The Use of Cruelty in Society

A qualified appreciation of Michel Foucault.

The two biggest influences on how I think about history and social theory have been F. A. Hayek and Michel Foucault. They’re a better pairing than you might think, and my goal here will be to describe how I make sense of some of Foucault’s major ideas. (I’ll write about Hayek later.) For now I’d just like to offer an accessible introduction to the parts of Foucault’s work that I’ve found interesting and/or useful, and how they inform my perspective on some current issues.

What a philosopher wrote is seldom up for debate. What they meant is a different question. And how we use their work is still another. I’m aiming between the second and third of those. I’m not being comprehensive, and I’m not surveying all the other interpretations of Foucault that are out there. And there are a few places in what follows where my engagement with Foucault’s ideas just can’t account for some things he said and did, and some I find blameworthy. I’m not tying up all the loose ends, and there’s no substitute for going back to the source, which you absolutely should do if you find this stuff interesting.

Marx’s Shadow

I often hear lately that Foucault was a Marxist. I disagreed in a Twitter thread a few weeks ago using quotes from The History of Sexuality vol I to show that Foucault rejected Marxist ideas about historical causality and the direction he thought history was headed. Foucault’s academic milieu was full of Marxism, and his early works show him uncritically accepting some, though far from all, Marxist assumptions. These, though, he moved away from rather decisively in his later career. It might be fair to say that Foucault grew out of being a Marxist, so showing how he differs from Marxism also shows how his thinking developed.

Marxist theories of history generally urge that we should see laws, mores, and practices as the outgrowths of an ongoing class conflict. The material requirements of production—that is, the ways that we make what we consume—will lead most societies to be divided into social classes. In capitalism, the workers work, but they own neither the machines they work on nor the products they produce. The capitalists own the machines—but they do not labor. The differences in the material conditions of the classes go on to shape their interests, values, and ideas.

One goal for a historian in the Marxist tradition is to show the links between the economic “base” and the “superstructure” of phenomena that the economic system produces and sustains. History, to a Marxist, is not ultimately a story of dynasties or ethnic groups, but of class conflict. And that conflict will continue until, sometime in the future, the workers’ revolution overthrows capitalism. At that hoped-for time, man’s exploitation of man will come to an end, and so will history. 

That approach was very much standard among Marxists in the 1950s and 60s, and thus very much standard in the French academy where Foucault worked. Socioeconomic class was a real phenomenon for Foucault—as it should be for any good social scientist or theorist—and so was class antagonism. But he didn’t view the socioeconomic classes as locked in an epochal, all-important struggle over the means of production that would ultimately liberate humanity. 

That’s a lot to leave out. Foucault was skeptical of such grand narratives: Marx was wrong not just about the direction that history is headed, but about whether history has any necessary direction at all. The nineteenth century had had a lot of class conflict, and it was right for theorists at the time to focus on it, but the result was a warped perspective on the rest of history: “Marxism,” wrote Foucault, “exists in nineteenth-century thought like a fish in water, that is, it is unable to breathe anywhere else.”¹

Still, under capitalism, capitalists do often collude, usually by means of the state, to make life worse for workers and better for themselves. Of course they do. Even an American libertarian can sometimes see it, and when they do, they will probably—and correctly—call it a case of rent-seeking. It’s just that rent-seeking isn’t nearly all of what capitalism does—really, it’s the smaller part—and ending it won’t end history.²

Cultural historians often ask about the origins of practices and institutions. For a Marxist, a question like that is nearly answered before it is asked—it’s class oppression, stupid, and the historian’s task is just to figure out how to link up the two. The differences in historiography among different kinds of Marxists tend  not to be about whether practices and institutions are rooted in class conflict; they agree on that. They tend to disagree mainly about how to fit the model of class conflict onto the data of historical events. 

For Foucault, however, classes and institutions often err. Sometimes they work against their own Marxist interests. At times they are less the agents of history—and more like the conduits through which historical developments flow. Consider, for example, the prudishness of the Victorian era. Marxist historians claimed it was the work of the bourgeoisie, who wanted to shape the proletariat into a tractable labor force, one that wouldn’t trouble anyone about their sexual desires or the risks they brought.

But Foucault noted that the bourgeois medical and psychological authorities of the late eighteenth and early nineteenth centuries first targeted the bourgeoisie itself, and not the proletariat, with a new sexual ethos:

If one writes the history of sexuality in terms of repression, relating this repression to the utilization of labor capacity, one must suppose that sexual controls were the more intense and meticulous as they were directed at the poorer classes; one has to assume that they followed the path of greatest domination and the most systematic exploitation: the young adult man, possessing nothing more than his life force, had to be the primary target of a subjugation destined to shift the energy available for useless pleasure toward compulsory labor. But this does not appear to be the way things actually happened. On the contrary, the most rigorous techniques were formed and, more particularly, applied first, with the greatest intensity, in the economically privileged and politically dominant classes.³

The nineteenth century’s prudery came from the influential members of the bourgeoisie first and most strictly disciplining themselves—the better to demonstrate, to themselves and their peers, their own worthiness. They also sought to recreate that worthiness in the rising generation using the latest methods science had to offer—and making them up when science fell short. They would eventually enlist the government in that project, and the result was eugenics. Reformers were not hoping to subject proletarians to immiserating labor. They were hoping to uplift the bourgeoisie and purify it through social technology.

Repression started with the repression of the powerful. I might add: It may well have spread by emulation, and by the attempt to climb society’s ranks—an act that Marxist ideology didn’t think could happen at any great scale, although in reality it certainly did. To join the middle class is possible even if Marxists don’t like to talk about it, but the price of admission may be to take on middle-class sexual mores.

In conversation I’ve often characterized Foucault as a perverse functionalist. In functionalist sociology, phenomena are explained with reference to their tendency to perpetuate the society in which they are found. Emile Durkheim, one of the founders of that school, argued that even suicide served a social purpose, for it revealed the points of tension in a society’s own operations; a suicide was a social signal that ultimately helped that society to stabilize itself and address the underlying problems that had prompted the suicide. 

In a perverse functionalism, phenomena, including many that are arbitrary, cruel, and immiserating, perpetuate themselves without regard to the continuity or well-being of the society in which they are found. These phenomena are contingent on previous phenomena for their content and development; they are not made from scratch as the products of our rational deliberation.  Social phenomena care neither about individuals nor about society. They are replicated, modified, redeployed, and ramified, but not because they play any great role in a long historical drama. Why are they there? That’s something we still need to explain, and neither Marxism nor functionalism can do it alone.

The Use of Madness

Consider Foucault’s observation, in his early work Madness and Civilization, that “Language is the first and last structure of madness, its constituent form … That the essence of madness can be ultimately defined in the simple structure of a discourse … gives it a hold over the totality of soul and body.” [Emphasis added.]

There’s a lot to unpack in that. In the early modern era, “madness” was generally identified with, and by, how people spoke, wrote, and thought. Madness was usually identified with unreason, and so one knew madman through the lack of reason in his speech. Words, after all, are both how we make sense of the world—and also how lies and unreason are born.

But here’s where things get interesting. “Madness” had a big range of associations. It had many forms and could manifest in many ways. Foucault demonstrates with a wide variety of primary source texts how these theories shifted over time, often in bewildering ways. “Madness” could be said of the mind, in asserting that a belief was mad, or that a belief was maddening (either was possible). Or madness could be said of the body, by claiming an organic cause for a madness, though sometimes the body was reflective of the madness. Or “madness” could even be said of the soul, in claiming a moral cause for madness—and certainly madness had moral consequences. All of these beliefs can be found in the early modern era, and often, to our way of thinking, they have been strenuously and bizarrely confounded.

Yet all of this ambiguity made the discourse of madness a convenient way to intervene in the lives of others. A truly excellent tyrant always wants to control you in body, mind, and soul, and for a time, “madness” delivered. Not only that, but “madness” as a discourse could even control the lives of people who were otherwise completely sane: Don’t masturbate, said Dr. Tissot. It’ll make you crazy. Generations grew up believing that, and they disciplined themselves accordingly.

Foucault would say that our notion of mental illness is a retreat from the more expansive category of madness; the former has surrendered the latter’s moral or spiritual dimension. Madness in the old sense is seldom discussed clinically today, and it shows up only by implication in the way that American law treats insanity. But it remains interesting to historians, and especially to historians of liberty, because it shows the kinds of work that language can do in support of arbitrary power.

How’s that? When “madness” is thought to be constituted by language—issued by the madman, interpreted by those who would classify and cure him—it permits us to apply a wide range of treatments to a wide range of symptoms: We can electroshock people so that they say the right things, and we know we’ve shocked them enough when they do so. We can talk to them in the hopes of curing a physical malady, and again, we know it’s cured when they respond in certain ways, both in behavior and speech. We can give them drugs to help their souls, or prescribe moralized exercises—like forced labor—and then say it’s for their own good. (They will thank us later.) Physical, mental, and moral can blur whenever discourse is the benchmark of success. The concept is sloppy, so the power that flows from it is arbitrary.

In the end, the point of saying that language is the first and last structure of madness is not to assert that madness is a made-up thing we can all ignore, or that the insane are just people like you and me, or that psychiatry is a put-on. And still less does it say that nothing is real. I’m trying to be charitable, but these are very, very shallow caricatures of what Foucault believed. As I’ve said elsewhere, many things are socially constructed—and at the same time they’re tyrannical and immovable. What I’ve said of gender, I could also say of madness:

The institutions and practices of gender are not to be trifled with. Gender may be socially constructed, and of course any constructivist is committed to saying so, but this does not mean that gender is arbitrary or freely chosen: Gender is constructed, but so are prisons.

To say “X is socially constructed” is not to say “X is arbitrary and easy to change.” It’s more like “Our society uses power in the defense of X. That means it’s real now, not just theoretical. Tamper with X at your peril.” 

The American psychiatrist Thomas Szasz is often likened to Foucault for his critiques of involuntary psychiatric commitment, which Foucault also opposed. Szasz practiced psychiatry only by consent, and he condemned the state’s entanglement in his profession, again like Foucault. And even the way he talked about mental illness was similar, in a way; Szasz called mental illness a “metaphor for human problems in living.” Again, the claim is that it’s more about language than about anything else.

Foucault’s book Madness and Civilization emphasized how the psychiatric establishment had literally begun behind bars, when, in 1656, Louis XIV founded the Hôpital Général de Paris. Out of this strange institution there eventually grew both the modern idea of insanity and the modern insane asylum.

But the Hôpital was not a hospital in our sense; it was an archipelago of workhouses. Beggars were sent there for the offense of being indigent. In another dissent from Marxism, Foucault noted that the workhouses of Europe, and particularly those of the Hôpital, seldom turned a profit, and often there was no work for the inmates to do. An institution like that doesn’t exploit anyone in the economic sense, it just confines them. Exploitation would suggest a forced transfer of the value of labor power, which wasn’t occurring.⁴

Very soon after the institution’s founding, something like one percent of the entire population of Paris had been locked away in the Hôpital, a development Foucault termed the Great Confinement. We might even call it the first mass incarceration in western history, for the Hôpital understood itself to be a place of punishment and correction. The Hôpital was an extraordinary development in another way: the West’s previous institutions of mass seclusion for moral reasons had all been religious, and at least nominally, they had been voluntary. The Hôpital Général was neither.

Much like Paris in the Old Regime, the United States is also a mass incarcerating society. In the mid-twentieth century we removed most forcibly confined mental patients from their dedicated institutions, partly thanks to Szasz. But a lot of those released inmates soon became homeless, and so we confined them once again, this time in prisons.

The mentally ill aren’t more criminal than the rest of us, but they perform worse on the witness stand, they’re worse at keeping the terms of parole, and they’re worse at obeying the arbitrary rules that prevail in prison. So, once again, they are lumped in with the criminals. The poor, too, end up disproportionately in prison, even when they’re not mentally ill. Mass incarceration also disproportionately affects people of color in ways that in some ways recall the biases that the system has shown toward the poor and the mentally ill. In a way, we’ve stumbled back into keeping those people in an Hôpital Général. We moderns love our confinement.⁵

Power over Life

Foucault had a complex, multifactor account of historical causality. He was also pretty ad hoc about it at times, but some patterns do seem clear. One of the biggest, longest-term themes in Foucault’s work, and one that  is especially clear in his later writings, is that in the ancient world a polity’s chief task was to allocate and preserve the licit, socially sanctioned right to kill; this right was essentially identical with sovereignty.  That’s just what the Romans did, as we saw in our last installment; the Romans gave that right to patriarchs, who may kill the members of their households. Patriarchs brought them into the world; patriarchs can take them out of it. 

More recently the right to kill has been held in reserve. Its bearers are restricted to a genetic line of individuals (kings), a given social class (slavemasters), or a given segment of the state bureaucracy (police). The favored ones can usually delegate the power to others, but not without some constraint. Death is threatened more often if we count the indirect threats, but it’s imposed much less.

That’s because the right to kill has been deliberately kept at the far end of the system. From the sovereign right to kill, and from the all too predictable desire of the subjects to go on living, we can easily see the next step in a crucial sequence: a “society in which power was exercised mainly as a means of deduction (prélèvement), a subtraction mechanism, a right to appropriate a portion of the wealth, a tax of products, goods and services, labor and blood, levied on the subjects.”⁶ 

After that, they could occupy themselves with all the things that may be done to a person or a population once they are accustomed to the idea that if they misbehave, someone else may kill them or take their stuff. Foucault writes:

Since the classical age [essentially what we would call the early modern era], the West has undergone a very profound transformation of these mechanisms of power. [Prélèvement] has tended to be no longer the major form of power but merely one element among the others, working to incite, reinforce, control, monitor, optimize, and organize the forces under it: a power bent on generating forces, making them grow, and ordering them, rather than one dedicated to impeding them, making them submit, or destroying them. There has been a parallel shift in the right of death, or at least a tendency to align itself with the exigencies of a life-administering power and to define itself accordingly.⁷

Again, this is rather weirdly libertarian: The government that can kill you can just about always pick your pocket. And the government that can pick your pocket can just about always force you to eat your vegetables. That is, it can treat you and all other citizens as an optimization problem. 

Foucault used the term knowledge-power to describe the networks of learning and teaching, and examining and reforming, that use other human beings for their raw material. He viewed modern government as a nearly comprehensive set of institutions bearing different kinds of knowledge-power. Professional psychiatry, organic medicine, anthropology, linguistics, penology—these sciences have been professionalized and brought into symbiosis with the state. Foucault termed this reconfiguration of authority and knowledge biopower, and it was for him one of the signal developments of modernity.

But the powerful state was dangerous: “It is as managers of life and survival, of bodies and the race, that so many regimes have been able to wage so many wars, causing so many men to be killed… If genocide is indeed the dream of modern powers, this is not because of a recent return of the ancient right to kill; it is because power is situated and exercised at the level of life, the species, the race, and the large-scale phenomena of the population.”⁸ 

Often the web of power extends into the mind of the subject itself. That’s why Foucault used the celebrated example of the panopticon in Discipline and Punish. Few willingly enter a panopticon. It takes the right to kill, which is to say a government, to put them in. But once there, they get used to the idea of being watched and judged by an unseen outside force, one that may or may not be observing them at any given moment and without their knowledge. The conscience that the authority wishes to inculcate is soon internalized, and the prisoner becomes docile. That’s a much more useful trick than just killing someone or taking their money. And even if it fails, there’s nothing saying that you have to let them out. 

Modernity also has a disconcerting tendency to cast all of society not just as a family, but as one body with one will—an exceptionally dangerous metaphor, one that inclines its users to regard anything unexpected as a disease. And, even more than a population, surely a single body has only a single will, or else it is touched by madness.

We have no unitary will, but it’s usually the case that a few people can push enough of the others around that it looks that way. Belief in social solidarity helps: As Foucault put it,  “I believe the great fantasy is the idea of a social body constituted by the universality of wills.”⁹ His was a radical individualism; consider that, for Jean-Jacques Rousseau, the very founding act of a society was an agreement by all of a given people to be bound in every particular by a unitary will, without which he held that there could be no society to speak of. I’m not aware that Foucault ever wrote about the Contrat Social, but it’s hard to imagine him saying anything good about it. Rousseau would close off every escape hatch in the machinery of power.

Foucault’s individualism is also why, rightly understood, his work sits poorly with identity politics. Populate the social body however you like—say it’s made up of any race, gender, or orientation—the universality of wills remains a fantasy, and the fantasy only sustains itself by cruelty to dissenters. The category of “homosexual”—created, as Foucault observed, as the subject of clinical study in the late nineteenth century—offers no escape from social conflict, not merely because of its origin, but because it grants very little, and nothing necessarily, of a shared will. It’s a category for examining and directing a population. It’s no proper foundation for social solidarity. The shared will of those caught in a cold, clinical category that was made up to study me? It might be good not to commit to that.

The Will Not to Be Governed

It’s hard to take Foucault seriously and not end up an anarchist. Yet here’s where I should probably acknowledge that Foucault’s critiques of Maoism were disconcertingly mild, and at times he collaborated with Maoist activists. He also rather amorally cheered the Islamic Revolution in Iran. “One thing must be clear,” he wrote as that revolution progressed. “By ‘Islamic government,’ nobody in Iran means a political regime in which the clerics would have a role of supervision or control.”

The claim seems shockingly naïve. Who could fail to see that Islamic government was just another fantasy of a universal will? How could a pro-gay and philosemitic academic see this revolution as an inspiring event? It may have been a fine thing to impartially record the inchoate hopes of a people, but Foucault also recorded the crowds chanting "Khomeini for King,” and he interviewed Khomeini himself, and perhaps he should have thought a bit more about the whole business before he wrote.

Like a lot of French intellectuals of his era, Foucault also spent some time, earlier in his career, in the Communist Party; he later renounced it, and there’s even a cheeky case for a Foucauldian anti-communism. The Soviets infamously turned political dissidents into psychiatric prisoners, and that could easily have kicked off a sequel to Madness and Civilization. Few societies have pursued the discipline of the individual more thoroughly than the Soviet Union, and few have exercised biopower more ruthlessly. (Maoist China is preeminent among them.)

It’s a fun riff, and Foucault is always a good guide to the use of cruelty in society, but it’s hard to keep it up for too long when the man himself kept the associations he did.  What Foucault appears to have been seeking in all of these failures, and certainly with way more optimism than warranted, was the will not to be governed. As one recent scholar put it, “the problem of obedience has become more ambiguous in late modern post-industrial societies, where the breaking down of traditional structures of authority … seems to coincide with ever greater levels of obedience and docility.” 

We obey not because we fear the strongman, but because we have acquired the habit of obedience. Docility in the Foucauldian sense is close to learned helplessness. But how do we escape docility, which is an interior disposition, one that presents itself to us as authentic, one that feels right? To put it mildly, I would hope there are more promising forms of revolt than Maoism or Shi’ite religious government. A better escape may first take a good deal of introspection, as one recent writer put it:

an ethical transformation of the self, even a form of ethical self-discipline. For Foucault, the subjectivity that power creates for us is also the material from which we can resist power and from which we can fashion for ourselves new ways of being. But this involves an ethical work conducted upon the self, a conscious practice of self-constitution.

A good anarchist should look inward, care for himself, and choose reflectively the time and manner of his revolt. Don’t look to the headlines, and don’t even trust your friends. If you’re doing it right, your opportunity probably won’t be a part of an organized movement at all. Perhaps nobody but you will notice. But you’ll be all that matters. (You always are, by the way. Any bad faith in this regard will always, always have an observer, who is you.) If you go down the political road, especially in the modern world, success may just mean that you’ll end up being the one who makes up new excuses for the panopticon. Don’t be surprised when they turn it against you.

But it would be a mean trick to drag you all this way just to leave you at the examined life. You want politics? Fine, here goes.

For a long time I’ve been a part of an American movement that’s at least nominally dedicated to individual liberation. Yet while the American libertarian movement often rightly recoils from biopower, I along with many others have found that it’s apt to rush headlong into a revival of the right of death: There’s a lot of classical republicanism in the average American libertarian, and we’ve seen what that’s like.¹⁰ Worse, the Libertarian Party was also recently taken over by its longstanding ethnonationalist faction. These folks don’t hate biopower. They can’t; they want to achieve ends to which only biopower is fit: restricting immigration, suppressing expressions of same-sex desire and gender nonconformity, directing women to more traditional lifestyles. If the modern, all-directing state does not exist in their utopia, it will be necessary to invent it.¹¹

The standard rap on Foucault is that he’s interesting to think about, and yet there’s nothing left to be done. Everything is power; to understand is to lose hope. But we can sketch a Foucauldian activism from themes in his work: Reverse flows of power. Where resources—including people—have been gathered and organized for the use of those who hold power, scatter and disorganize them. Realize that identities and narratives are sometimes conduits through which both power and resistance can flow. Resist surveillance. Don’t engage in civil disobedience. Break unjust laws. Turning yourself in to show that you’re loyal to your society isn’t virtue. It’s just biopower: long ago, the state shaped your conscience, and now your conscience is turning its body in for physical punishment.

In American history, the Underground Railroad was a resistance movement that exemplified all of these unapologetic tactics long before Foucault. The flow of human beings and their confinement in the South was reversed; the Railroad sent them not along one route in public, but among many routes in secret. Narratives from the Bible were deployed to subversive intent. Trust and authority among the conspirators amounted to a network in competition with the ruling one. 

Foucault’s own programmatic politics is dated and disreputable, and that would hardly have surprised the man himself. His ideas on power survive best when they describe not an ideology, but a style of activism. There they’re open to all—a gift with unfathomable implications.  What would happen if Foucauldian tactics became a much more frequent answer to power? What aggregate effects will they produce when they are put toward many different forms of resistance at once, some of them at cross purposes with one another? (Are we finding that out right now?) My critique of Foucault isn’t that he offers no hope. It’s that he offers much more tactically than he does programmatically, and it’s not clear where all of it will lead. 

1. Michel Foucault, The Order of Things: An Archaeology of the Human Sciences. Tavistock, 1970, p 262.

2.  This and many other examples have led some sharp economists to claim that critical theory is just a sloppy, qualitative account of public choice economics. There’s some truth to this, but I would suggest that the economists consider that critical theory is a vast field, and not much of it has been described as rigorously as those parts that correspond to public choice. There might be a lot more to do here.

3.  Foucault, History of Sexuality vol I, p 120.

4.  By “value” here I mean strictly in the Marxist sense—I mean the value of labor power extracted by the capitalist class. But in another sense, value was indeed extracted, and viciously: the inmates’ free time was taken from them, and the king and the administrators of the Hôpital converted that free time into reputational rewards for themselves. In present-day terms, they were virtue signaling by controlling other people’s lives. This too is an extraction of value, and it may serve as a bridge from the narrow, Marxist idea of extraction to a much more inclusive one.

5.  “Hospital” went on to denote another institution, one which Foucault discussed in The Birth of the Clinic.

6.  Foucault, History of Sexuality vol 1, p 136.

7. Ibid.

8. Ibid., p 137.

9.  Foucault, Michel 1980. Power/knowledge: selected interviews and other writings 1972-1977. Translated and edited by Colin Gordon et al. New York: Pantheon Books, p 55.

10. It shows in how they pick their arguments, and I am not cheered when proponents of school choice advance their cause, which I also support, by threatening parents with the specter of a gay public schoolteacher, a dangerous source of effeminacy who threatens to destroy the family with his strange, foreign, drag queeny ways. But that’s just what a patriarchal argument would look like: the traditional fathers should be the authorities, and they will reinstate tradition. Never mind the families who would welcome a gay schoolteacher, and never mind the fathers like me who aren’t traditional. There is no room for either.

Meanwhile vaccination is an especially stark case of the clash between the two different views of the proper deployment of power: Proponents cite statistics on what is, for them, usually a technical question of population management. They seldom speak the language of natural rights. Those who start from natural rights meanwhile tend to vindicate parents’ choices about their children’s medical treatment unconditionally. That’s one among several possible solutions that a natural rights approach could yield, and it’s certainly the most patriarchal. But in general they are quick to discount the externalities associated with an unvaccinated population. Formalizing rights theory and treating it as a science of government was a project of the twentieth century, but it’s not one that has been completed yet, and it shows. Usually these two sides have talked past one another.

11.  I haven’t been a member of the Libertarian Party for many years, but lately I don’t even want the adjective. Words are supposed to inform, and at some point the Latinate roots of the sounds we utter have got to yield to the facts on the ground. Given the state of its institutional referents, “libertarian” is just plain confusing. It’s not a trait to recommend in a word, and the confusion benefits those with something to hide.