The original French edition of the first, introductory, volume of Michel Foucault’s History of Sexuality appeared as long ago as 1976 and the expectations it generated were very high. Here was one of the most distinguished and sophisticated practitioners of the history of ideas and institutions turning his attention to yet another fundamental issue in the development of modern European sensibility. What he had already done for such subjects as madness, the clinic, and the prison, he was now to attempt to do also for “sexuality,” a notion which may seem utterly familiar, but which only becomes explicit for the first time in the nineteenth century.

In each of his major earlier studies he analyzed an idea or an institution which is nowadays often taken very much for granted, investigating its background or antecedents and the ways in which it is validated or legitimated. In the process he demolished or at least undermined many cherished assumptions about the self-evident rationality of Western culture and society. The earlier works on which his brilliant reputation was based had in each case not just suggested a new understanding of the subjects they discussed, but also advocated, and practiced, a new methodology in the history of ideas. The projected multivolume work on sexuality can be seen as, in some respects at least, continuous with some of his earlier pre-occupations. Some brief remarks on those earlier works will serve as a reminder of some of Foucault’s distinctive insights and style and help to locate the starting point for the study of sexuality.

“A History of Insanity in the Age of Reason” is the English subtitle and the principal topic of Madness and Civilization, his first major study, in which he pursued the issue of just how the Enlightenment understood madness and treated those it deemed mad. As in his subsequent works, the investigation involved not just the analysis of the changes and interplay of concepts, but more the analysis of the institutions that gave these concepts concrete expression. Among other things, this study showed how in an age of reason, and indeed in the name of reason, confinement came to be used increasingly as a method of dealing with those who were considered not to conform to the norms of rational behavior. Foucault charted, in particular, the foundation and the varying fortunes of the great Parisian asylums, such as the Hôpital Général and La Salpêtrière, and the spread of institutions based on these models—and of course their eventual decline. He analyzed the types of justification put forward by those responsible for setting them up and used by those who ran them—the protection of society, but also often the good of the inmates themselves—and he described the transformations that occurred in the perception of madness once such institutions existed to give, as it were, visible proof of society’s verdict that those confined were mad. Nor did he fail to point to the exaggerations in the idea that no sooner were those asylums closed than a vast improvement occurred in the understanding and treatment of the mad—both issues whose controversiality shows no signs of diminishing today.

Not the least of Foucault’s achievements was to have insisted that a society’s conception of madness cannot be understood in isolation from the political, economic, religious, legal, and philosophical factors that all contribute to the notions of order and normality at work in that society. For example, what types of sanctions are available and used to control deviants, how do expectations about normal behavior vary with the economic status of the people whose behavior is being judged, and what kinds of implicit or explicit beliefs about human nature influence or dictate the assumptions of members of that society about tolerable deviation?

Other similarly wide-ranging studies explored not only those other fundamental institutions of modern civilization, the clinic and the prison, but also the development of the human sciences, such as political economy, and even of science in general. While the starting point, on each occasion, may look easily recognizable as a possible topic of conventional historical study, whether in the history of institutions or in that of ideas, in each case Foucault transformed the problem, notably by insisting on the complex inter-relations of the manifestations of power, on the one hand, and knowledge, on the other. It was, of course, one of Foucault’s major recurrent preoccupations to explore the exercise of power well beyond its more obvious manifestations in, for instance, the domain of politics, including the definition and transmission of what passes for scientific knowledge. Whatever some scientists in the past or today might say, science itself was not a wholly disinterested pursuit of objective truth; no more, to be sure, was education the innocent transmission of such truth; and while others had made similar points about science and education before, no one had drawn attention so emphatically to the varieties of ways in which they acted as the sources of control and influence.


Even when he studied the development of penal law in Discipline and Punish, where others might have taken this as a more or less self-contained subject, Foucault insisted on the larger perspective. The book discussed punitive methods not simply as consequences of legislation, nor again just for what they revealed about social relations, but against the background of his far-reaching notion of power. In that perspective prisons came to be seen as one example, though no doubt a quite distinctive one, of the methods developed in modern times for the control of others. The breadth of the perspective, and the ambition of the project, come out clearly in Foucault’s own statement that his book is a history not just of penal systems, but one

of the modern soul and of a new power to judge; a genealogy of the present scientific-legal complex from which the power to punish derives its bases, justifications and rules; from which it extends its effects and by which it masks its exorbitant singularity.

Conversely, in his investigations of the development of the human sciences, first in The Order of Things and then more clearly in The Archaeology of Knowledge, Foucault dismantled many common assumptions relating to the conduct of the history of ideas. Much conventional intellectual history was flawed, in his view, by a failure to pay sufficient attention to the complex social, cultural, and even political conditions within which ideas are produced. Ideas as such have no history and can be understood in isolation neither from the individuals or groups who proposed them nor from the questions of why they proposed them, how they justified them, and what use they made of them. Foucault investigated three important changes in particular: the way in which natural history was superseded by biology; the way in which the inquiry known as general grammar gave way to philology; and how reflections on wealth and trade came to be supplanted by what we know as political economy. In each case the center of interest was not themes or topics, not even authors or works, but the whole nature of what he called the discourse that constituted a science as it was practiced at particular historical junctures.

For instance, how did those who engaged in natural history construe their inquiry? What types of data were considered relevant? How was the inquiry related to past learning? Or to other genres? Above all, what assumptions were made about the nature of the natural world under investigation? The discourses in question were thus each, broadly speaking, governed by certain sets of conventions or rules, and it was the business of the historian to unearth these in what Foucault came to advocate as a type of “archaeology.” In the process, where many writers had interpreted each of those three changes very much in terms of gradual transitions and continuities, Foucault stressed rather the specific nature of each of the disciplines he compared, and the breaks or discontinuities marking the development of the new human sciences.

This is not the occasion to attempt to summarize the strengths and the limitations of these earlier studies, but certain general observations are in order. First there are important respects in which all his major studies form a continuous whole, though certainly not in the sense that they are a set of investigations planned and predicted at the outset—for that he explicitly and vehemently denied. Yet they are so in the sense that certain problems and preoccupations recur, especially those to do with the complex inter-relations of power and knowledge.

The second point concerns Foucault’s capacity for self-criticism. Like other highly innovative thinkers, he has, naturally enough, attracted his fair share of criticism, for instance from those who have objected to his theses concerning the radical breaks marked by the development of the new human sciences. (Thus the rise of political economy has recently been the subject of sophisticated studies by Istvan Hont and Michael Ignatieff that modify the picture presented by Foucault.) But so far as the methodology of his investigations goes, he had probably been his own most unsparing critic. Thus the introduction to The Archaeology of Knowledge devotes several pages to a radical critique of Madness and Civilization, The Birth of the Clinic, and The Order of Things, criticizing the last of these, for example, on the grounds that at the stage at which it was written he had not made fully explicit the methodology on which the study of the subjects it investigated had to be based. Indeed, what passes for a conclusion to The Archaeology of Knowledge takes the form of an imaginary dialogue in which the very possibility of the investigation it attempts is questioned. With characteristic honesty, even if perhaps with just a touch of disingenuousness, Foucault there expresses the intention of the book as being simply to overcome certain difficulties that are preliminary to his strategic enterprise.


This takes me to the third point. Foucault’s effort to refine the methodology, to probe to the limit the question of just how the subjects concerned can be investigated, to unmask prevailing assumptions about the nature of historical inquiry itself, is surely among the most notable and durable contributions he has made. Whatever we may think about some of his specific conclusions, the type of critical inquiry he engaged in will, or should, never be the same again, not least because of the way he challenged conventional boundaries between disciplines, and demanded methodological self-awareness. He was not the first to make such challenges and demands, to be sure, but by calling attention to the fundamental methodological and definitional problems that need to be solved, his statement of them acquired exceptional force.


These three general points all bear on the project on sexuality. First, however, it should be explained what Foucault intended by this project. He was always scrupulous to define it both negatively and positively, as neither a history of sexual behavior and practices nor a history of the ideas, whether moral, religious, or scientific ideas, that have been used to understand, describe, and explain such behavior and practices. It was to be, precisely, a history of “sexuality” itself, the process by which this familiar notion of twentieth-century Western society came to be the familiar notion it is, invoked repeatedly both in our description and in our understanding of ourselves—both as the objects of the sexuality of others and as subjects possessing such sexuality ourselves—and the topic of much explicit comment, discussion, and theorizing among psychologists, sociologists, sociobiologists, and the like. Foucault rejected the assumption that sexuality is some kind of constant: it is not just that the term itself is an invention of the nineteenth century, but our modern awareness of the issues we discuss under the rubric of sexuality is the outcome of certain historical developments, changes, or transitions that Foucault set out to unearth.

What the project on sexuality has in common with his earlier studies is, first of all, that here too there is an evident interplay of power and knowledge. This can be seen both in relations among people, where the exercise of power often takes the form of the exploitation of the inferior or weaker by the stronger, and in would-be scientific understanding, when appeal is made to notions of normality in order to support recommendations concerning behavior.

But in the course of this project Foucault’s self-criticism and his insistence on methodological self-awareness were also very much in evidence and indeed led him to make serious revisions both in the scope of the study and in the approach he adopted. When the French edition of the second and third volumes finally appeared, eight years had elapsed since the first; Foucault completed the fourth and final volume only shortly before his sudden death, in 1984, at the age of only fifty-eight, and it is to be published posthumously.

Foucault explained his second thoughts about the project and its methodology in the introduction to the second volume, L’Usage des plaisirs, translated now under the title The Use of Pleasure. Initially he had treated the topic of sexuality very much as he had his other studies on the themes of power and knowledge, and up to a point he continued to seek to build on that earlier work—as he put it, to use the tools they provided. Certainly there was no going back on the idea of the need to see power and knowledge, in their manifold manifestations, in relation to each other. On the other hand, in the case of the concept of sexuality he came to be quite dissatisfied with an investigation limiting itself to the early modern period, that is to say to the striking changes and developments of the last two hundred years.

What was required, he says, was a new “theoretical shift”—one that involved a broad extension both of the time scale of the study and of the subject matter to be covered—that would engage what he called the “history of desiring man,” a complex and difficult but important notion that provides the key to the revised project. That history was not just of desires, that is to say of kinds of desire, but also of the person or self as the agent of desire. It may be taken to include both the history of the ideas and attitudes held about desires (for example how far they were deemed natural, how far dangerous, how far even evil), and the development of, and changes in, the concept of the person as the seat of desire—the “subject” of whom desires are predicated. Ultimately, then, it is a study of how human nature, the good life, and the goals for human beings were viewed.

Not content with a brief historical survey of the theme of desire Foucault chose to engage, to a degree he had never attempted before, in detailed investigations of pagan antiquity and of early Christianity. To decide thus to extend the original project took some courage and determination, for it meant long immersion in a body of literature—the corpus of extant Greek and Latin literature, no less, down to and including the early Christian fathers—with which as a self-confessed nonspecialist he was initially fairly unfamiliar. The project was inevitably delayed; at the same time the potential benefits to be gained were considerable, because, of course, if the study of these earlier periods could be carried through successfully, it would give his account of the eventual rise of sexuality in modern times altogether greater historical depth.

The question is how far Foucault succeeded in doing for pagan antiquity what he did for our understanding of aspects of the Enlightenment, and I shall concentrate here on his discussion of the classical and early Hellenistic periods in The Use of Pleasure and allude only briefly to its sequel, the study of late pagan antiquity in Volume III, called Le Souci de soi.

Foucault’s wide-ranging discussion deals first with the central question of what he calls the “problematization of pleasure.” This is not to be understood as suggesting that the ancient Greeks had any difficulty in expressing the concept of pleasure (they did not) or again in their capacity to enjoy or experience it. Rather, Foucault explored the kinds of pleasure that could be subject to moral disapproval or, more generally, the ways in which pleasure could be the source of anxiety. Negative answers to the questions thus raised turn out to be easier to give than positive ones, for it was clearly not the case that the ancient Greeks saw sexuality as such as a problem, or as something to be censured or feared; nor did disapproval or fear stem from a preoccupation with any equivalent to the later Christian concept of flesh. Many Greek authors, of course, Plato, Xenophon, and Aristotle not the least among them, wrote of the need to control the desires for food, drink, and sex; but food, Foucault claims with only a slight degree of exaggeration, was often a more prominent preoccupation than sex. They often expressed concern about the correct use of ta aphrodisia, the “things of Aphrodite,” especially about timing, appropriateness, and moderation: making love was all right, to paraphrase a recurrent motif in Aristotle’s ethics, at the right time, in the right place, with the right people, and in the right way—and not too much. The center of anxiety and of disapproval was not sexual desire itself, but the threat of being dominated in one of two ways, either by becoming a slave to one’s own desires, or—far worse—by being made subject to someone else’s dominion, to be used by that other person to give him or her pleasure, sexual or otherwise, at his or her whim.

Foucault’s subsequent analysis deals mainly with four further topics: attitudes toward the body, to the institution of marriage, to the relationship to boys, and to the existence of wisdom (the relationship to “truth”). On the first of these he uses the early medical writers—that is, the anonymous authors of the treatises in the Hippocratic corpus—or at least some of them, to point to a recurrent concern with the question of determining a regimen to secure physical well-being. Along with the advice they offer on food, drink, exercise, and so on, these writers, who are mostly themselves medical practitioners, not infrequently allude also to sexual intercourse, specifying, for instance, that sexual activity should be increased or decreased according to the season. (More sex, for example, for males in winter.)

Foucault rightly remarks that while they are preoccupied with how much sex people had, they show no signs of being interested in what form it took: intercourse was intercourse in whatever shape or kind. There is, however, in many Hippocratic writers, more of a continuity between a concern for physical, and one for psychical, well-being than Foucault’s discussion appears to allow.

The main treatise he uses, On Regimen, ends with a book devoted to the interpretation of dreams: this medical practitioner saw the well-being of the soul, as testified by the patient’s dreams, as just as much a part of the doctor’s task as the health of the body. Moreover, secondly, Foucault sticks almost entirely to the dietetic treatises which are mainly concerned with males. He ignores what the extensive Hippocratic gynecological works have to say on—as they see it—the radically different way in which the woman’s body works.

Thirdly, neither in The Use of Pleasure nor in Le Souci de soi does Foucault deal at all adequately with what pagan Greek and Latin authors had to say on the topic of abstinence, sexual or otherwise. From the Hellenistic period, beginning with Theophrastus and continuing with many later writers in the Neo-Pythagorean or Neoplatonist traditions especially, abstinence in one or another form found increasingly articulate spokesmen. Aline Rousselle dealt in detail with this topic in a monograph (somewhat misleadingly entitled Porneia) published in 1983. Foucault cites this in a footnote, but no doubt it appeared too late for him to use it to the full. But he thereby missed an important connection between paganism and early Christianity, since Rousselle was able to show that there were important pagan antecedents to some expressions of the Christian ideal of the ascetic life of the holy man.

Although aspects of the documentation Foucault collects are thus open to fairly serious objections, the principal conclusion that emerges from his treatment of this theme of attitudes to the body—as from his other chapters—is a broadly convincing one. This is the negative point that for the Greeks of the classical and Hellenistic periods sexuality as such was not an important issue. They mapped the issues that interested them quite differently from the way we do when we subsume a variety of phenomena and experiences under the general category of sexuality. Health was important, to be sure, and sexual behavior had a part to play in maintaining it, but no more important a part than many other aspects of a regimen of physical well-being. Marriage, too, was important, but what counted there was not the sexual relationship between the marriage partners, but purely and simply the production of legitimate children. Love too was important, both in heterosexual and in homosexual relations, but even here sexual fulfillment was not the only, or even—often—the primary, concern, for honor and self-esteem were also generally at stake.

In a careful discussion of that last topic Foucault resists the temptation to argue that heterosexual love was always subordinated, in pagan antiquity, to homosexual; above all he insists, on the whole correctly, on the inappropriateness of attributing to the Greeks the notion that homosexual and heterosexual desire are two kinds of desire. Appetite was directed, rather, at beautiful human beings, of whatever sex. If male homosexual love (other than with male prostitutes) was sometimes particularly highly prized, this was in part because it was seen—by lovers, if not also by the beloved—as more precarious than heterosexual. Unlike married women or girls considered as potential wives, whose spheres of activity and influence were within the home, boys and young men moved in the public world outside the house.

So too did female prostitutes, of course, but unlike them, in turn, young males were free agents able to accept or refuse lovers—and therefore a greater prize for those whom they accepted. To persuade them to accept, lovers had to show some tact and discretion, as examples from such dialogues of Plato as the Lysis indicate: they might even engage in something akin to a courtship of the beloved, though (as Foucault did not fail to point out) if this was courtship, it still fell a long way short of such later examples as medieval courtly love in the degree of elaboration of its rituals. Nevertheless sexuality here too was not an overriding concern, for the desire for consummation had sometimes to be balanced against the need to maintain the honor of the beloved.

Finally in the most ambitious and speculative section of his work, in his discussion of the topic of “truth,” Foucault took famous texts in Plato, the Phaedrus and the Symposium especially, as signs of the beginnings of what he called a “hermeneutics of desire.” His argument here is that Plato, more than any other pagan writer, foreshadows Christian ideas concerning the need for self-examination (as Plato made Socrates say: the unexamined life is not worth living), the need to scrutinize and to be self-conscious about desires especially, the need finally to purify oneself of desires, to renounce them, and to treat life as a preparation for death and immortality. On how far Foucault sustains his thesis on this issue with respect to Christianity, judgment should be reserved, for a fuller discussion of the early Christian period is promised in the fourth and final volume—Les Aveux de la chair—still to be published. The question for us here is how far his reading of Plato appears plausible. Foucault himself is careful not to assimilate Plato too closely to later Christian positions, even though he sees a concern for self-examination, even some of the language of purification, as common ground. But his chief reason for expressing reservations is that for Plato, as for many other pagan Greeks, the good life was seen in aesthetic terms, as a matter of the pursuit of the beautiful. Yet that may miss the main point, or one of them, which is that, for Plato, the good life is devoted to the philosophical quest for truth, and to the practice of dialectical argument with that aim; and truth, for him, was not a matter of revelation or faith.


No brief survey can do justice to the richness, complexity, and detail of Foucault’s discussion. Certainly that complexity and detail are among the strengths of this book. Another, already noted, is that his general argument is convincing on the negative point, the absence, in pagan antiquity, of a unifying concept equivalent to our notion of sexuality itself. Thirdly—despite the reservations I have just expressed about the discussion of “truth” in Plato—Foucault usually strikes the right balance on the question of the principal similarities and differences between pagan and Christian attitudes to sex, if not on the sanctions supporting them. Since pagan antiquity produced nothing quite to equal St. Paul’s fulminations against sex, it is easy—but superficial—to see a simple antithesis between pagan sexual tolerance and Christian repression. Foucault avoids that trap. He has no difficulty, for instance, in illustrating the expression of what he calls a “very ancient fear” concerning the expenditure involved in sexual activity—the loss of semen in ejaculation is a loss of the life force—or again of finding examples in pagan authors (such as Aristophanes) where homosexual behavior is depicted in ways that could hardly be described as tolerant. At the same time Foucault knew his pagan Greek and Latin authors well enough to appreciate that if it was oversimplified to treat the sexual morality of paganism as the polar opposite of that of Christianity, it was also a mistake to treat the two as forming a continuity.

Nevertheless in addition to the criticisms I have already expressed, the book suffers from certain serious defects or shortcomings. First, surprisingly in view of the great methodological sophistication already noted in Foucault’s discussions of the general issues of power and knowledge, his account of classical pagan attitudes exhibits two important weaknesses in method. The first concerns the extent to which he was prepared to pay attention to the differences in genre and diversities of interests in the chief sources on which he drew. The problems of evaluating sources are not totally ignored: for example, Foucault drew attention to major gaps in the written evidence available, the difficulties of reconstructing Pythagorean beliefs and theories for instance; and he appreciates the idiosyncrasies of such an author as Aristophanes. Yet the attempt to generalize about classical pagan views and attitudes as a whole—and to contrast these with both Christian and modern ones—repeatedly misleads Foucault into taking as typical what is clearly not.

Of course reconstructing what some hypothetical classical Greek man, woman, or slave in the street thought about these matters is impossible; but Foucault does not sufficiently acknowledge that most of the written evidence comes from the literate elite and a good deal of it from a particular section of it, the moralists. Plato and Aristotle can, of course, be taken as evidence for the expression of certain themes and problems; but at many points they are typical of no one but themselves. Yet Foucault incautiously attempts to generalize about Greek views of desire on the basis of philosophical texts (such as Plato’s Republic or Aristotle’s Nicomachean Ethics) whose authors were conscious of innovating, either by modifying and criticizing popular beliefs or indeed by developing their own philosophical theses in reaction to those of their predecessors.

Even more misguidedly Foucault stays almost exclusively within the bounds set by the written testimony and he neither cites nor illustrates the very extensive evidence that Greek vase painting yields. That evidence poses its own problems of interpretation, of course. But as K.J. Dover showed in his study of homosexuality (a text Foucault refers to more than once)* if we want to understand Greek lovemaking techniques and sexual practices generally—for instance the position the young man was expected to adopt in male homosexual intercourse—then the visual representations on pots provide direct evidence on many points where Foucault, sticking to the writers, remains puzzled by their “reticence.” Reticent is certainly not how the vase painters can be characterized, for they depict the widest possible variety of sexual practices, whether homosexual or heterosexual, with the greatest gusto.

The final general criticism that must be made is again a surprising one, in view of Foucault’s earlier work. This is that in The Use of Pleasure he largely ignores the political setting of the problems he discusses. The one exception is some speculations on Greek expectations about the sexual behavior, and especially the homosexual relations, of their political leaders, where Foucault suggests that the combination (real or imagined) of a dominant political role and a passive sexual one was, for the Greeks, a particular source of outrage. But he mentions only in passing the major contrast between paganism and Christianity presented by the fact that in the latter views on sex came to be made the subject of teaching by an established Church—which could and did control dissidents by the use of religious sanctions for which there was no equivalent in pagan antiquity.

Just as he takes little account of the diversity of interests of authors of different types, so too he fails generally to consider how discussion of some of these issues in classical Greek authors belongs to a certain style of self-conscious, open, and at points competitive debate. The classical Greek polis prided itself—or at least many of its foremost spokesmen from Herodotus down to Aristotle did—on settling disputes, in whatever domain, by reasoned argument. The challenge to give an account, of actions, attitudes, opinions, and theories, was frequently issued and taken up, and in the evaluation of the presentation of a position or thesis what counted was not any appeal to authority—certainly not to the type of religious authority represented by the teaching of the Church—so much as the strength and validity of the arguments adduced, at least as these appeared to the participants in the debate.

The difference this should make to our assessment of the classical Greek contributions to the issues Foucault discusses—health, desire, even the goals of human life—is that in one respect much less was at stake in those ancient debates than later, in another, much more. Much less was at stake in the sense that an element of debate for its own sake can be detected in some of those contributions—certainly in some of the early medical writers and in the Sophists, and even Aristotle is not above giving instruction about how to win an argument by deception. Ideas were sometimes tried out, half playfully, or at least without the total commitment of those who proposed them. (An example would be Gorgias’ account of agency in his Encomium of Helen.) But again more was at stake in another sense, in that, without a body of sacred texts to appeal to as ultimate authority, the resources of merely human reason and imagination had to be made to do more work, both in inventing theories and explanations and in justifying those proposed.

Foucault’s foray into the field of the interpretation of classical antiquity is, then, only partially successful. It is more valuable for the problems it sets than for the answers it offers, though it establishes well enough that our preoccupations with sexuality were not those of the ancient Greeks: they had no such sciences as psychology pronouncing on sexuality, no popular obsession with the subject, let alone any equivalent to its modern commercial exploitation. But when he deals with the themes that are made explicit in ancient debates, pleasure, desire, love, and so on, while he has much to say that is subtle and penetrating, one of the chief weaknesses of his analysis turns out to be one that we might least have expected from a critic who had earlier insisted on the need for a total archaeological excavation of the past. It is as if long immersion in unfamiliar source material diverted his attention from one of the most important lessons he had himself taught, about the need to study ideas in the full complexity of their political setting. That is to turn one of Foucault’s chief earlier methodological insights against himself. The Use of Pleasure at least establishes an agenda of central problems for the historian, even if it does not itself satisfactorily carry through their analysis.

This Issue

March 13, 1986