness of law (1977:221-224). The law in a way follows discipline. The judges, for instance, become obsessed by a desire for prison; they determine the penalties, but are merely employees of the prison apparatus (1977:282). Law is secondary; it sustains, legitimates, as a type of discourse, what technologies of power have already established. Juridical systems enabled the sovereignty of the State to be democratized (from the King to the social body), although and after this dated conception of sovereignty was already undermined by disciplinary coercion. "In reality, the disciplines have their own discourse... Disciplines are the bearers of a discourse, but this cannot be the discourse of right. The discourse of discipline has nothing in common with that of law, rule, or sovereign will... The code they come to define is not that of law but that of normalization" (1890:106). An exclusive orientation to law would be too one-sided: "The law always refers to the sword... I do not mean to say that the law fades into the background or that the institutions tend to disappear, but rather that the law operates more and more as a norm, and that the judicial institution is increasingly incorporated into a continuum of apparatuses (medical, administrative, and so on) whose functions are for the most part regulatory. A normalizing society is the historical outcome of a technology of power centered on life. We have entered a phase of juridical regression." (1984:266).

Finally, Foucault's discussion of police also indicate this relative unimportance as compared to the practice and technology of power. Again, Foucault does not deny that the (official) police is a matter of control by the State. Historically the police was at first even the direct expression of royal absolutism. At that time, the economic, criminal, urban and political surveillance functions were al united and centralized. But even then "the type of power that it exercises, the mechanisms it operates and the elements to which it applies them are specific" (1977:213). To be sure, the police played a crucial role in the prison system. Isolated illegality in the prison is not possible without the police, which surveyed everything, and managed the documentary system (description of criminals and arrest warrants). Police, prison, and delinquency support one another (1977:280-281). In addition, with the spread of discipline, the police even become a police of the entire society; everything that happens must be checked with "thousands of eyes posted everywhere, mobile attentions ever on the alert, a long, hierarchized network" (1977:214). The police thus always unites the King's powers and the lowest levels of power in all of society. The police was thus instrumental in generalizing discipline throughout society, but this does not mean that the state absorbed all the disciplinary functions once and for all. The fear of crime makes the police acceptable: "What makes the presence and control of the police tolerable for the population, if not fear of the criminal? This institution of the police, which is so recent and so oppressive, is only justified by that fear" (1980:47).

Foucault sometimes defines police more broadly than what it usually refers to (the previous comments seem to apply to the official, state-controlled police only). With respect to confinement, for instance, he defines police as "the totality of measures which make work possible and necessary for all those who could not live without it" (1984:128). Police, he says, is often misunderstood; it means "a program of government rationality. This can be characterized as a project to create a system of regulation of the general conduct of individuals whereby everything would be controlled to the point of self-sustenance, without the need for intervention;... [D]uring the eighteenth and nineteenth century, there appeared... this idea of a police that would manage to penetrate, to stimulate, to regulate, and to render almost automatic all the mechanisms of society. This idea has since been abandoned. The question has been turned around... That is, what is the principle of limitation that applies to governmental actions such that things will occur for the best, in conformity with the rationality of government, and without intervention?" (1984:241, 242).

c) What Has To Be Done?
"My point is not that everything is bad, but that everything is dangerous"(Foucault, 1984:343). 
From the foregoing the picture that emerges from Foucault's analysis of power seems to offer little room for human beings to fight power: if technologies of power operate so vigorously, and are spread all over society, is counter-power then impossible, or does Foucault offer a way out? There seems to be a way out, but the fight against power can no longer be conceived in terms of the fight against State, Capital or Law. In line with his discontent with totalitarian, global theories, Foucault asserts that criticism should be local, presenting a "multiplicity of genealogical researches" (1980:83), or a history of struggles, as an attempt to emancipate historical knowledges from subjection (1980:85). Theory has to be connected with practice "within specific sectors, at the precise points where their own conditions of life or work situate them (housing, the hospital, the asylum, the laboratory, the university, family and sexual relations)" (1980:126). In this process of localized struggles the intellectual only provides instruments of analysis: the fighters have to do their own fighting. The philosopher can only give a survey of the battlefield (1980:61-62).

Asked whether these local struggles are not in danger of remaining useless attempts, Foucault says that the weakest links have to be attacked so that the whole structure could be demolished (1980:143). "The role for theory today seems to me to be just this: not to formulate the global systematic theory which holds everything in place, but to analyse the specificity of mechanisms of power" (1980:145). Nevertheless, Foucault admits that one may run "the risk of being unable to develop these struggles for lack of a global strategy or outside support" (1980:130). Therefore, he considers cooperation across local struggles useful: everybody can participate in "a global process of politicisation of intellectuals" (1980:127). Particularly in recent time this seems necessary: "since the nuclear threat affected the whole human race and the fate of the world, [the intellectual's] discourse could at the same time be the discourse of the universal. Under the rubric of protest, which concerned the entire world, the atomic expert brought into play his specific position in the order of knowledge" (1980:128). Instead of developing global projects, one must employ specific tests.


References

Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison. New York: Vintage Books.

____. 1980. Power/Knowledge: Selected Interviews and Other Writings 1972-1977 (edited by Colin Gordon). New York: Pantheon Books.

____. 1984. Selected chapters ("The great confinement; The birth of the asylum; Space, knowledge, and power; Right of death and power over life; Preface to The History of Sexuality, Volume II; On the genealogy of ethics; Politics and Ethics.") In P. Rabinow (ed.), The Foucault Reader. New York: Pantheon Books.