3

Reframing the Theory: Biopower and Governmentality

Much of Foucault’s initial analysis of power has now been presented—we’ve got the general theory or framework, and we’ve seen Foucault’s analysis of disciplinary power, a form of power found at the molecular level of modern society. But Foucault came to realize that disciplinary power alone is not adequate to explain the entirety of modern power. He thus introduces a new element, which he initially termed “biopower,” to encompass important macrophenomena that cannot be characterized as “sovereign power” or in terms of the “juridico-discursive model.” This in turn forced him to revise his initial hypothesis that all macro forms of power could be derived entirely from micro forms. Instead, Foucault’s analysis of power is reframed—most explicitly in the 1978 course at the Collège de France—so that macro and micro forms of power can be understood as interrelated but not necessarily reducible the one to the other. To effect this reframing Foucault introduces two new central concepts into his analysis of power: biopower and governmentality.

Biopower denotes certain macro forms of power, especially those made possible through statistical analysis, power that is exercised at the level not of individuals (as was the case in disciplinary power) but populations. Norms (a point of difficulty or tension for Foucault’s earlier analyses, which failed to appreciate the independence of the macro level), then, can be more adequately understood as one of the important conduits or interfaces between macropowers of populations and micropowers of individuals.

Governmentality emerges in Foucault’s discussions as a sort of umbrella term—encompassing both macro and micro forms of power, both the institutions of the state and more local institutions like schools, hospitals, and barracks—that also includes within its penumbra “the government of oneself,” an orientation that will loom increasingly important in Foucault’s later work. (One could even use a phrase such as “the logic of government” as a rough synonym for “governmentality.” But the term “governmentality,” in both English and French, contains the term “mentality,” thus perhaps suggesting that the mechanisms described are not just a logic but a mentality—something more pervasive than a mere logic, something like what William James terms a “temperament” [James 1907, 8ff.]. Admittedly, this observation is a bit “cutesy,” but its suggestiveness is helpful for seeing how a Foucauldian ethics will orient itself and, in particular, how such an ethics is rooted in an attitude of critique.)

My agenda in this chapter is to “complete” our survey of Foucault’s analysis of power, so that we can see how it constitutes the “background” or framework for his ethical thinking and how it provides certain resources for an ethical project. We shall see how tensions in the account so far—centered on the emergence of disciplinary techniques and their diffusion throughout society—lead Foucault to supplement and reconfigure this analysis with an entirely different “macro” level of power relations, ones not reducible to the microrelations of discipline. This new, more general understanding of how power works serves to reframe its problems in terms of “governmentality,” thus providing an opening for ethical possibilities that would seem to have been eliminated at the end of Discipline and Punish.

Foucault’s move away from the bleak portrait of his Hobbesian hypotheses would be clarified by 1978 in his Collège de France course that year, Security, Territory, Population. But tensions in the closed account that he attempted to articulate in Discipline and Punish would begin to emerge almost immediately. The clearest indication of these tensions—and the new directions that his analysis of the “power of normalization” will take—emerge in two texts from 1976: his Collège de France course that year, “Society Must Be Defended,” and the fifth (and final) part of La volonté de savoir (The History of Sexuality, volume 1, whose fourth part, as we saw in Chapter 1, produced a critical articulation of his general theory of power), in both of which he explicitly explores the extent to which politics can be understood as “the continuation of war.”

POLITICS AS THE CONTINUATION OF WAR

We can hear this theme in Discipline and Punish: “it must not be forgotten that ‘politics’ has been conceived as a continuation, if not exactly and directly of war, at least of the military model as a fundamental means of preventing civil disorder” (1975ET, 168). Indeed, Foucault had been thinking along these lines since at least 1973: in the November 7, 1973, opening lecture of his Collège de France course, for example, he described the dynamics of power and knowledge in the asylum as “actually a battlefield” (CdF74ET, 7). But the theme of politics (and society) as a kind of war becomes his explicit focus in the 1976 Collège de France course: the final lecture summarizes that “I have been trying to raise the problem of war, seen as a grid for understanding historical processes” (CdF76ET, 239). Arnold Davidson, in his introduction to this course, agrees that this course “is Foucault’s most concentrated and detailed historical examination of the model of war as a grid for analyzing politics” (Davidson 2003, xviii).

Foucault opens this lecture course with the question, “What is power?” (CdF76ET, 13), to which he answers (following the argument I’ve presented in the first two chapters) that “it is primarily, in itself, a relationship of force” (CdF76ET, 15). Which leads to a new question—“shouldn’t we be analyzing it first and foremost in terms of conflict, confrontation, and war?”—and a new hypothesis to be tested: “Power is war, the continuation of war by other means. At this point, we can invert Clausewitz’s proposition and say that politics is the continuation of war by other means” (CdF76ET, 15). Giving us the theme for the year’s lectures, Foucault makes this hypothesis quite explicit:

I would like to try to see the extent to which the binary schema of war and struggle, of the clash between forces, can really be identified as the basis of civil society, as both the principle and motor of the exercise of political power. (CdF76ET, 18)

Is something analogous to war really “the basis of civil society,” its “principle and motor,” as would be suggested by Foucault’s Hobbesian hypothesis and the reduction of all power to a disciplinary “relationship of forces”? This is the hypothesis to be tested in these courses. And while we might anticipate that his examination would confirm this hypothesis, on the contrary, in fact, it will lead us to reexamine and reassess it.

Foucault opens the second lecture by reiterating that this course constitutes “a series of investigations into whether or not war can possibly provide a principle for the analysis of power relations: can we find in bellicose relations, in the model of war, in the schema of struggle or struggles, a principle that can help us understand and analyze political power” (CdF76ET, 23). He then notes several “methodological cautions” that should guide these investigations—cautions that reiterate the elements of his general theory of power. Two of these are particularly important.

First, although “the system of right and the judiciary field are permanent vehicles for relations of domination, and for polymorphous techniques of subjugation” (CdF76ET, 27), we must “not regard power as a phenomenon of mass and homogenous domination” (CdF76ET, 29). Thus, while our institutions can serve to dominate us, power is not exclusively or essentially something that dominates.

Second,

we should make an ascending analysis of power, or in other words begin with its infinitesimal mechanisms, which have their own history, their own trajectory, their own techniques and tactics, and then look at how these mechanisms of power . . . have been and are invested, colonized, used, inflected, transformed, displaced extended, and so on by increasingly general mechanisms and forms of overall domination. (CdF76ET, 30)

Begin, in other words, with the local microrelations of power, and then look to see how they combine to constitute macrorelations such as a general domination.

These cautions remind us of the contrast with other, inadequate and incomplete, analyses of power, which distinguishes Foucault’s analysis as a “theory” of power. More particularly, they fit with—and bring into relief—the implications that Discipline and Punish left us with: that power operates at an infinitesimal or micro level upon individuals in ways that lead to larger macroeffects of domination. However, what Foucault proposes here is that our method should begin with infinitesimal microrelations; this methodological caution does not necessarily entail that macro power relations are entirely analyzable in terms of microrelations. This is the insight toward which Foucault is beginning to work. It will open up both a way out of the apparent “iron cage” and an intriguing ethico-strategic possibility, with which the lecture closes:

Truth to tell, if we are to struggle against disciplines, or rather against disciplinary power, in our search for a nondisciplinary power, we should not be turning to the old right of sovereignty; we should be looking for a new right that is both nondisciplinary and emancipated from the principle of sovereignty. (CdF76ET, 39–40)

The first two lectures had been published independently of the rest of the course as “Two Lectures” (DE193, DE194). They offer a nice summary of the stakes and scope of Foucault’s analysis of power, but in themselves they do not allow us to see how Foucault will ultimately assess this Hobbesian hypothesis. The fifth lecture, however, almost halfway through the course, explicitly addresses Hobbes on war. “Hobbes . . . does, at first glance, appear to be the man who said that war is both the basis of power relations and the principle that explains them” (CdF76ET, 89). For Hobbes, war is what characterizes the “state of nature” out of which sovereignty is born. We create sovereignty out of fear, to protect us from others; thus, “the will to prefer life to death: that is what founds sovereignty” (CdF76ET, 95) and constitutes the right of the sovereign to take life or let one live. Thus, “for sovereignty to exist there must be—and this is all there must be—a certain radical will that makes us want to live” (CdF76ET, 96). This effects a sort of reversal in how we should read Hobbes, Foucault argues: “far from being the theorist of the relationship between war and political power, Hobbes wanted to eliminate the historical reality of war, as though he wanted to eliminate the genesis of sovereignty” (CdF76ET, 97). By making war the “state of nature” left behind in the establishment of sovereignty, by making sovereignty a kind of perpetual given, war becomes in a certain sense “off the table” for contemporary struggles.

What would such an elimination accomplish? “The enemy—or rather, the enemy discourse Hobbes is addressing—is the discourse that could be heard in the civil struggles that were tearing the State apart in England at this time” (CdF76ET, 99). And so “What Hobbes is trying, then, not to refute, but to eliminate and render impossible . . . is a certain way of making historical knowledge work within the political struggle” (CdF76ET, 98). But Foucault’s entire project is devoted first of all to showing that and how historical knowledge and historical particularities do work within political struggles, and second and just as importantly, to harnessing and using historical knowledge within ongoing struggles. His analysis of Hobbes—and of what Hobbes was arguing against (indeed, Foucault is more interested in Hobbes’s discursive opponents’ insights)—shows, Foucault thinks, that “the logical and historical need for rebellion is therefore inscribed within a whole historical analysis that reveals war to be a permanent feature of social relations” (CdF76ET, 110). So taking war to be a “permanent feature of social relations” has a particular value: it entails a “need for rebellion,” that is, it implies that the status quo cannot become entirely closed in domination, that historical knowledge and circumstances can be exploited in resistance.

Foucault’s insight contra Hobbes here—that understanding social relations on the model of war makes it possible to cognize and strategize resistance to those social relations—echoes a point Simone de Beauvoir had made in The Ethics of Ambiguity:

To the idea of present war there is opposed that of a future peace when man will again find, along with a stable situation, the possibility of a morality. But the truth is that if division and violence define war, the world has always been at war and always will be; if man is waiting for universal peace in order to establish his existence validly, he will wait indefinitely: there will never be any other future. (de Beauvoir 1948, 119)

Beauvoir’s point (oriented not toward a mythical past, as in Hobbes’s account, but a mythical future) is that we will never be “free” from the struggle and division that make society seem to be a sort of ongoing war. She rejects the view that “the possibility of morality” would obtain only in some “future peace” because such a peace will never come to be. But Beauvoir also insists that we can “establish [our] existence validly” in our current (and inevitable) circumstances of ongoing struggle—indeed, these are the conditions upon which, for her, morality must be created. I note this parallel here because Foucault will make a very similar move (and we can already see the first step in his discussion here): ethics becomes possible within the circumstances of struggle, within the network of power relations in which we are situated. Indeed (this is the first step), that those power relations can be understood in terms of struggle opens up certain possibilities for what kinds of ethics might emerge.

However, Foucault will recognize that this analogy of political power, and social relations more generally, to war is, while useful, incomplete and limited—especially if those power relations are understood only in terms of discipline. He acknowledged or suspected these limitations even at the beginning of the course: “I think that the twin notions of ‘repression’ and ‘war’ have to be considerably modified, and ultimately, perhaps, abandoned” (CdF76ET, 17). We saw in Chapter 1 that “repression” as an exhaustive model of power clearly must be abandoned. But the model of “war,” too, must be modified. He had attempted to understand power in the light of this analogy as “strategic relations” best exemplified in disciplinary techniques. But by the course’s conclusion, in fact, Foucault is already articulating a major addition and alteration to his understanding of disciplinary power:

And I think that one of the greatest transformations political right underwent in the nineteenth century was precisely that . . . sovereignty’s old right—to take life or let live— . . . came to be complemented by a new right which does not erase the old right but which does penetrate it, permeate it. . . . It is the power to “make” live and “let” die. (CdF76ET, 241)

This new, inverted, “opposite” “right” reveals the “inner logic,” one might say, of what he will call “bio-politics.” This is especially intriguing, given that Foucault had closed the second lecture by noting that we needed just such a “new right” to “struggle against disciplinary power” (CdF76ET, 39–40). A new right, corresponding to a kind of power that cannot be reduced to discipline, will (even if it poses new dangers of domination) at the very least add new dimensions for any struggles of resistance and new openings through which to escape a seemingly “complete and austere” prison.

Though we will return shortly to this final lecture of his 1976 course, we can get a clearer picture of this new biopolitical right “to ‘make’ live and ‘let’ die” in the final part (entitled “Right of Death and Power Over Life”) of La volonté de savoir: “one would have to speak of bio-power to designate what brought life and its mechanisms into the realm of explicit calculations and made knowledge-power an agent of transformation of human life” (1976ET, 143). As Foucault presents it, “this power over life evolved in two basic forms”—the micro form of discipline and the macro form of biopower—“these forms were not antithetical, however; they constituted rather two poles of development linked together by a whole intermediary cluster of relations” (1976ET, 139). Foucault is here reorganizing the landscape of his analysis of power relations: having just completed an analysis premised upon the assumption that end-forms like the state can and must be analyzed in terms of more general micro forms in Part 4, he is now altering that analysis to allow for a macro form that is also “basic”—one that is not antithetical to but linked together with disciplinary micropower.

Foucault explicitly identifies disciplinary power as the first of these two forms:

One of these poles—the first to be formed, it seems [an interesting caveat, given how Foucault’s analysis has developed]—centered on the body as a machine: its disciplining, the optimization of its capabilities, the extortion of its forces, the parallel increase of its usefulness and its docility, its integration into systems of efficient and economic controls, all this was ensured by the procedures of power that characterized the disciplines: an anatomo-politics of the human body. (1976ET, 139)

He thus recharacterizes disciplinary power as an “anatomo-politics” to highlight the parallelism between it and the second form, “bio-power” or “bio-politics,” which is addressed not to the individual anatomy but to the bios, life in general:

The second, formed somewhat later, focused on the species body, the body imbued with the mechanics of life and serving as the basis of the biological processes: propagation, births and mortality, the level of health, life expectancy and longevity, with all the conditions that can cause these to vary. Their supervision was effected through an entire series of interventions and regulatory controls: a bio-politics of the population. (1976ET, 139)

Thus, Foucault has begun to recognize a macropower, aimed at the species and populations through regulatory mechanisms, distinct from disciplinary micropower. As he describes it in the final lecture of “Society Must Be Defended,” “This new technique does not simply do away with the disciplinary technique, because it exists at a different level, on a different scale, and because it has a different bearing area, and makes use of different instruments” (CdF76ET, 242). A “different level,” a “different scale,” and a “different bearing area” are all ways of saying that this power functions between and upon macrophenomena. And so this macropower “does not exclude disciplinary technology, but it does dovetail into it, integrate it, modify it to some extent, and above all, use it by sort of infiltrating it, embedding itself in existing disciplinary techniques” (CdF76ET, 242). The two modes of power are distinct and irreducible to the other; they are also—even if he is not yet sure exactly how—interwoven and integrated: “the two sets of mechanisms—one disciplinary and the other regulatory—do not exist at the same level. Which means of course that they are not mutually exclusive and can be articulated with each other” (CdF76ET, 250).

Together, in their mutual articulation, “this great bipolar technology—anatomical and biological . . . —characterized a power whose highest function was perhaps no longer to kill, but to invest life through and through” (1976ET, 139). Power at all levels is, in other words, fundamentally productive. And the two technologies, the micro- and the macro-, are conjoined and integrated not “at the level of a speculative discourse, but in the form of concrete arrangements that would go to make up the great technology of power in the nineteenth century: the deployment of sexuality would be one of them, and one of the most important” (1976ET, 140).

This is the background that enables us to understand the importance assumed by sex as a political issue. It was at the pivot of the two axes along which developed the entire political technology of life. On the one hand it was tied to the disciplines of the body. . . . On the other hand, it was applied to the regulation of populations through all the far-reaching effects of its activity. (1976ET, 145)

Foucault defines “sex” not as a drive but as “the most internal element in a deployment of sexuality” (1976ET, 155), which is itself “an especially dense transfer point for relations of power: between men and women, young people and old people, parents and offspring, teachers and students, priests and laity, an administration and a population” (1976ET, 103). Note the pairings that Foucault has included here—these indicate the scope of power relations to penetrate something as seemingly intimate as sexuality and the different ways that sexuality can become a vehicle for both micro- and macrorelations. “Sexuality exists at the point where body and population meet” (CdF76ET, 251–252). Foucault continues that “sexuality is not the most intractable element in power relations, but rather one of those endowed with the greatest instrumentality: useful for the greatest number of maneuvers and capable of serving as a point of support, as a linchpin, for the most varied strategies” (1976ET, 103). Indeed, “it is through sex—in fact, an imaginary point determined by the deployment of sexuality—that each individual has to pass in order to have access to his own intelligibility . . . to the whole of his body . . . to his identity” (1976ET, 155–156).

In Part 4, Foucault sketched four especially significant such points of support: “it seems that we can distinguish four great strategic unities which, beginning in the eighteenth century, formed specific mechanisms of knowledge and power centering on sex” (1976ET, 103). Foucault identified these four strategies or “deployments” as “hysterization of women’s bodies,” “pedegogization of children’s sex,” “socialization of procreative behavior,” and “psychiatrization of perverse pleasure” (1976ET, 104–105). All four of these strategies are composed of both disciplinary techniques and larger macro-level power relations such as the regulation of populations and the construction of racial identities. We can thus situate the planned trajectory of Foucault’s subsequent volumes in the History of Sexuality series—studies that this first volume was meant to inaugurate but which were never published; each was to focus on one of these “deployments of sexuality.”

In these two 1976 texts, Foucault does begin to articulate how the connection between disciplinary micropowers and biopolitical macropowers is effected—through the vehicle of norms:

There is one element that will circulate between the disciplinary and the regulatory, which will be applied to body and population alike, which will make it possible to control both the disciplinary order of the body and the aleatory events that occur in the biological multiplicity. The element that circulates between the two is the norm. . . . The normalizing society is a society in which the norm of discipline and the norm of regulation intersect along an orthogonal articulation. (CdF76ET, 252–253)

And so, for example, sexuality will be one of the points or lines of orthogonal articulation. He adds that biopower “needs continuous regulatory and corrective mechanisms. . . . Such a power . . . effects distributions around the norm” (1976ET, 144). Thus, to pull this together, “a normalizing society is the historical outcome of a technology of power centered on life” (1976ET, 144). Foucault will have more to say about the details of how norms work in his 1978 course, which we’ll turn to shortly. But he makes another important connection here, linking these norms with the function of racism.

Racism “is primarily a way of introducing a break into the domain of life that is under power’s control. . . . It is a way of separating out the groups that exist within a population” (CdF76, 245–255). Kenneth Stampp’s discussion of slavery, which I earlier used to illustrate disciplinary power, supports this analysis as well: “Slavery was, above all, a method of regulating race relations, an instrument of social control” (Stampp 1956, 387). Stampp goes on to connect his analysis with the framework that had driven Foucault’s investigations here (bringing our discussion full circle): to the hypothesis of understanding power relations (and society in general) in terms of war. He notes that slaves’ basic attitude toward whites (and especially enslavers) was “an attitude of deep suspicion.” He continues: “When this was the Negro’s basic attitude, the resulting relationship was an amoral one which resembled an unending civil war; the slave then seemed to think that he was entitled to use every tactic of deception and chicanery he could devise” (380). The most basic relationship structuring the social life of the South could be construed as a sort of “war,” which authorized slaves to employ whatever tactics that were available for their resistance.

I suspect that Foucault’s continuing analysis of these examples—the deployments of sexuality and racism—may have been what helped Foucault understand a key error in his analysis up to this point, namely that many of these macrorelations cannot be entirely analyzed in terms of micropractices. This realization will drive the next great theoretical shift in Foucault’s understanding of power and in Foucault’s understanding of the contexts for ethics. For as he notes here:

It is the agency of sex that we must break away from, if we aim—through a tactical reversal of the various mechanisms of sexuality—to counter the grips of power with the claims of bodies, pleasures, and knowledges, in their multiplicity and their possibility of resistance. (1976ET, 157)

We will return to these remarks—which herald one launching point for Foucauldian ethics—in Chapter 4.

However, elements remain in both of these texts from 1976 that could be construed to support the bleak view of power as an all-encompassing iron cage. Consider two passages from The History of Sexuality. First, a longer passage that ends on a bad note:

If the development of the great instruments of the state, as institutions of power, ensured the maintenance of production relations, the rudiments of anatomo- and bio-politics, created in the eighteenth century as techniques of power present at every level of the social body and utilized by very diverse institutions (the family and the army, schools and the police, individual medicine and the administration of collective bodies) . . . acted as factors of segregation and social hierarchization . . . guaranteeing relations of domination and effects of hegemony. (1976ET, 141)

This list of institutions includes those that he had studied in Discipline and Punish (the army, schools, and police), in The History of Sexuality (the family), and in another text that I will discuss in the next section, “The Politics of Health in the Eighteenth Century” (individual and collective medicine). Thus, it would not be unreasonable to hear, when Foucault says these techniques “guarantee” domination and hegemony, a continuation of the bleak analyses of the end of Discipline and Punish. He continues (in the second passage I want to highlight in this regard) that “power would no longer be dealing simply with legal subjects over whom the ultimate domination was death, but with living beings, and the mastery it would be able to exercise over them would have to be applied at the level of life itself” (1976ET, 142–143). So, it seems, hegemony and domination are guaranteed through mastery of living being. These passages seem to cohere with one from the final lecture of the Collège de France course:

Beneath that great absolute power, beneath the dramatic and somber absolute power that was the power of sovereignty, and which consisted in the power to take life, we now have the emergence, with this technology of biopower, of this technology of power over “the” population as such, over men insofar as they are living beings. . . . Sovereignty took life and let live. And now we have the emergence of a power that I would call the power of regularization, and it, in contrast, consists in making live and letting die. (CdF76ET, 247)

These three passages could be read, as I’ve suggested, in a negative way. In saying that that power guarantees domination and hegemony and that its mastery now extends over all life, by making it live and letting it die, Foucault could be understood to connote that “war” organizes all of our relationships, that our options have been foreclosed by the emergence of these new technologies of power. But this interpretation is not the only one available. And Foucault gives us a clear indication that, while he may have seriously explored that possibility, it is a mistaken reading:

The normalizing society is therefore not, under these conditions, a sort of generalized disciplinary society whose disciplinary institutions have swarmed and finally taken over everything—that, I think, is no more than a first and inadequate interpretation of a normalizing society. (CdF76ET, 253)

THE IMPORTANCE OF POPULATION

In both the 1976 Collège de France course and La volonté de savoir Foucault has clearly begun a process of autocritique, revising and complicating the understanding of power whose articulation he had just completed. The tension that emerges in both of these texts, but is not yet fully controlled, inheres in the claim that all power could be analyzed in terms of disciplinary relations. Indeed, Foucault was already beginning to recognize that this reduction cannot be maintained and that there are certain macrophenomena (and macro power relations) that must be explained in other terms. The notion of “population” is central for this clarification.

We can see how Foucault’s conception of population evolved by contrasting his discussions of it between 1974 and 1979. In a lecture given in October 1974 in Rio de Janeiro, Brazil (lectures that reflect Foucault’s then-current view of power as principally disciplinary),1 Foucault discusses “populations” as essentially reducible to individuals embedded in microrelations:

The individual thus emerges as an object of medical knowledge and practice.

At the same time, through the same system of disciplined hospital space, one can observe a great number of individuals . . .

Thanks to hospital technology, the individual and the population simultaneously present themselves as objects of knowledge and medical intervention. (DE229.1, 151, trans. mod.)

For Foucault in 1974, populations are essentially “a great number of individuals,” an epiphenomenon or byproduct, as it were, of disciplinary power’s effects upon discrete individuals, that emerges simultaneously with them. He still holds this view in 1975’s Discipline and Punish, defining a population as simply constituted by a “multiplicity of bodies and forces” (1975ET, 77–78). In 1976, as we saw in the preceding section, Foucault’s understanding has begun to evolve, and “populations” point to larger phenomena that are not simply or strictly reducible to “groups of individuals.” Finally, in the 1979 essay, “The Politics of Health in the Eighteenth Century,” he is describing populations as discrete entities requiring their own level of analysis (and rebutting his 1974 remark that populations were merely “great numbers” or a “sum” of individuals):

In appearance, it [population] is a question of nothing but the sum of individual phenomena; nevertheless, one observes there constants and variables which are proper to the population; and if one wishes to modify them, specific interventions are necessary. (DE257, III-730-731; DE257.1, 117)

As the tensions we saw in his Collège de France course and La volonté de savoir illustrate, 1976 was a critical year in Foucault’s evolving understanding of populations. A key moment in this transformation is indicated in his review of a biological text by Jacques Ruffié, De la biologie à la culture [From Biology to Culture]. Foucault takes a number of insights from Ruffié, most importantly his understanding of populations not simply as collections of individuals but rather statistical entities, “ensembles of variations” that “are unceasingly formed and dissolved” and that can function independently of the individuals they encompass (DE179, III-97; DE179.1, 129). Importantly, especially given his remarks about racism in the final lecture of his Collège de France course, he also draws from Ruffié’s work for his understanding of “race” as a concept to be challenged and rejected—especially as it is employed by states as a mechanism of control:

Hemato-typology now authorizes the dissolution of the idea of human race. With a whole series of supporting evidence from prehistory and paleontology, it can be established that there never were “races” in the human species; but at the very most a process of “raciation,” tied to the existence of certain isolated groups. This process, far from having succeeded, reversed itself beginning with the Neolithic era and, through the effect of migrations, displacements, exchanges, and diverse interminglings, it was succeeded by a constant “deraciation.” We must conceive of a humanity not as juxtaposed races, but as “clouds” of populations that are interwoven together and combine a genetic inheritance that is all the more valuable the more its polymorphism is accentuated. (DE179, III-96; DE179.1, 129)

“Race” is a kind of fiction that may come to seem permanent, perhaps even a priori, but in fact “race” and “races” are contingent historical constructs that can be dissolved through other historical shifts. Foucault finds Ruffié’s text valuable because, as he closes his review,

one sees very clearly formulated here the questions of a “bio-history” that would no longer be the unitary and mythological history of the human species across time, and a “bio-politics” which would not be one of divisions, self-preservation, and hierarchies but of communication and polymorphism. (DE179, III-97; DE179.1, 129)2

For Foucault, “bio-politics” and biopower are addressed not to particular individuals but to larger shifts in human populations—and, in contrast to the disciplinary techniques of division and hierarchy, they open a space for “communication and polymorphism” (important notions for a Foucauldian ethics).

Perhaps the best illustration of how Foucault’s rethinking of the concept of population allows his larger analysis to advance can be seen in the changes he made in two versions of an essay published in 1976 and 1979—initially as he is just discovering the inadequacies of his current analysis, and later, after he has reconstructed his theory. The essay is “The Politics of Health in the Eighteenth Century.”

The bibliographical history of this text is tricky. Both versions were published in a monograph, which itself has the same title in both editions—Les machines à guérir [Curing Machines]. The 1976 edition (DE168) was published in Paris by the Institut de l’Environnement, however, whereas the 1979 edition (DE257) was published in Brussels by Pierre Mardaga. (An English translation of the 1976 version was published in the early collection Power/Knowledge [DE168.1]; the 1979 version did not appear in English until 2014 [DE257.1].) Despite superficial similarities, the second version has been substantially rewritten. Though they are approximately the same length, and the second halves of the two essays are virtually identical (one paragraph from the 1976 version is omitted in 1979), the essays’ first halves, however, differ in significant ways.³ Foucault has rewritten the first half of this essay in light of his insights in the intervening years about populations as macrophenomena to which correspond nondisciplinary forms of power—and the essay is, as a result, frankly much clearer.

His aim in this essay is to articulate the emerging interconnections between “private” medicine (clinical practice between doctors and patients) and state-supported “assistance” (public health and welfare programs). Already in 1976, Foucault recognizes that these are distinct but related: that version of the essay begins by noting that “what the eighteenth century shows, in any case, is a double-sided process. The development of a medical market in the form of private clienteles . . . cannot be divorced from the concurrent organization of a politics of health” (DE168.1, 166). But his attempts to articulate the framework in which both obtain are strained. His “solution”—inelegant and unclear—is to speak of what he calls “noso-politics,” literally medical- or hospital-politics. “The most striking trait of this noso-politics . . . no doubt consists in the displacement of health problems relative to problems of assistance” (DE168.1, 168). This neologism does little to help us understand how these two trends are interwoven; nevertheless, it is the organizing theme of the 1976 version. He does speak of “populations” in this version but still reads them as the cumulative effects of multiple applications of power upon individual bodies rather than as autonomous (DE168.1, 171–172).

Whereas Foucault was struggling (unsatisfactorily) to use “noso-politics” as a kind of framing concept in 1976, this term has entirely disappeared from the 1979 version. As I noted above, Foucault now recognizes populations as discrete entities (not merely Benthamite “sums of individuals”) upon which discrete forms of power may operate. As this concept allows him to analyze different kinds of power relations at different levels, Foucault’s problem of articulating the interrelationships between these levels (the clinical and the public) becomes much more straightforward—as is reflected in his prose and his abandonment of awkward neologisms. Thus, the later version begins by asserting that

what is important is rather the specific manner in which, at a given moment and in a specified society, the individual interaction between the doctor and the sick person is articulated upon the collective intervention with respect to illness in general or to this sick person in particular (DE257, III-726; DE257.1, 114).

This clarity was facilitated by his realization that populations function at a macro level in a way analogous (but not reducible) to individuals at the micro level:

But an element appeared at the center of this materiality, an element whose importance unceasingly asserted itself and grew in the seventeenth and eighteenth centuries: it was the population, understood in the already traditional sense of the number of inhabitants in proportion to the habitable area, but equally in the sense of an ensemble of individuals having between them relations of coexistence and constituting therefore a specific reality. (DE257, III-730; DE257.1, 117)

Populations constitute “a specific reality” “at the center of this materiality,” which is to be managed through other, nondisciplinary techniques—“collective interventions” that he terms “the functions of police.” And so the “politics of health” in the eighteenth century is characterized by the interweaving of two related but distinct forms of power—disciplinary power over (sick) individuals and a macropower over populations, whose aim was not to treat but prevent illnesses.

As the changes in this essay illustrate, his shift away from an exclusive emphasis upon individualizing microfunctions of power toward the dynamics of populations and power at a larger scale will enable Foucault to analyze better social processes such as normalization and to spell out a new aspect of power: biopower. Indeed, my explication of the latter risks jumping too far ahead, as it presupposes concepts that Foucault had developed in more detail in the intervening years. Articulating the concepts that facilitate this shift in his understanding of power was the primary task in his 1978 Collège de France course, Security, Territory, Population.

APPARATUSES OF SECURITY

After a sabbatical year in 1977, Foucault’s 1978 course at the Collège de France gave him the opportunity to articulate the major alteration to the theory of power that he had been working out over the past several years. Here he articulates the irreducibility of macrophenomena to microphenomena, even as he begins to show some of the ways in which they are interwoven. This analysis builds upon Foucault’s recognition of “populations” as irreducible macrophenomena; it also brings a renewed subtlety to his analysis of the processes of normalization. He articulates what we could characterize as a “structural” analysis of biopower, explicitly contrasting its mechanisms and orientation with those of disciplinary power, describing this macropower as one oriented toward “security,” in particular, the security and the preservation of the modern state. But this task constitutes only the first of three central, interconnected foci that emerge in Security, Territory, Population. Further, Foucault is able to trace the emergence of biopower back to much earlier, premodern forms of power that developed in the eastern Judeo-Christian tradition—this constitutes the second focal theme; he calls this premodern form “pastoral power.” (This connection also shows how the application and utility of Foucault’s theory of power is not restricted to modern Western societies and cultures.) The third theme of the course—to which I will devote less attention—returns to the original concern with states’ security, identifying two particular technologies or mechanisms by which states were able to exercise this macropower: what Foucault terms “raison d’état” and “police.” (The latter of which, in particular, constitutes an interface between biopolitical and disciplinary modes of power.) With the modifications and reelaboration that Foucault accomplishes here, his theory of power becomes much more complex—and, as a corollary, so do the landscape of and possibilities for resistance, self-articulation, and ethics within these multifaceted relations of power. Reconceived in this way, the problem of “power” becomes instead the “problem of government,” or (as he will call it) “governmentality.” This problem of governmentality explicitly thematizes certain ethical dimensions—it thus constitutes a background, a stable social analysis, within which a Foucauldian ethics can begin to be articulated.

Indeed, the very first remark Foucault makes to inaugurate the course is that “This year I would like to begin studying something that I have called, somewhat vaguely, bio-power” (CdF78ET, 1). He then defines it in two ways. First, biopower is “the set of mechanisms through which the basic biological features of the human species became the object of a political strategy, of a general strategy of power” (CdF78ET, 1); it can also be understood as “how, starting from the eighteenth century, modern Western societies took on board the fundamental biological fact that human beings are a species” (CdF78ET, 1). These two definitions, while largely overlapping, are not identical. The second includes both a temporal specification—modernity, especially since the eighteenth century (which fits with the temporal focus of his 1976/1979 essay on the politics of health)—and a geographical specification—Western, that is, European, societies—that are not present in the first. Indeed, the origins of biopower, he will argue, come from beyond both of these specifications; nevertheless, modern Western society is of particular importance because it constitutes our situation, the circumstances within which we must act.

Foucault next defines “security,” or the “apparatuses of security,” which represent a modulation or shift effected upon certain disciplinary techniques, so that they now function in a different way, upon a different object. He illustrates this with two examples: law and medicine, in each of which we can see several modulations. Law, for example, may begin with a simple prohibition such as “Do not steal” and its accompanying punishment. He here calls this the “legal or juridical mechanism” (CdF78ET, 5). It can then be modulated by disciplinary techniques: it is still the same law, “but now everything is framed by . . . a series of supervisions, checks, inspections, and varied controls that, even before the thief has stolen, make it possible to identify whether or not he is going to steal, and so on” (CdF78ET, 4). Finally, there is a biopolitical modulation, “based on the same matrix, with the same penal law, the same punishments, and the same type of framework of surveillance on one side and correction on the other, but now [all this] will be governed by” (CdF78ET, 4) a different organizing principle: “The general question basically will be how to keep a type of criminality, theft for instance, within socially and economically acceptable limits and around an average that will be considered as optimal for a given social functioning” (CdF78ET, 5). Medicine, too, offers an illustration of these different modulations. Biopolitical medicine is oriented not toward particular sick individuals but rather “the statistical effects on the population in general” of medical practices such as inoculation and vaccination (CdF78ET, 10). “In short, it will no longer be the problem of exclusion, as with leprosy, or of quarantine, as with the plague [a disciplinary medicine], but of epidemics and the medical campaigns that try to halt epidemic or endemic phenomena” (CdF78ET, 10). These “modulations,” Foucault immediately notes, however, do not displace or eliminate the earlier forms; rather they are overlaid upon each other: “there is not a succession of law, then discipline, then security, but that security is a way of making the old armatures of law and discipline function in addition to the specific mechanisms of security” (CdF78ET, 10).

Having thus illustrated how “security” or biopower functions simultaneously with, alongside, disciplinary power, Foucault devotes the remainder of the first three lectures of this course to articulating four “general features of these apparatuses of security” (CdF78ET, 11), in each case showing how biopower is distinct from disciplinary power. These four characteristics concern the way these different kinds of power deal with space (or geography), their respective treatments of uncertainty, “the form of normalization specific to security which seems to me to be different from the disciplinary type of normalization,” and “the correlation between the technique of security and population as both the object and subject of these mechanisms of security” (CdF78ET, 11). Of these four—space, uncertainty, normalization, and population—the latter two are clearly the most important. For Foucault had closed Discipline and Punish by noting that it “must serve as a historical background to various studies of the power of normalization” (1976ET, 308), and as I have argued, Foucault’s reconceptualization of “population” had compelled and facilitated his analysis of biopower as a distinct type of power. Speaking of the historical transformations that he is tracing in governmental technologies and practices, but in language that is equally applicable to the development of his own understanding of power, Foucault observes that:

it is thanks to the perception of the specific problems of the population, and thanks to the isolation of the level of reality that we call the economy, that it was possible to think, reflect, and calculate the problem of government outside the juridical framework of sovereignty [or discipline]. (CdF78ET, 104)

We can briefly discuss the first feature (space), and then will turn to a more extended presentation of normalization and populations—the heart of the contrast—before finally returning to the second feature (uncertainty, which is importantly linked to possibilities for freedom). In constructing a panopticon, a prison, a barracks, or any architecture designed to facilitate observation and hierarchy, “discipline works in an empty, artificial space that is to be completely constructed” (CdF78ET, 19). By contrast,

Security will rely on a number of material givens . . . this given will not be reconstructed to arrive at a point of perfection, as in a disciplinary town. It is simply a matter of maximizing the positive elements, for which one provides the best possible circulation, and of minimizing what is risky and inconvenient, like theft and disease, while knowing that they will never be completely suppressed. [Thus,] one works on probabilities. (CdF78ET, 19)

Thus, biopower does not aim at a complete control but begins with “givens,” historical realities that can shift and change independently and thus can only be maximized or minimized, never entirely eliminated or perfected. It cannot control every individual in a population; instead it aims at probabilities rather than totalities. As a result, biopower “works on the future, that is to say, the town will not be conceived or planned according to a static perception that would ensure the perfection of the function there and then, but will open onto a future that is not exactly controllable, not exactly measured or measurable” (CdF78ET, 20). In this contrasting approach to problems of space, we see the emerging importance of both norms—those not completely controllable but probabilistically predictable patterns—and population—the set of human beings taken not as discrete individuals but as a collectivity.

The lecture of 25 January 1978—the third lecture of the course—is devoted to these two topics. First, Foucault corrects and clarifies his own understanding of norms by introducing a distinction between what he terms “normation” and “normalization”—a distinction corresponding to micro- and macro forms, though the micro form, what he now terms “normation,” is what he had earlier (in Discipline and Punish) called “normalization.” The second task is to begin to excavate the significance of “population”—a concept that as we’ve seen is centrally organizing for his understanding of these macro forms of power. The purpose behind both of these tasks is, he notes at the very beginning of the lecture, “to emphasize the opposition, or at any rate the distinction, between security and discipline” (CdF78ET, 55). (We can hear in this softening, from “opposition” to “distinction,” the movement toward complementing rather than replacing the original analysis of disciplinary power.)

So, strictly speaking, what is this distinction between “normation” (Foucault’s own neologism, which he himself describes as a “barbaric word” [CdF78ET, 57]) and normalization? Discipline “normalizes” (in the loose sense) in a five-step process. First, it analyzes and isolates discrete “individuals, places, times, movements, actions, and operations” (CdF78ET, 56). Next, these discrete analytical units (individuals, behaviors, etc.) are classified “according to definite objectives” (CdF78ET, 57). Then optimal sequences, links, and coordinations are established between them—in other words, the individuals are sorted and hierarchized in light of their classifications. Fourth, “discipline fixes the processes of progressive training and permanent control” (CdF78ET, 57) in light of the given objectives. Finally, “on the basis of this, it establishes the division between those considered unsuitable or incapable and the others. That is to say, on this basis it divides the normal from the abnormal” (CdF78ET, 57). This analysis is itself very schematic, but Foucault has already done the detail work: recall Part 3 of Discipline and Punish, and you will find a rich empirical analysis of this process in armies, in prisons, and in schools. But, as Foucault now observes,

Disciplinary normalization consists first of all in positing a model, an optimal model that is constructed in terms of a certain result [step 2 of the five he has just delineated], and the operation of disciplinary normalization consists in trying to get people, movements, and actions to conform to this model, the normal being precisely that which can conform to this norm. (CdF78ET, 57)

The force of this insight is that disciplinary power presupposes something beyond its own techniques, something that gives content to its distinctions—something that Foucault here refers to as a “model.” This model determines the “definite objectives” as well as the optimal sequences and hierarchies. But the disciplinary micropower itself does not and cannot determine what that model is; it merely operates with the values (or variables, if you’d like to think about this mathematically) that are provided by that model. “In other words,” Foucault notes, “it is not the normal and the abnormal that is fundamental and primary in disciplinary normalization, it is the norm” (CdF78ET, 57). The norm, then, is the model that is presupposed by disciplinary normalization, and so:

Due to the primacy of the norm in relation to the normal, to the fact that disciplinary normalization goes from the norm to the final division between the normal and the abnormal, I would rather say that what is involved in disciplinary techniques is a normation rather than normalization. (CdF78ET, 57)

How, then, are these norms—norms that are given externally and prior to the operation of disciplinary apparatuses—determined? The identification and emergence of norms is the product of a macrorelation, and Foucault finds this macrorelation at work in the emerging medical practices of vaccination and inoculation. This macroprocess involves several elements. First, there is the “case,”

which is not the individual case but a way of individualizing the collective phenomenon of the disease, or of collectivizing the phenomena, integrating individual phenomena within a collective field, but in the form of quantification and of the rational and identifiable. (CdF78ET, 60)

A case can be identified only if individuals are considered not as discrete individuals but as tokens of a type within a larger field. Along with “cases” come the elements of “risk” and “danger”: variations in individual circumstances (all children in France, or those in towns compared those in the country, or adults compared to children, etc.) will account for a greater or lesser risk of contracting, for example, smallpox. The quantitative calculation of these risks

shows straightaway that risks are not the same for all individuals, all ages, or in every condition, place or milieu. There are therefore differential risks that reveal, as it were, zones of higher risk and, on the other hand, zones of less or lower risk. This means that one can thus identify what is dangerous. (CdF78ET, 61)

We are not speaking of discrete individuals but of groups, patterns, and populations when we speak of these zones of higher or lower risk. This macroapparatus “is not the division between those who are sick and those who are not” (CdF78ET, 62), which would be the disciplinary technique. Rather,

It takes all who are sick and all who are not as a whole, that is to say, in short, the population, and it identifies . . . the normal expectation in the population of being affected by the disease and of death linked to the disease. . . . Thus we get the idea of a “normal” morbidity or mortality. (CdF78ET, 62)

It then subdivides the population as a whole “to disengage different normalities in relation to each other” (CdF78ET, 63). “It is at this level of the interplay of differential normalities [all established in terms of populations, not individuals] . . . that . . . the medicine of prevention will act” (CdF78ET, 63). This analysis at the level of populations gives us

a plotting of the normal and the abnormal, of different curves of normality, and the operation of normalization consists in establishing an interplay between these different distributions of normality and [in] acting to bring the most unfavorable in line with the more favorable. . . . These distributions will serve as the norm. (CdF78ET, 63)

Norms emerge, then, through macroanalyses, macrorelations of power and knowledge that cannot be explained in disciplinary terms. Foucault is unambiguous on this point: “The government of populations is, I think, completely different from the exercise of sovereignty over the fine grain of individual behaviors. It seems to me that we have two completely different systems of power” (CdF78ET, 66).

(Of course, while he was unambiguous in the assertion that these are two “completely different” systems of power, he did speak here of an “exercise of sovereignty” when he should have said “of discipline.” That he is speaking of discipline is clear not only from the phrasing here of “the fine grain of individual behaviors” but because he is referring to the Panopticon, “a power that takes the form of an exhaustive surveillance of individuals” and is the paradigmatic architecture of disciplinary power.)

Foucault has now grasped that the macro power relations that determine norms are not reducible to disciplinary microtechniques. This macroprocess of determining norms is what Foucault now terms “normalization in the strict sense” (CdF78ET, 63). We should also note, however, that these two processes (disciplinary normation and macronormalization) are not isolated phenomena and can reciprocally influence each other. How the macronorms motivate microdiscipline is, I hope, already clear. But the influence can flow in the opposite direction, too—constituting a sort of feedback loop. Micropractices can, over time, produce new norms, or at least produce individuals that, when considered as part of a collective whole, shift the values of the norm in new directions. (Thus are constituted the “material givens” with which apparatuses of security must begin.) Norms inform and frame discipline’s classification of the normal and abnormal, but the new, altered individuals produced by these disciplinary practices can also shift the values of the norms. (This is, in effect, a process of evolution.)

And so, if his earlier analyses showed that an individual is the locus of, and in part constituted by, disciplinary micro forms of power, then we can now understand a population as the analogous object of the normalizing (in the strict sense) macro forms of power. For mechanisms of security, like preventative medicine, the “pertinent level of government is not the actual totality of the subjects in every single detail but the population with its specific phenomena and processes” (CdF78ET, 66). Populations come to be understood as “natural” and as a “set of processes to be managed” “not from the standpoint of the juridical-political notion of the subject, but as a sort of technical-political object of management and government” (CdF78ET, 70). This “natural phenomenon” is marked by several characteristics: it “is not the simple sum of individuals inhabiting a territory” (CdF78ET, 70); it “is not a primary datum; it is dependent upon a series of variables” (CdF78ET, 70) such as climate, commerce, etc.; and thus, “the relation between the population and sovereign cannot simply be one of obedience or the refusal of obedience, or obedience or revolt” (CdF78ET, 71). Nevertheless, a population “is constantly accessible to agents and techniques of transformation, on condition that these agents and techniques are at once enlightened, reflected, analytical, calculated, and calculating” (CdF78ET, 71). A population thus constitutes

a set of elements that, on one side, are immersed within the general regime of living beings and that, on another side, offer a surface on which authoritarian, but reflected and calculated transformations can get a hold.

. . . we have here a whole field of new realities in the sense that they are the pertinent elements for mechanisms of power. (CdF78ET, 75)

We can hear in this evocation of “the general regime of living beings” the perspective that led Foucault to define biopower initially as a power that targets humans as a species (CdF78ET, 1). We can hear, too, in the description of population as an object of management and rational, calculated government the origins of his term “governmentality.”

The discovery or recognition of these macro forms of power, and the central importance of this notion of population as the object of these macro forms, is a shift of profound significance in Foucault’s understanding of power relations. (Hence, perhaps, the proliferation of terms for this new macro form of power.) However, there are important continuities. Structurally or theoretically, the macro forms of biopower and disciplinary micropower are, as Foucault understands them, quite similar: they are both relational; neither is understood as a property to be possessed; they are both productive, not prohibitive in function; they both constitute and are constituted by knowledges; and both in their very relational logic necessarily presuppose a possibility for resistance—indeed, freedoms. It is in light of these profound commonalities that I think we can best understand biopower as a complement to, an enrichment of, Foucault’s analyses of discipline.

Foucault’s closing remarks in this third lecture offer some important suggestions about how this reelaborated analysis of power opens new possibilities for Foucault’s ethical thinking. He ends by noting that

Hence the theme of man, and the “human sciences” that analyze him as a living being, working individual, and speaking subject, should be understood on the basis of the emergence of population as the correlate of power and the object of knowledge. (CdF78ET, 79)

This “theme of man” is, of course, a famous trope that runs throughout Foucault’s work: In 1966 he closed The Order of Things with the quasi-hopeful/quasi-despairing claim that “one can certainly wager that man would be erased, like a face drawn in sand at the edge of the sea” (1966ET, 387); Discipline and Punish opened with the claim that “the man described for us, whom we are invited to free, is already in himself the effect of a subjection much more profound than himself” (1975ET, 30).

Discipline and Punish, which grasped disciplinary power in its details, mistakenly hypothesized that all power relations are reducible to microrelations, thus leading to its seemingly bleak portrayal of a “carceral society.” “This process that constitutes delinquency as an object of knowledge is one with the political operation that dissociates illegalities and isolates delinquency from them. The prison is the hinge of these two mechanisms” (1975ET, 277). But as Foucault can now recognize, his earlier analyses conflated disciplinary microprocesses (the former process) with macroprocesses (the latter political operation, which we can now understand in the terms he has introduced in 1978). If there is no escape from the ever-tightening net of discipline, if we are the effects of “a subjection much more profound” than ourselves, then resistance may be ultimately futile. But if these distinct forms of power are not “one with” each other but rather sometimes collaborative and sometimes opposed forces, then for those who find themselves enmeshed in these power relations (in other words, all of us), the possibilities for freedom have been greatly enlarged and multiplied. The power relations in which our subjectivities are constituted are not monolithic. The field has opened up. A framework is opening in which Foucault (and we) will be able to reexamine our subjectivities—not merely as passive products of “subjectivation” but also as active, “self-interpreting animals” (to borrow a phrase from Charles Taylor),4 engaged in what Foucault in 1984 will call “practices of liberty” (cf. DE356). This insight will guide the trajectory of his Collège de France courses from 1980 until his death.

The notion of freedom raised here also brings us back to the second feature that distinguishes biopower—its treatment of uncertainty. In the second lecture of the course (18 January 1978), Foucault explores this theme through the example of scarcity—food shortages and how governments could attempt to prevent or ameliorate them. A disciplinary approach would attempt to regulate finely all aspects of production—through quotas, import/export regulations, and price controls, for example—to ensure that there is always an adequate supply of grain. But this cannot account for the uncontrollable variability of “material givens” such as climate (drought, flood, or ideal conditions), harvest (a shortage or a surplus—both of which play havoc with the supply mechanisms), etc. But these techniques, frankly, typically failed to prevent shortages; they also often contributed to unrest and revolts. A biopolitical approach, “by contrast . . . ‘lets things happen’” (CdF78ET, 45). Rather than imposing infinitesimal and tight regulations, it is oriented toward solving the problem at the level of the collectivity or population and allows “market mechanisms” (CdF78ET, 40) to regulate themselves, solving the problems of excess supply or demand through price variations, increasing or decreasing exports, or planting of larger or smaller crops. Thus, the “problem” of scarcity dissolves itself and becomes a chimera:

It is a chimera when, in fact, people conduct themselves properly, that is to say when some accept to endure scarcity-dearness, others sell their wheat at the right moment, that is to say very soon, and when exporters send their product when prices begin to rise. This is all very well and we have here, I don’t say the good elements of the population, but behavior such that every individual functions well as a member, as an element of the thing we want to manage in the best way possible, namely the population. (CdF78ET, 43)

Note that this mode of power, too, attempts to impose certain conceptions of “conducting oneself properly”—but this proper conduct will be geared toward the ends of the population as a whole, not the discrete individual. Some will do without; others will raise or lower their prices, release more grain or alternatively export grain to reduce domestic supply: the integration of all these behaviors serve to regulate the supply of grain for the population as a whole. The mechanisms of security are not only markedly different from the mechanisms of discipline; in this case they actually work against each other.

The laissez-faire approach of security—accepting the uncertainty of given reality—“this fundamental principle that political technique must never get away from the interplay of reality with itself is profoundly linked to the general principle of what is called liberalism” (CdF78ET, 48).5 It thus brings to the fore what Foucault calls “this problem of freedom,” with which he closes this lecture. First, “it cannot be false . . . that this ideology of freedom really was one of the conditions of development of modern or, if you like, capitalist forms of the economy” (CdF78ET, 48). (Thus, an ideology of freedom is a kind of “condition of possibility” for a capitalist economy.) But second, this leads Foucault to make a significant revision to the portrayal given in Discipline and Punish.

I said somewhere6 that we could not understand the establishment of liberal ideologies and a liberal politics in the eighteenth century without keeping in mind that the same eighteenth century, which made such a strong demand for freedoms, had all the same ballasted these freedoms with a disciplinary technique that, taking children, soldiers, and workers where they were, considerably restricted freedom and provided, as it were, guarantees for this freedom. Well, I think I was wrong. I was not completely wrong, of course, but, in short, it was not exactly this. (CdF78ET, 48)

Whereas (following his Hobbesian hypothesis) he had earlier read freedom as a “mirage” produced through disciplinary normation of individuals, he now understands freedom in a wider context—it is not merely an epiphenomenon of discipline and control. But it remains fundamentally conjoined with power relations—the macrorelations of biopower as much as if not more than the microrelations of discipline: “this freedom, both ideology and technique of government, should in fact be understood within the mutations and transformations of technologies of power. More precisely and particularly, freedom is nothing else but the correlative of the deployment of apparatuses of security” (CdF78ET, 48). Of course, we should recognize this language of “correlates”—the soul, he had said, is “the present correlative of a certain technology of power over the body” (1975ET, 29)—but two very important shifts are being made here. First, freedom is primarily correlated with the macrotechniques of apparatuses of security, not discipline. But also, and for our purposes more importantly, “an apparatus of security . . . cannot operate well except on condition that it is given freedom . . . the possibility of movement, change of place, and processes of circulation of both people and things” (CdF78ET, 48–49). Freedom is a correlate of power relations, but those very power relations cannot function without that very freedom—freedom is also a condition for the possibility of these very power relations.

Let me close this discussion with an illustration from the eighteenth century—like my discussion of Stampp on slavery, this is an example that Foucault himself does not cite—a long and dense passage from the opening of Immanuel Kant’s 1784 “Idea for a Universal History with a Cosmopolitan Intent”:

History—which concerns itself with providing a narrative of these appearances, regardless of how deeply hidden their causes may be—allows us to hope that if we examine the play of the human will’s freedom in the large, we can discover its course to conform to rules as well as to hope that what strikes us as complicated and unpredictable in the single individual may in the history of the entire species be discovered to be the steady progress and slow development of its original capacities. Since the free wills of men seem to have so great an influence on marriage, the births consequent to it, and death, it appears that they are not subject to any rule by which one can in advance determine their number; and yet the annual charts that large countries make of them show that they occur in conformity with natural laws as invariable as those [governing] the unpredictable weather. . . . Individual men and even entire peoples give little thought to the fact that while each according to his own ways pursues his own end—often at cross purposes with each other—they unconsciously proceed toward an unknown natural end, as if following a guiding thread; and they work to promote an end they would set little store by, even if they were aware of it. (Kant 1784b, 29; Ak. 8:17)

Kant here articulates the interplay between individuals and populations that Foucault has been analyzing; Kant also shows how norms that seem to “obey natural laws” or “follow a guiding thread” emerge even through the interplay of free individuals. But Kant also takes this uniformity amid chaos as a reason for various kinds of hope—for Kant, hope for the progress of the human species and for our ability to discover the moral law within ourselves. Perhaps Foucault would hope for other things. But the emergence of this hope out of the interplay of freedom and norms shows, I think, a genuine alternative to the bleak and despairing outlook that Discipline and Punish could (from mistaken premises, as Foucault notes here) foster.

PASTORAL POWER AS THE “ART OF GOVERNING”

Foucault’s presentation of “security” or “biopower” in these first three lectures allow him to bring into focus a new problem—the modern problem of justifying the state’s claim to sovereignty, after the displacement of “sovereign power”—which emerged in discourses about “the art of government.” Indeed, he suggests, “the unblocking of the art of government was linked to the emergence of the problem of population” (CdF78ET, 103–104). “In fact,” he notes, “we have a triangle: sovereignty, discipline, and governmental management, which has population as its main target and apparatuses of security as its essential mechanism” (CdF78ET, 107–108). And at the pinnacle of this triangle are the arts of government, so that “from the eighteenth century, these three movements—government, population, political economy—form a solid series that has certainly not been dismantled even today” (CdF78ET, 108).7 This allows Foucault to specify the “challenge” or the “stakes” of these lectures: “Can we talk of something like a ‘governmentality’ that would be to the state what techniques of segregation were to psychiatry, what techniques of discipline were to the penal system, and what biopolitics was to medical institutions?” (CdF78ET, 120).

To answer this question—to assess what constitutes an “art of governing”—Foucault first attempts to trace this notion’s genealogy. “I will now try to show you how this governmentality was born from the archaic model of the Christian pastorate and, second, by drawing support from a new diplomatic-military model, or rather, technique, and finally, third . . . thanks to a set of very specific instruments . . . which is called . . . police” (CdF78ET, 110). These latter two elements are less important for my discussion; instead I will focus on its origins in the Christian pastoral and on Foucault’s understanding and use of the notion of “governmentality.”

He begins by examining the meaning of the verb “to govern,” looking not at its contemporary “political, rigourous statist meaning” (CdF78ET, 120) but its earlier, variable meanings from the thirteenth through fifteenth centuries. There are a number of material senses: first, a very simple, physical meaning, “to direct, move forward, or even to move forward oneself on a track, a road” (CdF78ET, 121). There is a broader sense of “supporting by providing means of subsistence” (CdF78ET, 121) to one’s family or to a city; finally, this second sense can be abstracted even farther, to encompass the sources of one’s subsistence, such that a town “is governed by” its principal industry. Beyond these material meanings, however, there are also a number of moral senses to this term. At its core, the moral element of “governing” involves the guidance or direction of someone’s conduct—in a spiritual sense of “the government of souls” or in a more secular sense of “imposing a regimen.” One could guide another’s conduct, as doctors direct their patients, or one can guide oneself, as a patient “who imposes treatment on himself, governs himself” (CdF78ET, 121). “Government” could be a synonym for conduct (a misbehaving daughter would be described as of “bad government”); “governing” could mean a variety of relationships, from command and control to engaging in conversation with someone. It could also refer to having sexual relations with someone. Thus, from its beginning, the notion of governing and government combine material elements encompassing the power to bring about a certain end and moral elements encompassing how to exercise this power well or poorly, for the good or not. “One thing clearly emerges through all these meanings, which is that one never governs a state, a territory, or a political structure. Those whom one governs are people, individuals, or groups” (CdF78ET, 122).

This notion of governing people does not come from the Western (Greek or Roman) political traditions. Foucault engages in an extended reading of Plato’s Statesman to show how in this exemplar of the Western tradition such an understanding of governing is mentioned only to be quickly dismissed. Rather, this notion of government came to the West from Christianity in the Mediterranean East—from the practices of the pastorate, which offered the shepherd as the model of good governing:

The Christian Church coagulated all these themes [which Foucault had just listed, and I will discuss next] of pastoral power into precise mechanisms and definite institutions, it organized a pastoral power that was both specific and autonomous, it implanted its apparatuses within the Roman Empire, and at the heart of the Empire it organized a type of power that I think was unknown to any other civilization. (CdF78ET, 129–130)

This pastoral power is distinguished by four key characteristics. As Foucault enumerates them, it is exercised over a “multiplicity in movement” (CdF78ET, 125); second, it “is fundamentally a beneficent power” (CdF78ET, 126); its form is of “someone who keeps watch” (CdF78ET, 127); and, finally, it “is an individualizing power” (CdF78ET, 128). Each of these is quite significant; however, I will discuss the second trait only after I have discussed the other three.

First, pastoral power is exercised over “the flock in its movement from one place to another” (CdF78ET, 125). In contrast to Greek conceptions, in which gods are tied to a particular place or territory, a shepherd (and the Hebrew God) governs a group that is constantly in motion, constantly in search of new grazing. Thus, this power is not “exercised on the unity of a territory”(CdF78ET, 126) but rather over a group—the multiplicity, the flock as a whole, as it moves from pasture to pasture. This first characteristic anticipates the structure of biopower, which will be concerned with populations, groups of individuals, taken as a whole, populations that can fluctuate and are in constant motion.

Further, “the form it takes is not first of all the striking display of strength and superiority” (CdF78ET, 127), as, for example, would be the case in Greek gods’ manifestations or in the exercise of “sovereign power.” Rather, the shepherd “will keep watch over the flock and avoid the misfortune that may threaten the least of its members. He will see to it that things are best for each of the animals of his flock” (CdF78ET, 127). Rather than drawing others’ attention to the shepherd’s majesty and glory, pastoral power is exercised by directing the shepherd’s attention to its flock. He is concerned with the flock as a whole and with each individual member (even “the least”) of the flock, keeping watch over them to avoid dangers. Here we can recognize rudimentary elements of both biopower (again, in the concern for the entire flock, anticipating dangers to be avoided) and discipline (in its watchful attention to each individual sheep).

Thus, pastoral power is individualizing. The shepherd must take note of each individual sheep, not just the flock as a whole. He does so by counting the sheep, morning and night, and may even be compelled to put the rest of the flock at risk to rescue a single stray sheep. In its individualizing function, as well as through its mechanism of keeping watch, we can hear early elements of what will become disciplinary power.

Pastoral power is an ancestral form of what will be bifurcated into micro and macro forms, discipline and biopower. This tension is captured in what Foucault calls “the paradox of the shepherd,” which takes two forms. First,

the shepherd must keep his eye on all and on each, omnes et singulatim, which will be the great problem both of the techniques of power in Christian pastorship, and of the, let’s say, modern techniques of power deployed in the technologies of population I have spoken about. (CdF78ET, 128)

Further, just as the shepherd must be willing to sacrifice the entire flock to save even the least of the sheep, he must be willing to sacrifice himself to save the entire flock. This constitutes a “moral paradox”: “the sacrifice of one for all, and the sacrifice of all for one, which will be at the absolute heart of the Christian problematic of the pastorate” (CdF78ET, 129). And so the problematic revealed in this paradox—how to exercise pastoral power in the government of Christians’ souls—will become for Saint Gregory of Nazianzus in the fourth century C.E. “the ‘art of arts,’ the ‘science of sciences’” (CdF78ET, 151). This early precursor of “the art of governing” thus “was the art by which some people were taught the government of others, and others were taught to let themselves be governed by certain people” (CdF78ET, 151).

This paradox only arises—the question of whom to sacrifice only becomes an issue—because of the final (Foucault’s second) feature of pastoral power: “pastoral power is fundamentally beneficent power” (CdF78ET, 126). Foucault puts this simply and unambiguously: “Pastoral power is a power of care” (CdF78ET, 127). And this feature is fundamental: “pastoral power is, I think, entirely defined by its beneficence; its only raison d’être is doing good, and in order to do good. In fact the essential objective of pastoral power is the salvation of the flock” (CdF78ET, 126). This power exists on ethical grounds—it is exercised precisely and “entirely” for the good, the salvation, of those upon whom it is exercised. It thus constitutes “a duty, a task to be undertaken” for the one who exercises it, the shepherd (CdF78ET, 127). “The shepherd directs all his care towards others and never towards himself. This is precisely the difference between the good and the bad shepherd” (CdF78ET, 127–128). The prominence of this term “care” here is important for at least two reasons. First of all, this form of power, which is a precursor to both discipline and biopower, has an ethical dimension at its core. Second, as we shall see, some of Foucault’s own ethical trajectories will explore the possibilities for such caring within a framework marked by the later, more developed forms of power.

Foucault presented this analysis of pastoral power again in his October 1979 Tanner Lectures at Stanford University (DE291; in effect, these two lectures serve as a condensed version of the 1978 Collège de France course). Two key points emerge quite clearly in the Tanner Lectures rearticulation.

First, the theme of care is underscored: “The shepherd’s role is to ensure the salvation of his flock. . . . It’s a matter of constant, individualized, and final kindness” (DE291.4, 302). And so his “duty,” his “shepherdly kindness[,] is much closer to ‘devotedness’” (DE291.4, 302), which manifests in a constant concern to keep watch over the flock: “He pays attention to them all and scans each one of them. He’s got to know his flock as a whole, and in detail” (DE291.4, 303). In pastoral power, the ethics and the techniques of power are interwoven. And the final phrase here, “as a whole and in detail,” again reminds us that pastoral power is at the root of both biopolitical and disciplinary forms of modern power. It can even be seen in contemporary political struggles: “The well-known ‘welfare state problem’ . . . must be recognized for what it is: one of the extremely numerous reappearances of the tricky adjustment between political power wielded over legal subjects and pastoral power wielded over live individuals” (DE291.4, 307).

The second important point that Foucault brings out in the Tanner Lectures is pastoral power’s legacy in the construction of modern subjectivity. For in the history of the Christian Church, the original theme of pastoral power was to be rearticulated and reformulated in various techniques and institutions. Most important among these are the practices of self-examination and confession. The shepherd has a certain responsibility to “render an account—not only of each sheep, but of all their actions” (DE291.4, 308). So each member of the flock must confess its sins to the shepherd, so that he can, in his turn, render an account. Correspondingly, each sheep must be obedient to the shepherd, taking on a “personal submission to him” (DE291.4, 309). Taken together, these relations of responsibility and obedience serve to constitute “a peculiar type of knowledge” (DE291.4, 309) of each sheep’s particular needs—but also of each one’s sins. We can hear in this description an anticipation of what Foucault will call in Discipline and Punish “the modern ‘soul.’” Understanding this legacy—the constitution of the modern subject—will also guide Foucault’s continuing itinerary. The title of the 1982 Collège de France course, The Hermeneutics of the Subject, is explicitly devoted to the genealogy of this subjectivity; in fact, it is a central theme of many of his courses in the 1980s.

A particular example can be seen in another 1982 text, “The Subject and Power” (DE306). Returning to the questions of “the state” that had initially guided his 1978 Collège de France inquiry, Foucault again situates pastoral power at the root of these developments: “we can see the state as a modern matrix of individualization, or a new form of pastoral power” (DE306.4, 334). In enumerating how pastoral power is present in the state and in the forms of modern subjectivity, he concludes by evoking the twin poles of macro biopower and micro disciplinary power:

Finally, the multiplication of the aims and agents of pastoral power focused the development of knowledge of man around two roles: one, globalizing and quantitative, concerning the population; the other, analytical, concerning the individual. (DE306.4, 335)

Significantly, immediately following this observation, Foucault cites Immanuel Kant’s 1784 essay “What Is Enlightenment?” as “an analysis of both us and our present” (DE306.4, 335). Foucault continues that “the task of philosophy as a critical analysis of our world is something that is more and more important” (DE306.4, 336). And he notes that this has important ethical implications:

The conclusion would be then that the political, ethical, social, philosophical problem of our days is not to try to liberate the individual from the state, and from the state’s institutions, but to liberate us both from the state and from the type of individualization linked to the state. We have to promote new forms of subjectivity through the refusal of this kind of individuality that has been imposed on us for several centuries. (DE306.4, 336)

Freedom, liberation, is here identified as the task toward which philosophy ought to work. Indeed, Foucault closed his Tanner Lectures in noting that “Liberation can come only from attacking not just one of these two effects [individualization and totalitarianism] but political rationality’s very roots” (DE291.4:325). And as he noted in the 1978 course, “there has never been an anti-pastoral revolution” (CdF78ET, 150).8 Exploring what possibilities there could be for “new forms of subjectivity” will be one of Foucault’s tasks in the years to come—one of the trajectories of a Foucauldian ethics. But just as importantly, we can note that this freedom, the telos of philosophy, is also—as he has shown in the first part of the 1978 course—a condition for the possibility of the power relations from which we seek to be liberated. This is an important insight, to which we shall return.

GOVERNMENTALITY

Pastoral power, then, “is the prelude to this governmentality in two ways” (CdF78ET, 184). It establishes new types of relationships, and it gives rise to, constitutes, a specific sort of subject. And so, Foucault explains, “by this word ‘governmentality’ I mean three things” (CdF78ET, 108).

First, it encompasses

the ensemble formed by institutions, procedures, analyses, and reflections, calculations, and tactics that allow the exercise of this very specific, albeit very complex, power that has the population as its target, political economy as its major form of knowledge, and apparatuses of security as its essential technical instrument. (CdF78ET, 108)

As the last half of this passage makes explicit, Foucault means biopower, the macropower that he has been describing in the first lectures of the course. But if we were to consider only the first half of this passage (ending with “this very specific, albeit very complex, power”), we could rightly infer that Foucault was talking about modern power more generally—not only biopower but also disciplinary power—for they each are constituted through all of the elements and techniques he has adumbrated.

Second, “governmentality” means

the tendency, the line of force, that for a long time, and throughout the West, has constantly led towards the pre-eminence over all other types of power—sovereignty, discipline, and so on—of the type of power that we can call “government” and which has led to the development of a series of specific governmental apparatuses on the one hand, [and, on the other] to the development of a series of knowledges. (CdF78ET, 108)

We can note the same bivalence in this statement, as well. On the one hand, Foucault is clearly specifying biopower in this description and even suggests that it has become preeminent over other forms, including discipline. There is a sense in which this is correct: biopower as a specific technique emerged somewhat later than discipline and was able to “piggyback” on disciplinary modes in the articulation of its knowledges about populations. However, two points can be noted in response. First, Foucault has been quite clear (and correctly so) that biopower does not “displace” but rather “overlays” disciplinary power—the latter is not superseded; rather, they both simultaneously function in the same social field. Second, as Foucault’s discussion of pastoral power (which we have already seen but would actually be given in the lecture following this one) makes clear, discipline and biopower share a common ancestor in pastoral power. So the “tendency” or “line of force” that leads to the preeminence of a particular form of power might be better understood as referring not to biopower in its specificity but rather to the modern distribution of power, which is only adequately described when both forms are interwoven.

Third and finally, “governmentality” connotes “the result of the process by which the state of justice of the Middle Ages became the administrative state in the fifteenth and sixteenth centuries and was gradually ‘governmentalized’” (CdF78ET, 108–109). Again, in the strict sense, Foucault is referring to a state whose administrative apparatus is aimed at the population, that is, biopower. But in a broader sense, the administrative institutions and techniques that constitute the state are also represented by what Foucault calls “police,” and “police” techniques are disciplinary through and through. (Indeed, “police” represents the interface where disciplinary and biopolitical techniques are integrated for the management or government of the state.) Discipline is, just as much as biopower, the result of these administrative transformations of the state since the Middle Ages.

So with each of these specifications of “governmentality,” Foucault could be speaking exclusively of biopower, or he could be speaking of a framework of modern power that incorporates discipline and biopower into the constitution and management of states. As Michel Senellart notes, “the concept of ‘governmentality’ progressively shifts from a precise, historically determinate sense, to a more general and abstract meaning” (Senellart 2004, 387–388). While we do need to be aware of that “determinate” sense (the forms of raison d’état and police, which I mentioned earlier, are such historically specific developments), the more general meaning is in fact the more important.

Even by the end of these lectures, though the specific sense has not been entirely abandoned, “governmentality” comes to function as a sort of umbrella term that brings the different kinds of power (sovereign, disciplinary, and biopower) into a coherent frame; it also underscores, in the multiple senses of “governing,” the capacities or resources for ethics implicit within his analysis of power—resources that Foucault will continue to excavate for the remainder of his life, most explicitly in his final Collège de France courses. In the very last paragraph of the entire course, Foucault sums up its work:

All I wanted to do this year was a mere experiment of method in order to show how starting from the relatively local and microscopic analysis of those typical forms of power of the pastorate it was possible, without paradox or contradiction, to return to the general problems of the state, on condition precisely that we [do not make] the state [into] a transcendent reality whose history could be undertaken on the basis of itself. It must be possible to do the history of the state on the basis of men’s actual practice, on the basis of what they do and how they think . . . there is not a sort of break between the level of micro-power and the level of macro-power, and that talking about one [does not] exclude talking about the other. (CdF78ET, 358)

We must not take the state as a transcendent reality. On the contrary, we must talk about the micropowers and macropowers—disciplines and biopowers—together. As we move from the local to the global and back, “on the basis of what men do and how they think,” “governmentality,” and “the arts of governing” serve well as a framework and background for this critical work. Using this concept of “government” as a frame will, in fact, help Foucault clarify his own thinking as he turns from the analysis of power to questions of ethics within power. We can see such a schema being sketched in Foucault’s January 1984 interview, “The Ethics of the Concern for the Self as a Practice of Freedom”:

We must distinguish between power relations understood as strategic games between liberties—in which some try to control the conduct of others, who in turn try to avoid allowing their conduct to be controlled or try to control the conduct of others—and the states of domination that people ordinarily call “power.” And between the two, between games of power and states of domination, you have technologies of government—understood, of course, in a very broad sense that includes not only the way institutions are governed but also the way one governs one’s wife and children. (DE356.4, 299)

“Governing,” guiding the conduct of others, is an integral part of the games of power. “States of domination” are not “nation-states” but rather reified situations in which no resistance would be possible. The “technologies of government,” which Foucault notes here should be understood in the broadest sense, encompass both states (“governmentality” in the strict biopolitical sense) and microrelations such as families. (And lest we mistakenly accuse Foucault of antifeminism here, we should note that he drew these examples from his study of ancient Roman society, in which one did speak of governing one’s wife.) In 1982’s “The Subject and Power,” Foucault had spoken of liberation “both from the state and from the type of individualization linked to the state” (DE306.4, 336). Here, in 1984, he will note that

I believe that this problem [how power is to be used] must be framed in terms of rules of law, rational techniques of government and ethos, practices of the self and of freedom.

I believe that this is, in fact, the hinge point of ethical concerns and the political struggle for respect of rights, of critical thought against abusive techniques of government and research in ethics that seeks to ground individual freedom. (DE356.4, 299)

The framework with which we understand power—“governmentality,” encompassing government of the state, the economy, and the self—is a hinge point for ethical concerns as well as political struggle. This new frame will be indicated in the titles of several of his later Collège de France courses: from 1980’s “The Government of the Living” to “The Government of Self and Others” in 1983 and 1984. The 1980 title reflects a lingering emphasis upon biopolitical themes, concerned with the management of macrogroups through institutions such as the state (which had been his focus in 1979). It nevertheless continues two important and parallel shifts, initially indicated in Foucault’s 1978 discussion of pastoral power’s roots in early Christianity: first, a shift away from contemporary transformations to earlier, ancient practices—as can be seen in his opening image of the Roman emperor Septimus Severus—and second, a deemphasis of states in favor of individuals. As Foucault stressed at the beginning of the third lecture of the 1980 course, “This is what I’d like to study a little this year . . . the element of the first person, of the ‘I,’ of the ‘autos,’ of the ‘myself’ in what could be called alethurgy or veridiction of oneself” (CdF80ET, 48; trans. mod. based on the audio recordings; cf. CdF80, 48).

Thus, the structural elaboration of how power works is essentially complete, with the emergence of this notion of “governmentality.” We are given a clear and succinct (if dense) statement of Foucault’s “mature” view in the 1982 Collège de France course:

In other words, what I mean is this: if we take the question of power, of political power, situating it in the more general question of governmentality understood as a strategic field of power relations in the broadest and not merely political sense of the term, if we understand by governmentality a strategic field of power relations in their mobility, transformability, and reversibility, then I do not think that reflection on this notion of governmentality can avoid passing through, theoretically and practically, the element of a subject defined by the relationship of self to self . . . the analysis of governmentality—that is to say, of power as the ensemble of reversible relations—must refer to an ethics of the subject defined by the relationship of self to self. Quite simply, this means that in the type of analysis I have been trying to advance for some time you can see that power relations, governmentality, the government of the self and others, and the relationship of self to self constitute a chain, a thread, and I think it is around these notions, that we should be able to connect together the question of politics and the question of ethics. (CdF82ET, 252; trans. mod.; cf. CdF82, 241–242)

Inevitably, there will be details to be revised in this analysis of power and governmentality, but the core of the theory of power has been articulated, and (as these passages from 1982 and 1984 nicely encapsulate) this theory marks a key intersection between empirical and ethical concerns. We are now, therefore, in a position to identify the linkages that Foucault points to between his analyses of power and ethics.

ETHICS WITHIN POWER

We have achieved a number of insights. First, the mechanisms of modern (or any) power cannot be analyzed exclusively in terms of disciplinary micro practices. On the contrary, macropowers aimed at populations are coextensive with and irreducible to the disciplinary micropowers. Indeed, both forms of power share a common ancestor in premodern Christian pastoral power. Second, since there are multiple, irreducible forms or modes of power constantly at play with and against one another, the field of power relations has become exponentially more complex. This complexity actually means that there is more “freedom” within the system—multiple kinds of power that are overlaid upon each other create more possibilities for resistance against any particular form. Thus, the “Hobbesian hypothesis,” as I’ve termed it—that since all social relations are inflected by power we are trapped in an inescapable cage—is unsustainable. Consider again one of Kenneth Stampp’s observations about slavery:

In truth, no slave uprising ever had a chance of ultimate success. . . . The bondsmen themselves lacked the power to destroy the web of bondage. They would have to have the aid of free men inside or outside the South. (Stampp 1956, 140)

Even though the slaves themselves could not completely break their state of domination, there were other vectors of power, recourse to which would bring the institution of slavery to an end. (Of course, even the end of slavery was not the final end of all struggle but the inauguration of a new one that continues today in the civil rights movement and Black Lives Matter a century and a half later.) While we should be suspicious of power, we need not despair. On the contrary, the very sources of power and resistance can give us cause for hope in a struggle against domination.

This brings us to two absolutely critical and interwoven Foucauldian insights.

First, this analysis of power provides a framework and basis for ethical concerns. On the one hand, this means that we cannot engage in an ethical analysis without first understanding how power relations have shaped and conditioned our selves. On the other hand, this means that ethics must be articulated within a framework of power relations—ethics is not “outside of” or “beyond” power relations but rather emerges explicitly and necessarily within and through those very relations.

This point is intimated in the very opening remarks of Foucault’s 1978 Collège de France course—at the end of which he remarks, “Now I would like to begin the lectures” (CdF78ET, 4). In these prefatory remarks, he lays down a series of “indications of choice or statements of intent” (CdF78ET, 1) that delineate how he wants to situate his developing analysis of power. These general indications implicitly give the outline for a new shift that will reorient virtually all of Foucault’s subsequent thinking—a new explicit foregrounding of ethical questions that will displace power relations from their central place. These indications also articulate power relations’ newly emerging status as the context for, the inescapable frame or condition of possibility within which these ethical questions obtain.

Foucault’s five general indications are as follows: First, his ongoing analysis of power is “not in any way a general theory of what power is” (CdF78ET, 1) but rather “could and would only be at most a beginning of a theory” (CdF78ET, 2) of mechanisms of power’s functions. Of course, we have heard this claim that Foucault is not doing a “theory of power” before. But in this iteration, he qualifies and specifies the de rigeur disclaimer in interesting ways.

The second “indication of choice” is thus that power relations “are not ‘self-generating’ or ‘self-subsistant’; they are not founded upon themselves” (CdF78ET, 2). Power relations are not sui generis; on the contrary, they are interwoven with a wide variety of and different types of social relations. “Mechanisms of power are an intrinsic part of all these relations and, in a circular way, are both their effect and cause” (CdF78ET, 2).

This leads us to the third indication: because power relations are interwoven with other kinds of relations throughout society, “the analysis of these power relations may, of course, open out onto or initiate something like the overall analysis of a society” (CdF78ET, 2). In other words, this analysis can certainly begin (as it has) with an analysis of power, but it must then go beyond power relations to the other kinds of relations that constitute society. And, Foucault adds, this means that “what I am doing is something that concerns philosophy” (CdF78ET, 3). The analysis of power relations that he has developed can be linked and integrated with “history, sociology, or economics” (CdF78ET, 3), but for Foucault, the important connection that he wants to develop is to philosophy, “that is to say, the politics of truth, for I do not see many other definitions of the word ‘philosophy’ apart from this” (CdF78ET, 3). When Foucault speaks of “the politics of truth” he means adopting a certain kind of critical attitude—which is, as he will make clear in later years, the heart of an ethical project. Thus the connection to ethics, though only implicit, is nonetheless beginning to be articulated.

It becomes much clearer, if still tentative, in the final two indications. The fourth indication is that all “theoretical or analytical discourses” are “permeated or underpinned in one way or another by something like an imperative discourse” (CdF78ET, 3). This is, quite frankly, strong and surprising language: the analytical work that Foucault has been doing to articulate a theory of power, he tells us, is “permeated or underpinned” by imperative discourses—value claims. In other words, normativity, or some sort of ethical or moral discourse, is unavoidable. He immediately qualifies this statement, however, adding that such an imperative “seems to me, at present at any rate, to be no more than an aesthetic discourse that can only be based on choices of an aesthetic order” (CdF78ET, 3). We can hear in this remark the roots of Foucault’s later exploration of the possibilities available within a “merely” aesthetic approach, of “one’s self as a work of art,” for example. And we can recognize a larger pattern of argument being repeated here: suppose a most-limited-case scenario, and only when it is demonstrated to be inadequate can we safely posit a stronger claim—one that we wanted but were initially reluctant to embrace. (This was, after all, how the Hobbesian hypothesis was refuted.) For “the dimension of what is to be done can only appear within a field of real forces that cannot be created by a speaking subject alone” (CdF78ET, 3)—that is, ethical work can only be done within a field analyzable in terms of power relations.

This has direct implications for how Foucault will conceive his own work.

So, since there has to be an imperative, I would like the one underpinning the theoretical analysis we are attempting to be quite simply a conditional imperative of the kind: If you want to struggle, here are some key points, here are some lines of force, here are some constrictions and blockages. In other words, I would like these imperatives to be no more than tactical pointers. (CdF78ET, 3)

If Foucault’s own imperatives have the status of mere “tactical pointers,” then he will be able to resist efforts to make his word a sort of gospel—a role he consistently denied. “You don’t have to do (or think) this, but you might want to try X,” not “Here is what must be done.” There are many reasons for Foucault’s reluctance to make such proclamations; among the most important is his profound appreciation for the contingency of any particular situation and the rich and unpredictable results of resistance in “a field of real forces”—what one would recommend now could in fact make things worse, and other strategies might turn out to be more productive toward a given end. And so, as he will put it later, “My point is not that everything is bad, but that everything is dangerous” (DE326.4, 256). Yet even as Foucault marks his ethical norms as mere “conditional imperatives” and “tactical pointers,” he will be more willing to make them explicit. And they will thus pull him deeper into what he terms “the circle of struggle and truth, that is to say, precisely, of philosophical practice” (CdF78ET, 3), for as he later notes, “ethics is a practice” (DE341.1, 377).

This brings us to Foucault’s fifth and final indication: “this serious and fundamental relation between struggle and truth . . . becomes emaciated, and loses its meaning and effectiveness in polemics. . . . So in all of this I will therefore propose only one imperative, but it will be categorical and unconditional: Never engage in polemics” (CdF78ET, 3–4). Even as it claims to be “merely methodological,” this imperative, too, carries rich ethical implications. For as he will note in a 1983 interview, “polemics allows for no possibility of an equal discussion” (DE342.2, 112).

The polemicist, on the other hand, proceeds encased in privileges that he possesses in advance and will never agree to question. . . . For him, then, the game consists . . . of abolishing [an interlocutor] from any possible dialogue; and his final objective will be not to come as close as possible to a difficult truth but to bring about the triumph of the just cause he has been manifestly upholding from the beginning. (DE342.2, 112)

In explicit contrast to “polemics,” Foucault will distinguish his own work—indeed his “attitude”—as investigations into the politics of truth, that is, philosophy.

My attitude isn’t a result of the form of critique that claims to be a methodological examination in order to reject all possible solutions except for the one valid one [as the polemicist would do]. It is more on the order of “problematization”—which is to say, the development of a domain of acts, practices, and thoughts that seem to me to pose problems for politics. (DE342.2, 114)

So these indications, at the very opening of the 1978 course, delineate a movement and trajectory that brings the ethical into the field of power relations, in terms (especially the final term, “polemics”) that will be further clarified as late as 1983.

Though beginning from an analysis of power, Foucault explicitly indicates that such an analysis cannot stand alone and constitutes just a piece of a larger social analysis. That larger analysis has to be reciprocally grounded in a normative context—even if in 1978 what seems immediately possible in that context is merely aesthetic and not yet ethical. Nevertheless, this inescapable imperative discourse pulls us (and Foucault) toward the deeper, fundamental philosophical relations between struggle and truth, in other words, toward the ethical.

Of course, Foucault often recast earlier work in terms of his current preoccupations—we saw in Chapter 2 how in the 1974 Collège de France course he recast History of Madness as “really about power.” But the opening to the ethical through—beginning in—the analysis of power that Foucault introduces here (and his characterization of it as an “indication of choice” is revealing) is not exactly the same kind of move: it is much more speculative. He doesn’t yet know where this arc or trajectory will take him; he can only indicate its underlying or framing elements. Foucault is still focused on understanding power relations—and they will remain present, sometimes even central, in his later analyses (as is only appropriate, since ethics emerges within a framework of power relations)—but the dawn is rising on a new horizon.

Further, these opening pages suggest not only that a concern for ethics is on Foucault’s horizon but also an important insight into how to approach ethical questions or problematics in a Foucauldian way. And this key—that ethics is not outside of (as a Habermasian might characterize it) but essentially and intensively embedded within power relations—helps us see how Foucault’s various subsequent ethical strands of inquiry (aesthetic self-fashioning, friendships, sadomasochism, parrhesiastic courage, and the government of oneself and others) can be coherently integrated into an ethical vision.

So the second critical insight—interwoven with that first insight that power relations constitute a framework for ethics—is that this general analysis of power gives us concrete resources (as well as problems) for an ethics. As he characterizes his project in a later interview,

I am attempting . . . to open up problems that are as concrete and general as possible, problems that approach politics from behind and cut across societies on the diagonal, problems that are at once constituents of our history and constituted by that history. (DE341.1, 375–376)

Along his ethical itinerary, Foucault will develop a number of specific resources—possibilities and sites of resistance emerging from the particular domains under analysis, for example, sexuality. Indeed, most of Foucault’s initial ethical forays are explorations of this kind of specific resource. And just as his overarching understanding of modern power had to evolve, so will the strategies and tactics that he explores as ethical possibilities. We have already heard the first of these specific resources at the close of La volonté de savoir, when Foucault asserted that “the rallying point for the counterattack against the deployment of sexuality ought not to be sex-desire, but bodies and pleasures” (1976ET, 157). His exploration of “bodies and pleasures,” particularly in the explicitly power-laden practices of sadomasochism, will lead him to a new set or source of specific resources, which would fall under the rubric of friendship or “care” (that fundamental feature of pastoral power).

But Foucault’s development of these specific ethical resources is not unproblematic. Indeed, some of his most acute critics take him to task precisely because of these resources’ or resistances’ specificity: the challenge is how any ethical or normative stance that emerges from a contingent and specific situation could offer something broader than merely context-specific justification.

However, Foucault’s analysis of power also gives us more general resources that authorize a response to these critics. The first of these is the insight that has already emerged, that the multiple vectors of micro and macro power relations guarantee a certain amount of play, so that there is always room for resistance, and power relations can never be absolutely fixed into permanent states of domination. Beyond that, and essentially, the very basic structure of power relations—that power is only possible on the presupposition of freedom (and hence, resistance)—provides the resources for normative justification. This justification will obtain even though, given the contingency of any particular situation and the ongoing nature of the struggle, any course of action will remain dangerous.

These two critical insights—that an analysis of power relations necessarily frames ethics and that it simultaneously provides resources for such an ethics—are both reflected in Foucault’s own intellectual itinerary: it was only on the basis of his fully developed analysis of power that he was able to begin to speak explicitly about ethical concerns, without which his critical theory would remain incomplete. For as he noted in an 1978 interview with Duccio Trombadori, “I’ve never argued that a power mechanism suffices to characterize a society” (DE281.2, 293). Of course, Foucault always maintained a certain reticence about making proclamations—in the language of the first lecture of the 1978 course, he wanted his “imperatives” to be mere “tactical pointers,” or, in Kantian language, mere hypothetical imperatives. What, then, is the content of these imperatives? What constitutes the itinerary or trajectory of a Foucauldian ethics? Answering these questions, understanding Foucauldian ethics—its framework and the trajectories that emerge from this engagement with power—will be the work of our next chapter.