Bearing the lightning of possible storms: Foucault’s experimental social criticism

Abstract

This paper argues that Michel Foucault explicitly rejected the model of critique by which he is often understood—by both his defenders and detractors. Rather than justifying norms that could be said to represent “the people;” judging institutions, norms, and practices accordingly; and creating programs for others to enact, he theorized and practiced an experimental social criticism in which specific intellectuals help people work through “intolerable” situations by multiplying the ways they can think about and act upon them. As Foucault’s work with the prisons in France shows, one way intellectuals can be part of the experimental transformations social bodies carry out upon themselves is through genealogical work describing the ways problems have come to be identified—and can thus be transformed. This account of criticism undercuts the problem of justifying a standpoint of critique that has plagued philosophers and suggests a few concrete means of better aligning theory and practice.

After decades of dispute, major questions remain about the normative status and critical import of Michel Foucault’s work, though his readers, whether defenders or detractors, have tended to fall into two camps. Some have argued that Discipline and Punish, for example, is purely descriptive and has no basis for critiquing social norms or widespread institutions.¹ Others have suggested that Foucault’s work does a kind of “immanent critique” of Enlightenment values, continuing them in some important ways and modifying them in others, as, for example, “bio-power” might change the meaning of “freedom” but need not reject it.² Despite their differences, both interpretations assume that critique requires philosophers to provide a justification of the norms that could be said to represent the interests of those for whom they write, be it a group, a community, or humanity at large. These norms could then be used to evaluate institutions, norms, or practices in a representative manner. Foucault’s divided reception—that he does or does not endorse a normative standpoint—is based upon the shared assumption that critique depends upon a theoretical justification of such a standpoint that can be given in advance of the judgment of particulars.

Rather than adding yet one more argument either side of this debate, I would like to suggest that Foucault provides a different model of evaluation that undercuts both readings. Few philosophers have clearly articulated the theory and practice of criticism³ in which he took part,⁴ though it becomes manifest when his work and reflections upon it are connected with two terms: the “specific intellectual” and “problematization.” For Foucault, criticism is not primarily an intellectual act carried out by philosophers—whether in the form of defining, justifying or applying norms—because the application and evaluation of values cannot take place outside of action taken to deal with an actual problem with which people are struggling. Criticism involves the shared inquiry and action of those dealing with an “intolerable” situation, or what Foucault calls the “work within and upon the very body of society.”⁵ Which norms might be relevant and which may need to be rethought will depend upon the possible ways a problem can be dealt with effectively. This problem-based experimentalism⁶ may require radical changes in the theory and practice of intellectual work, for the investigation of norms should be a concrete affair involving action undertaken by those dealing with a problem, not a theoretical task carried out only by philosophers.

Foucault’s belief that intellectual work should be oriented towards and evaluated by its use in dealing with social problems is made clear in a conversation with Gilles Deleuze, who says, “A theorizing intellectual, for us, is no longer a subject, a representing or representative consciousness.”⁷ Rather than try to represent “the people,” specific intellectuals engage in a field where people already “know” what the problem is. This does not mean that intellectuals have no resources that might make them better suited for dealing with certain aspects of problems, for they may be able to help rethink a problem’s “problematization”—the conceptual framework and practices that make the problem possible—and suggest attuned re-problematizations⁸ with which people can experiment. There are many ways that problems can be problematized or re-problematized, only a few of which are best handled by intellectuals. Still, finding new ways of dealing with problems may often occur through differential historical research, or what Foucault terms “genealogy.” This is his tool of choice, but it is not the only one.

In order to defend my reading of Foucault against common interpretations, I will show how his interventions into the prison system can be understood as attempts to take part in experimental—not representative—social criticism. There has been intense debate over whether or not Discipline and Punish, in particular, is a descriptive or normative work. By drawing upon Foucault’s remarks in interviews and providing a re-reading of this classic text, I will show that Discipline was meant to provide a re-problematization of a problem with which many people were struggling. Instead of either justifying or assuming a representative standpoint or simply furnishing a factual account, the new framework he provides multiplies ways of thinking and acting so that people can experimentally determine what represents them.⁹ This kind of criticism would “bear the lightning of possible storms,”¹⁰ scattering possibilities for interpretation and transformation rather than leaving both to the province of philosophers. Foucault provides a number of suggestions for intellectuals to take part in such work, but I will conclude by pointing to some institutional changes that may help to realize experimental social criticism on a broader scale.

1 Representative critique

Before outlining Foucault’s experimental model of social criticism, it will be useful to sketch a widely held conception of critique, which I will label “representative.”¹¹ I do not mean to provide an exhaustive characterization or criticism of a vast range of philosophers but rather to show some of the shared assumptions and problems that lead to, in my consideration, misunderstandings of Foucault’s account, which I will call “experimental.” In brief, I suggest that representative critique is characterized by the assumption that social criticism requires the theoretical justification of a normative standpoint that can then be used to evaluate institutions, norms, and practices. This assumption creates the problem of providing a normative standpoint or ground—however shifting—that represents the interests of a community, nation, or humanity at large. Many standpoints have been suggested. For example, John Locke argues that human nature shows that life, liberty, and property are the criteria by which the legitimacy of governments can be judged.¹² Some philosophers, like John Rawls and Jürgen Habermas, have argued that legitimate critique is grounded in a certain kind of reasoning; others, such as Michael Walzer and Richard Rorty, believe that its authority derives from a specific connection to a community. The number of standpoints that have been provided suggests that it is time to reconsider this model of critique.

One way to attempt to philosophically justify a standpoint of critique is by appeal to reasons that others could be shown to accept. Rawls and Habermas provide some of the most well-known examples of this kind of argumentation. For example, Rawls argues that “the principles of justice for the basic structure of society are the object of [a hypothetical] original agreement. They are the principles that free and rational persons concerned to further their own interests would accept in an initial position of equality.”¹³ In Between Facts and Norms Habermas makes the case that “the legitimating force of a discursive process of opinion- and will-formation, in which the illocutionary binding forces of a use of language oriented to mutual understanding serve[s] to bring reason and will together—and lead[s] to convincing positions to which all individuals can agree without coercion.”¹⁴ These thinkers share the assumption that a definition of universal reason is necessary in order to justify critiques of widespread institutions—and that philosophy can provide it. They do not, however, agree on what universal reason is, nor is it clear how such a debate could be arbitrated.

Another common way to define the standpoint of critique is by appeal to the norms of a community or language game. Rorty and Walzer exemplify this approach. Recognizing that different people hold different beliefs, they try to avoid the debates over “reason” by grounding the standpoint of critique in inherited norms. For instance, Rorty argues that, rather than justifying anything, political philosophers are simply trading idealizations. Rejecting truth for description, he claims that it is “impossible to say that one language of moral and political deliberation, and the set of social practices intertwined with that language, is more rational than another.”¹⁵ Instead of being more rational, his account of justice simply “seem[s] to cohere better with the institutions of a liberal democracy than the available alternatives do.”¹⁶ Walzer gives a more nuanced account of the differences among communities and experts, though he still defines the ends of critique by the beliefs of the experts of a community: “There is a tradition, a body of moral knowledge; and there is this group of sages, arguing. There isn’t anything else. No discovery or invention can end the argument; no ‘proof’ takes precedence over the (temporary) majority of sages.”¹⁷ If asked why a certain standpoint is valid, Rorty and Walzer respond: because it represents the ends we share—or, rather, those that the “sages” tell us we share.

Despite major differences between Rawls, Habermas, Rorty, and Walzer, they all engage in purely theoretical debates about the representativeness of one normative standpoint or another. If one assumes that critique must take this representational form, one will interpret Foucault in a number of possible ways. First, he might be seen as providing a merely descriptive account of power relations and thus not engaging in critique at all, as Walzer suggests when he claims Foucault is a “nihilist.”¹⁸ On this reading, Discipline and Punish is meant to show that we cannot criticize any norms or practices because they are all equally coercive. Second, Foucault may smuggle in some kind of standpoint without admitting it, as Rorty seems to believe, understanding Foucault to be exposing the dangers of contemporary society to liberal self-creation.¹⁹ On this reading, Discipline and Punish simply accepts the good of non-coercion and criticizes contemporary institutions that do not live up to “our” own standards. Third, Foucault might occupy some kind of uneasy tension between a descriptive and normative account. For example, Habermas finds only contradiction between the posing of a “critique of power, disabled by the relevance of the contemporary moment, to the analytic of the true in such a way that the former is deprived of the normative standards it would have to derive from the latter.”²⁰ Finally, Foucault may be operating under a modification of the Enlightenment standpoint, as Amy Allen claims, writing that one might “even suggest that Foucault spent his entire career reworking Kant’s famous four questions, historicizing and contextualizing them as he went.”²¹

In the remainder of this paper, I will present an alternative interpretation of Foucault that undercuts the problem of defining and justifying a normative standpoint for critique in theory by rethinking criticism as an experimental social practice. Experimental social criticism involves the shared inquiry and action of those dealing with a problematic situation—or what Foucault calls the “work within and upon the very body of society.”²² Theory is necessary for such work, but it does not justify anything in advance. Instead, the validity of the “standpoint” of any intellectual work is shown by its use in helping people create less intolerable norms, practices, and institutions. That is, Foucault turns this theoretical problem into an experimental one, which, as we will see, makes very different demands upon intellectual work. It is for this reason that Foucault has been interpreted in such diverse ways and often in strict opposition to what he himself said about his work. I will consider his interviews and historical analyses in order to explicate his radical reconstruction of critique and ground my reinterpretation of Discipline and Punish.

2 Experimental social criticism as “a work within and upon the very body of society”

Foucault’s alternative to representational critique has two main components. First, it consists of a theoretical and practical response to specific, felt problems. Foucault’s major innovation is to rethink critique not as an intellectual act but as a series of experimental transformations carried out by people for the sake of ameliorating a problematic situation. Intellectual work is essential for making such transformations intelligent and effective, but it is simply one part of a critical response. Second, the differential effects of such criticism for those dealing with their problems determine its validity, not a theoretically justified standpoint. A course of action or the conceptual tools used to direct it cannot be validated in theory or in advance of the transformations they make possible. Criticism is inherently experimental, in that the lived experiences of actual people are its final arbiters and that nothing can be assumed to be demonstrated before action is taken.

The first way that Foucault differentiates his experimental account from the representational model is by the distinction between the universal and specific intellectual. In the face of value conflicts, we often assume that a universal standpoint is needed to decide right from wrong. We need to know, before problems arise, how they should be solved. The universal intellectual represents universal moral truth, which administrators and policy-makers can apply when faced with tough decisions. On this “trickle down” model of moral expertise, a select group of people have access to truth and justice, while others merely instantiate their universally valid knowledge.²³ These applications have no bearing on the validity or formation of universal norms or procedures. For instance, whether or not a dictatorship of the proletariat has disastrous consequences has no bearing on the universal values that have history on their side. Intuitive as this view is, it faces a number of well-known theoretical and practical challenges. In particular, since intellectuals do not all agree, they spend a great deal of time arguing about which standpoint is truly universal or what might constitute the authority of a representative value. What would resolve these debates and how laypeople could contest the experts is unclear.

To contribute to a model of criticism that would not prioritize the experiences of those deemed experts, Foucault describes a new figure of expertise: the specific intellectual. The distinction between the universal and specific intellectual is often taken to be a claim about the subject matter of intellectual work, i.e., that it should pertain to institutions, norms, and practices, like prisons, health, or sexuality, rather than the Good, the True, or the Beautiful. Foucault encourages this interpretation when he says that the specific intellectual has “become used to working…within specific sectors, at the precise points where their own conditions of life or work situate them (housing, the hospital, the asylum, the laboratory, the university, family and sexual relations).”²⁴

However, I would like to suggest that subject matter makes up only one part of the specificity of the specific intellectual, who, first and foremost, is defined by her attempt to help people come to terms with situations felt by many to be “intolerable.”²⁵ Far from the Cartesian sentiment that “everything must be doubted,” Foucault explains that “for a domain of action, a behavior, to enter the field of thought, it is necessary for a certain number of factors to have made it uncertain, to have made it lose its familiarity, or to have provoked a certain number of difficulties around it.”²⁶ While the universal intellectual attempts to provide criteria for solving future problems in advance of them, the work of a specific intellectual should be, in some way, a response to existing situations. Rather than trying to demonstrate that issues in the penal system should be judged according to norms of humanity or efficacy, for example, specific intellectual work should respond to actual conflicts regarding penality. This is a prescription for practice as much, if not more, than it is for theory.

If the intellectual is not to be the judge of people’s experiences, by developing a program for what must be done according to universal criteria that could be said to represent a people’s interests, the effects of her work in helping to create situations that are less intolerable should determine its value. That is, intellectual activity is not critical unless it leads to critical effects—or, as Foucault aptly phrases it, “when critique has been played out in the real, not when reformers have realized their ideas.” On this model of criticism, which Foucault calls “experimental,” the specific intellectual is one participant in a whole body of inquiry and experimentation interested in transforming problematic institutions, norms, and practices. The value of her work depends upon the actions she makes possible for those who, like her, are working to transform the prisons, for example. Intellectual labor is only one part of “a whole social project, a work within and upon the very body of society.”²⁷

Experimental social criticism must be attuned to the differential effects of the courses of action taken to ameliorate a problematic situation. Though theoretical work should help to make predictions about what can be done, there will always be unintended consequences of institutional reform. Moreover, these effects will be different for different groups of people who must be brought into the critical process. Rather than paving over such differential effects by appeal to theoretical justification, we must find ways of giving them voice. Foucault explains,

We should transform the field of social institutions into a field of experimentation, in order to determine which levers to turn and which bolts to loosen in order to bring about the desired effects. It is indeed important to undertake a campaign of decentralization, for example, in order to bring the users closer to the decision-making centers on which they depend, and to tie them into the decision-making process, avoiding the type of great, globalizing integration that leaves people in complete ignorance about the conditions of particular judgments. We should then multiply these experiments wherever possible on the particularly important and interesting terrain of the social, considering that an entire institutional system, now fragile, will probably undergo a restructuring from top to bottom.²⁸

If intellectual work is to be a part of the much larger practice of institutional transformation it should be as an aid to experiments that people are already undertaking. That is, it should be attuned to the purposes, conditions, and effects of the projected transformations of intolerable situations. This, in turn, may require major changes in institutions of expertise, as I will later suggest.

Paying attention to effects is a theoretical and practical task. One important tack is to avoid the concepts and problems common to representative critique.²⁹ For example, Foucault rejects theoretical definitions of the rational, though not because he is a relativist or irrationalist. Gesturing towards Habermas, he explains,

I think the blackmail that has very often been at work in every critique of reason or every critical inquiry into the history of rationality (either you accept rationality or you fall prey to the irrational) operates as though a rational critique of rationality were impossible, or as though a rational history of all the ramifications and all the bifurcations, a contingent history of reason, were impossible.³⁰

Instead of uncovering reasonable reasons to which all reasonable people could reasonably assent or the pragmatic-ideal presuppositions of a rational society, Foucault affirms the possibility of experimentally evaluating reasons and their ramifications in reference to specific situations. Understanding the history of the present may help create new ways of thinking and acting, but these cannot be validated by an intellectual, even a specific one. Philosophical justification cannot guarantee that old conceptual frameworks and courses of action will help people deal with new problems. It is the experimentation and discovery of new problems and practices that is most likely to invent the new rationalities and technologies that are needed to address the intolerable.

Similarly, Foucault attempts to avoid the difficult questions that relying on a pre-existing consensus faces, such as how a community could come to share the same beliefs, how a critic could become representative of those values, and how it could be known that the courses of action they make possible would have the same validity for everyone in “the community.” In response to Rorty, Foucault argues that

the problem is, precisely, to decide if it is actually suitable to place oneself within a “we” in order to assert the principles one recognizes and the values one accepts; or if it is not, rather, necessary to make the future formation of a “we” possible by elaborating the question. Because it seems to me that the “we” should not be previous to the question; it can only be the result—and the necessarily temporary result—of the question as it is posed in the new terms in which one formulates it.³¹

Foucault does not claim to represent any community, because, on an experimental framework, the value of any transformation based upon even shared values cannot be assumed to have valid consequences for the entire community. If, for instance, Madness and Civilization allowed for effective criticism—that is, if it allowed psychiatric institutions to be successfully reconfigured according to certain points of view—the validity of such purposes is a product, not a postulate, and it will only be valid for certain people. Not philosophers but people themselves must weigh its differential consequences with those of other possible courses of action made possible by other ways of thinking and acting.

3 Philosophers as problematizers

Though there are many ways that intellectuals could be involved in the social body’s experimental work upon itself, one that is perhaps particularly pertinent to philosophers is what Foucault terms “problematization.”³² Rather than providing the solution to a social problem—such as more rights, less government, more autonomy, or less harm—which “laypeople” are supposed to carry out, it is an intellectual’s

problematization that responds to these difficulties, but by doing something quite other than expressing them or manifesting them: in connection with them, it develops the conditions in which possible responses can be given; [problematization] defines the elements that will constitute what the different solutions attempt to respond to.³³

It should be noted that problematization is not the province of the philosopher only but rather that it is “problematizations through which being offers itself to be, necessarily, thought.”³⁴ Still, to be involved in a more general “work within and upon the very body of society,” one thing intellectuals can do is help re-define the space in which problems can be recognized and addressed. Though there are many other factors that can make possible new ways of dealing with problems—such as natural disasters, new technologies, or linguistic events—intellectuals may often be best able to make suggestions attuned to the history of past problematizations and their effects.

In the next section I will show how Foucault attempts to re-problematize debates about the prison system in France. But first I will briefly outline some consequences of seeing problematization as merely one part of experimental social criticism rather than providing a specific plan for reform. Problematizations—like rights versus democracy, homosexuality versus heterosexuality, prison reform versus increased incarceration—are not solutions; they define the space of experimentation which can be undertaken within the work of transforming institutions, norms, and practices. The ways of thinking and acting a new problematization makes possible

cannot easily be resolved. Years, decades, of work and political imagination will be necessary, work at the grass roots, with the people directly affected, restoring their right to speak…. It’s a matter of working through things little by little, of introducing modifications that are able if not to find solutions, at least to change the given terms of the problem.³⁵

The distinction Foucault makes between problematizations and problems suggests that the very naming of a moral or political problem is an exclusionary or selective act that allows for certain courses of action rather than others. Social problems are not given, perennial, or solvable only by experts. Instead, experts bring with them problematizations that name situations in certain ways, as economists will be interested in “incentives,” political theorists will worry about “coercion,” and bureaucrats will demand to know about “the bottom line.” For this reason, new problematizations may be needed to change the space of a problem as initially recognized so that new voices and interests can be heard and new solutions can be tried.

Since people have already understood themselves and their problems along the lines of problematizations formed in response to and made possible by prior situations, new problematizations or re-problematizations should be attuned to these histories—that is, they should be genealogical “histor[ies] of the present.”³⁶ Imagining better ways of thinking and acting is likely to involve the renovation of past categories in a manner attuned to present conditions for the sake of effecting future transformations. For instance, Foucault suggests that the dilemmas of family-centered economies were transformed by the invention of the concept “population” and the application of statistics to it. Since many contemporary economic debates now occur within the problematization of the “national economy,”³⁷ the imagination of new ways of dealing with “economic” problems may be aided by such a history. To successfully deal with the antinomy between prison reform and increased incarceration, for example, “there would be a considerable amount of work to do in order to renovate the conceptual categories that inspire our way of approaching all of these problems…. For the moment, we lack completely the intellectual instruments to envisage in new terms the framework within which we could achieve our goals.”³⁸ Descriptions of the conditions and differential effects of these two problematizations—the national economy; prison reform versus multiplying deterrents—provide resources for re-problematizations. Genealogy or “[h]istory becomes ‘effective’ to the degree that it introduces discontinuity into our very being,”³⁹ but discontinuities are always defined in specific ways and in reference to specific problems. An effective history can aid in the imagination of better futures because it names what has been tried for which reasons and points towards attuned reconstructions.

Because new problems emerge, are understood, and are addressed on the terms of existing ways of thinking and acting, it follows that intellectuals should multiply possible problematizations. Rather than trying to show that everyone could accept their point of view, specific intellectuals would create new possibilities for institutions to be evaluated and contemporary practices reconstructed. This practice of criticism

would bear the lightning of possible storms… [It] would try not to judge but… would light fires, watch the grass grow, listen to the wind, and catch the sea foam in the breeze and scatter it. It would multiply not judgments but signs of existence; it would summon them, drag them from their sleep. Perhaps it would invent them sometimes—all the better.⁴⁰

Of course, 1984 multiplying problematizations does not require philosophers to infinitely problematize and re-problematize contemporary situations, for whether or not another round of problematization is necessary will depend upon the effects it would make possible and the conditions that exist. Though almost anything could be better, some issues are more pressing than others. Nonetheless, the multiplication of problematizations should often be useful for dealing with our felt problems, because, though we already have conceptual resources for understanding the present—such as utility, rights, community, difference—we frequently suffer from “inadequate means for thinking about everything that is happening.”⁴¹ Instead of trying to “prove” the necessity or universality of a particular problematization, intellectuals could be lightning rods—conduits for dispersing energies.

On Foucault’s account of problematization and social criticism, many political theories and moral philosophies could be reconstructed as possible spaces of experimentation, rather than as attempts to represent a group of people. For instance, Habermas’s ideal speech situation may be a useful tool for addressing certain problems, but its concrete validity is not determined by his (lengthy) arguments for it. They may explain in part why such an hypothesis might be useful, but a genealogical account is more likely to help people understand the probable effects it should make possible since it pays explicit attention to them.⁴² This methodological point is the greatest difference between Foucault and many contemporary theorists, for he rejects, not expertise as such, but the one-directional practices wherein experts reduce thought and action to their favorite problematization. It may appear as though Foucault intends that “power” should take the place of the “ideal speech situation” as the essential way of understanding and criticizing institutions, particularly because he does not often spell out the interests or purposes for which his genealogies may be useful. But Foucault asserts, “I’ve never claimed that power was going to explain everything…. For me, power is what needs to be explained.”⁴³ Power is a problematization whose value is not shown by theory alone, but by the askeseis it makes possible.

We must not imagine that it is so easy to shift problematizations because institutions of expertise fund certain ways of thinking and acting through their frameworks, norms, and practices. The complexity of the modern world would be intractable without colleges and universities, professional organizations, and bureaucracies. Yet such institutions may easily reproduce the practices of what Foucault calls the universal intellectual by narrowing the ways in which people can understand and transform their problems. For instance, political parties are institutions of expertise that severely limit the possibilities for understanding intolerable situations due to their rigid problematizations. Foucault writes, “One may wonder whether the political parties are not the most stultifying political inventions since the nineteenth century. Intellectual political sterility appears… to be one of the salient facts of our time.”⁴⁴ If philosophers are to interrupt such problematizations, they may need to experimental not only with new problematizations but with different practices as well.

4 Re-problematizing the prisons

I will now show how Foucault attempted to put experimental social criticism, in the form of problematization, into practice. As we have seen, many of Foucault’s detractors focus on his work on the prisons, particularly Discipline and Punish and its use of “power.” Many have argued that he either provides no standpoint for his critique of “disciplinary society” or that his genealogies can only be descriptive. That the normative/descriptive debate misses the point entirely, however, can be shown by Foucault’s own claims and writings. In order to defend my interpretation of Foucault against many common objections, I will show how Discipline and Punish undercuts the problem of justifying a normative standpoint in advance—which is assumed by practitioners of representative critique—by providing a genealogical re-problematization with which people themselves can experiment. That Foucault provides no answers about what should be done about the prisons is a virtue and indicates the type of the intellectual work that may be needed to shift the ways that public problems can be more effectively understood and addressed. He attempts to understand the beliefs, norms, and practices that have lead to the present in order to provide an alternative problematization to that which has dominated discussion about and reform of the prisons.⁴⁵

Like many of Foucault’s works, Discipline and Punish cannot be summarized easily, as it extends into many different fields, histories, powers, and problems. One of its most powerful contributions to our ways of understanding penality, however, is its suggestion that both penal reform and prison proliferation are two sides of the same coin. Though both positions assume that the prison is a necessary institution for contemporary society, only a modus vivendi between the two is possible due to their vastly different ends. The former, believing that the prison should correct people, seeks to remove obstacles to this goal, such as cruel and unusual punishment; the latter, concerned not with correction but deterrence, tries to multiply deterrents. The popularity of these problematizations has waxed and waned throughout the years, but their shared premise that the prison system is problematic—though for different reasons—has remained unchanged. Whichever of the two points of view wins out for a time will have more to do with external factors than with intelligent experimentation.

Foucault attempts to form a new problem space that might transform the deadlock between these two problematizations and allow for different experiments and possibilities. He does this in two ways: first, by exploring the productive aspects of power and looking to the ways prisons are predicated on a specific technique of power that Foucault calls “discipline.” This turns an interminable debate about ends into an analysis of effects. Second, by showing how, on these grounds, prisons are not failures at all but have actually succeeded for certain purposes, Foucault transforms the space in which we consider the effects of prisons. Rather than looking only to how prisons deter criminals or reform them, we can inquire into the ways they reproduce a specific form of power. Both moves are made possible by a careful genealogy of the shifting conceptions and practices of punishment that occurred in the 18th and 19th centuries.

After tracing the development of the theory and practice of disciplinary power in schools, hospitals, and, of course, prisons, Foucault shows that imprisonment came to be the choice form of punishment because of the ways it intensified discipline. In spite of the theory of reformers, who sought to match the punishment to the crime in an obvious way, prisons aided in the production of categories of criminals, the administration of techniques for normalizing individuals, and the experimentation of the human sciences that investigate the makeup of criminality. In particular, prisons created a new category of the individual: the delinquent. Foucault writes, “For the observation that prison fails to eliminate crime, one should perhaps substitute the hypothesis that prison has succeeded extremely well in producing delinquency, a specific type… of illegality.” Prisons contribute to a form of penality that “does not simply ‘check’ illegalities; it ‘differentiates’ them, it provides them with a general ‘economy.’ Foucault suggests that prisons have been exceptionally good at producing a class of people that are potential delinquents, “a relatively small and enclosed group of individuals on whom a constant surveillance may be kept.”⁴⁶

Imprisonment has a certain number of advantages, the greatest of which is that it helped to transform the monarchical system of illegality. While the bourgeoisie retained the illegality of right, the “rabble” was confined to the illegality of property and thus required constant policing. The prison system allowed the bourgeoisie to prevent the lower classes from committing more serious crimes, use them as tools for colonization and imperialism, profit from them, as in the cases of prostitution or drugs, or employ them as tools of the state.⁴⁷ In short, rather than seeing prisons as failures, and thus requiring either harsher or more lenient measures, Foucault’s genealogical analysis opens up a new problematization for understanding them. It does not reconcile the problematizations of prison reform or the multiplication of deterrents, but provides a new site of thinking and acting. Rather than polarized posturing and the continuation of old practices under a parade of new names, experimentations in penality might now be undertaken to realize a broader set of needs.

What exactly should be done is not clear, and it is not up to Foucault to decide whether or not this problematization is helpful—or critical—for those affected by prisons. He writes,

If prisons and punitive mechanisms are transformed, it won’t be because a plan of reform has found its way into the heads of the social workers; it will be when those who have a stake in that reality, all those people, have come into collision with each other and with themselves, run into dead ends, problems, and impossibilities, been through conflicts and confrontations.⁴⁸

Unless some common, perhaps long-term, ground can be found that transforms the narrow interests currently allowed expression, the most legitimate option for a group of people may be a modus vivendi or tyranny of the majority. It is likely that not everyone will find a voice nor will everyone be satisfied by the effects of a new problem space, but this simply means that new problematizations and new possibilities for amelioration may need to be attempted. Despite Foucault’s careful archival work, it is clear that contemporary systems of penality still produce many illegitimacies. This is to be expected, as true reform will require a great deal of work by everyone involved. It is always easier to maintain the status quo.

The uncertain or experimental nature of genealogical re-problematization can be seen by the way in which many theorists have simply appropriated Foucault’s language of “power” or “discipline” by statist problematizations that dichotomize coercion and legitimate political power. If, despite Foucault’s emphasis on its productive aspects, power is understood as coercion, Discipline and Punish would seem to argue that legitimate political power is a contradiction in terms.⁴⁹ Foucault himself admits that some of his earlier writings may have been unclear on this point,⁵⁰ yet “power” is simply a tool for historical analysis and is designed to name the plurality of ways in which institutions like the prisons arise and can be transformed. One cannot assume that these will take the form of coercion or consent, let alone a sovereign and a subject. Of course, “power” will not be the right problematization for every intolerable situation. This is why, rather than being a mystery to be solved, the variety of Foucault’s approaches evidences an awareness of the multiplicity of conceptual tools and methodologies that different problems may require.⁵¹ If “power,” and disciplinary power in particular, are useful conceptual tools, this is to be shown by the transformations they make possible.

5 Multiplying experiments

If, as Deleuze explains, theory “does not totalize,” if “it is an instrument for multiplication and it also multiplies itself”⁵²—that is, if intellectual work cannot predict its effects—one cannot assume that philosophers can demonstrate the value of any standpoint, set of ends, or intellectual work based on them. The specific intellectual’s task is not to discover what people believe but to help define spaces in which beliefs could be experimentally tested and revised, or to provide experiences that make possible the transformation of institutions in which means and ends can be determined to be true or false, valid or invalid.⁵³ Criticism would then describe this whole process, beginning with people’s felt problems, their re-problematization (though not necessarily by intellectuals), and complementary experimentation, transformation, and evaluation. For this reason, rather than trying to represent the oppressed or speak the truth about them, Foucault and Deleuze argue that specific intellectuals should help create, for example, “the conditions that permit the prisoners themselves to speak.”⁵⁴ This need not be an anti-theoretical, un-critical, or anti-intellectual position, though it suggests that intellectual work should be tested by its use in helping people deal with problems. According to Foucault, “work done at the limits of ourselves must, on the one hand, open up a realm of historical inquiry and, on the other, put itself to the test of reality, of contemporary reality, both to grasp the points where change is possible and desirable, and to determine the precise form this change should take.”⁵⁵

When, where, how, and for whom experimental social criticism actually leads to better practices will depend on concrete material and social factors. In other words, the value of experimental social criticism should itself be experimentally evaluated. However, realizing this kind of experimental social criticism on a large scale would require widespread changes in the ways theory is grounded in and responds to practice, for “grasp[ing] the points where change is possible and desirable” as well as the “form this change should take” should not be tasks limited to public intellectuals like Foucault. Nor must it depend upon complex genealogical problematizations. Experimental social criticism may require institutional changes to which Foucault draws our attention but does not furnish many resources. How could the research universities, media, and governments be better attuned to intolerable situations and more open to experimentation regarding the value of their knowledges?

These institutional questions are being tackled on a number of fronts, one of which is research into the professions. Albert Dzur makes a number of suggestions for rethinking the ways in which expert knowledge is brought to bear upon pressing social problems through a framework he calls “democratic professionalism.”⁵⁶ Journalists, for example, often help legitimate the status quo’s understanding of a situation. By experimenting with different ways of naming problems, they might provide opportunities for people to see themselves as actors rather than spectators, as is often the case in many political and economic affairs.⁵⁷ Another field of research has studied the ways in which universities might better partner researchers with people working through problems. Since land grant universities were designed to play a public role rather than accumulate “knowledge,” Michigan State University’s Office of University Outreach and Engagement, for example, has experimented with a number of ways of pairing academics and community groups, particularly those involved in economic development.⁵⁸ These are only a few examples, and if intellectuals are to help “bear the lightning of possible storms” by multiplying experimentations in new ways of thinking, acting, and valuing in response to intolerable situations, there is much work to be done.