Liberal institutions stop being liberal as soon as they have been attained: after that, nothing damages freedom more terribly or more thoroughly than liberal institutions. Of course people know what these institutions do: they undermine the will to power, they set to work levelling mountains and valleys and call this morality, they make things small, cowardly, and enjoyable – they represent the continual triumph of herd animals. Liberalism: herd animalization, in other words.
Nietzsche1
Eighteenth century liberalism constituted itself in opposition to the authoritarian (police) state and its regulatory ambitions. It shifted responsibility and authority to the economic subject and civil society. But, during the second half of the nineteenth century, an important transformation of governmental techniques occurred, in which the state acquired an entirely new meaning: no longer partisan and potentially despotic, it came to signify neutrality, standing above and outside of society and its conflicts. In this light, governmentality meant that the state does not produce political inequality, nor does it preserve outdated social structures. Instead, it represents an indispensable element in resolving political antagonism and guaranteeing social progress. By the end of the century, the state no longer faces society as a foreign body; it becomes the state of society: the social state.2
This historical transformation has been examined from numerous standpoints: in terms of the increasing juridification of social relations, as the product of contradiction between political equality and economic inequality, and as the victory of the working class and its organizations. The theoretical peculiarity of the works discussed in the following is that they set aside these models of explanation in favour of a ‘genealogy of the welfare state’ that focuses on analysing ‘political rationalities’.3 Their common point of departure is the assumption that historical changes to the state cannot be separated from an epistemologico-political shift in the objectivation of society, which took material form during the nineteenth century and differs markedly from the early liberal idea of ‘civil society’.
Liberal thinkers had conceived society as the (voluntary) association of individuals, which ultimately derives from the economic and political interests of particulars. For liberal governmentality, the central point of reference is not society, but individuals. In this conception, civil society is identical with the sum of its parts and social phenomena do not constitute a reality sui generis – that is, one different from the reality of political and economic subjects. Society and subjects occupy the same plane; they do not belong to separate realms. Society is not a subject, but an object produced by a multiplicity of economic individual interests and political preferences.4
But, beginning in the second half of the nineteenth century, ‘society’ acquired a density of its own, making it impossible to trace it back to the aggregation of individual wills. Seemingly an independent and self-sufficient subject, it now represents something other than the sum of individual subjects; it can no longer be derived from its parts. Henceforth, ‘society’ names an autonomous sphere of relations with specific regularities and objects distinct from both economic and political relations. A ‘specific organizational form of society’,5 or a ‘new political positivity’,6 emerges: the social.
For the authors discussed here – who were Foucault’s colleagues and collaborators – the most important reason for this shift followed from the complete rearrangement of liberal politics towards a new system of security in which the state played a central role. The simultaneous emergence of mass poverty and an organized labour movement exposed the insufficiency of liberal security mechanisms, which centred on the individual and his or her civil liberties. In addition, the separation liberalism had effected between politics and economy posed more and more of a problem in terms of economic inequality and political equality. The nineteenth century faced a dilemma for which the liberal security system had no answer, which was neither economic nor political: the social question.7
In History of Madness, Foucault describes the establishment of the ‘Hôpital General’, where the ill, those in need of medical care, the insane and the unemployed all were held. More and more, over the course of the eighteenth century, this generalized practice of confinement came to be viewed as irrational. Poverty came to stand as a category of its own, as a matter requiring a specific mode of treatment. Its specificity implied that economic infirmity should be set apart from other forms of destitution. Gradually, the mixed and polyvalent forms of support that had defined care for the poor up into the eighteenth century were abandoned.
They were replaced by a (liberal) rationality that faulted traditional forms of charity and welfare for economic wastefulness and technical inefficiency. This twofold disqualification bore on the sizeable sums that flowed into general pauper relief. Not only did this mean that the poor remained outside the realm of production and found no ‘economic’ use; more still, such efforts, reasoning held, do not get rid of poverty – they prolong it indefinitely. Liberal rationality tried to break through this negative mechanism by formulating a (positive) relation between poverty and labour and fragmenting the spectrum: the figure of the ‘sick poor’ vanished and gave way to an array of distinctions between good and bad paupers: those who were simply lazy and those who sought work, people capable of working and those who were not, and so on. By reorganizing the problem around the pole of work, the operation sought to break down poverty in order to separate the ‘real’ poor from the impostors and freeloaders.8
Studies by Giovanna Procacci have stressed that liberal government did not replace one perception of the problem with another, nor did it inaugurate a change of perspective. What is more, traditional practices of welfare and charity did not yield to more efficient or ‘more economical’ approaches, and no new treatment for the old problem of poverty was proposed. Liberal thinking did not focus on how the problem should be viewed or remedied, but on the problem itself. In various forms and with different degrees of intensity, poverty has always existed. There may have been different ways to look at it or treat it, but in the context of liberal government – which meant a free labour market and the guarantee of equal rights for all – poverty changed status. With the abolition of feudal and estate barriers, the institution of free markets, and the postulate of legal equality, poverty could no longer be approached by traditional means such as paternalism and/or repressive exclusion. On the contrary: the poor became equals among equals. Only within the framework of government practice linking market forces and the statute of general, legal equality – thereby staging a charged relationship between economic liberty and political sovereignty – did being poor turn into a problem.9
According to François Ewald’s analysis of legal politics, the problem of social inequality did not represent a ‘blind spot’ of liberal thinking. On the contrary, examining the reasons for, and consequences of, inequality in a society of equals stood at centre stage. As such, what is remarkable about liberalism is not the absence of the theme of poverty, or lack of interest in it, but the attention and effort given to treating it. It was impossible for liberal thought to be blind to the fact of social inequality, for it performs an important social function. In other words, inequality does not represent a foreign body so much as an integral component of the social world of liberalism. Inequalities constitute civil society, which defines itself through social differences, and they provide the index for the naturalness of social phenomena and the possibility of society in general. Social inequalities are not the result of a defective society, but a necessary and indispensable element of its everyday functioning:
Inequalities are natural, inevitable, and irreducible: they constitute a component of the order of creation, which is an order of variety and diversity; as rewards and punishments for the merits and shortcomings of particulars, they are necessary for social progress; finally, they are part of providence: as wellsprings of social cohesion, they stand at the origin of the existence of societies. Thus, not only can they not be prevented – one should not seek to make them disappear … The existence of inequalities underlies both the possibility and necessity of a liberal art of government.10
This conception of civil society as the sum of individual interactions allowed liberal rationality to dismiss any idea of a social system that went beyond the will of particulars. Liberalism’s political ‘achievement’ is to have rejected the very thought of social causality: between the wealth of the ones and the poverty of the others, there is no social connection organizing relations according to a necessary principle. The reason and cause for inequality does not follow from a generative system, but derives from the different ways that individuals use their freedom. Liberalism replaces any thought of antagonism pervading society with the postulate of continuous, equal liberties. As such, the freedom of individual will and the principle of responsibility occupy the centre of liberal thinking. The problem does not involve rectifying social injustice, but coordinating wills and ensuring the coexistence of liberties.
As Ewald presents matters, liberals consider the social world to be simultaneously natural and just. No distinction, no difference of level, exists between natural and social justice that would require corrective measures; no basis exists for society intervening in nature. Accordingly, any step towards righting social relations not only runs counter to the natural self-regulation of social mechanisms, but also contradicts the principle of justice. Only in one instance is intervention in the natural course of social relations possible (and necessary): when their naturalness itself is at stake. If ‘freedom’ names the universal principle defining the naturalness of society, any liberty that restricts other liberties is unjust. Thus, the coexistence of liberties constitutes the measure and the limit of rights and the law. From the perspective of liberalism, the law does not constrain; on the contrary, it serves to critique constraint. Necessarily, law has negative character, because it derives from freedom and is oriented on it: it is impossible to force people to do anything; one can only prevent them from acting improperly. Constraint is legal in nature when it foils something that unjustly obstructs liberty: the law solders and fixes, but it does not correct.11
Liberalism regulates social relations not by way of the law but through morality. Inasmuch as the social world is a just world characterized by coexisting liberties, any instance of inequality must derive from a (false) use of freedom. General freedom of the will has its counterpart in constant moral evaluation that subjects all actions to critical judgement. The liberal world knows perpetrators, but no victims. The reasons for poverty lie in moral failing; it results from individual wills and subjective dispositions. This perspective does not merely define a certain way of understanding the causality of events; at the same time, it formulates a positive programme: if poverty does not arise from the (defective) structure of society, but instead stems from a faulty personality structure, then poverty may be abolished only by heightening moralization, transforming personal will and changing behaviour. As such, moral evaluation functions both as an instrument for diagnosis and a mode of therapy, naming the problem and the course to follow in order to solve it.
According to Ewald, the concept of individual responsibility plays a key role in the liberal strategy of moralization.12 It affords a principle of objectivation as well as a standard of judgement: everyone is responsible for his or her own life. Assuring seamless communication between the law and morality, it provides a unified mode of culpabilization and brackets any idea of social causality that extends beyond personal misconduct. Liberal rationality does not view poverty as a social ill based in economic operations organized along capitalist lines, but as a moral disposition and an individual form of conduct. Thus, efforts to combat it cannot concentrate on reorganizing social mechanisms; rather, they must focus on the individual will (or unwillingness), to which this condition owes its existence; without a transformation of the will, it will come back again and again. Poverty is not the result of an objective reality, but the product and sign of a subjective volition (or its absence). Responsibility for this condition lies ultimately with the poor themselves; it is a matter of individual, not social, responsibility, and as a social responsibility it is strictly an individual matter.13
The moralizing imperative does not concern only those without property; it also applies to those with property. The corollary of assigning personal responsibility for poverty is the principle of moral responsibility on the part of the wealthy. Indeed, liberalism is notable for many and varied practices of welfare and charity, which form an integral component of its politics of social security. As such, it does not reject the idea of the social obligation of the rich towards the poor, nor does it contest the need for charitable works. However, it envisions a different form for such social duties: they must be a matter of voluntary, moral decision, not legal, binding obligation. Liberalism does not reject practices of aid and support so much as the idea of a ‘right to support’ that affirms the rights of the poor with respect to the social obligations of the wealthy.14
Although seemingly insignificant, this distinction proves decisive because it results from a strict separation between law and morality that defines liberalism and is indispensable for its functioning. For liberals, civil society precedes political society, and the two are not the same. Positive law only confirms the principles of natural law and lends them binding force. The law is not the artificial work of human hands, but retraces the foundations of natural morality, giving it explicit and codified form. There is no law without morality; instead, morality underwrites the law. Nor does liberalism limit social obligations to those subject to legal sanction. Because the law has the sole function of securing the coexistence of liberties, liberal reason makes only one social obligation the object of legal sanction: not to curtail the freedom of others. All other obligations are banished from the negativity of the law to the positive terrain of morality: ‘In the liberal perspective, the right to support represents a contradiction in terms’.15
The ‘liberal delimitation’16 that draws a line between law and morality replaces a confrontational legal claim – with rights on the one hand and corresponding obligations on the other – with a voluntary moral bond that unifies the very people whom economic interests hold apart. Liberal practices of charity, then, do not derive from legal duty, but from a moral imperative that fulfils a politically indispensable function of socialization. Charitable practices create a social tie between rich and poor, givers and takers, and replace an antagonistic relationship with a reciprocal bond. The lack of a legal anchor possesses further significance, too: if poverty is a moral disposition, then charitable practice must not be institutionally fixed; charity contains the principle of its own suspension: it must not guarantee services but should make itself unnecessary. In other words, charity must not be a right – otherwise, the poor cannot become full legal subjects again.17
The liberal objectivation of behaviour in terms of responsibility rests on an important epistemological precondition: establishing a strict, practical connection between work and poverty. Assigning poverty to the individual is closely tied to an economic doctrine that assimilates poverty to work: if individual misconduct alone leads to being poor, then, conversely, wanting to work must necessarily represent the way out. The strategy of moralization can only function inasmuch as it forges a negative bond between poverty and work.
Studies by Giovanna Procacci and Jacques Donzelot point to the historical failure of such a conception. A central paradox defines the liberal treatment of poverty: liberalism must presuppose poverty in the same measure that it claims to wish to abolish it. Poverty simultaneously serves as the basis of, and motor for, moral and economic progress that pretends to aim at making it disappear.18
From the liberal perspective, poverty results from a lack of willingness to work. As such, it cannot find a basis for any right to support. Indeed, any social policy seeking to elevate the standard of living for poor segments of the population proves unthinkable, since this would mean keeping the poor from doing what would eliminate their poverty: working. Moreover, legislating social policy is not only wrong in economic terms; it is also immoral, since it leaves the poor in a state of dependency instead of eliminating poverty itself.
At the same time, poverty must not simply disappear, since it performs a key function within liberal governmentality and possesses economic and moral significance.
Poverty is economically necessary. Instead of representing a space beyond economic calculation, it is built into the metabolism of production. The variously adumbrated distinctions between ‘real’ and ‘false’ poverty make it possible to integrate the poor who are fit for work into the labour market and thereby ensure a cheap workforce. And poverty is economical in a direct sense, too: it provides the counterpart of, and basis for, the ‘wealth of nations’ – the precondition and vector of economic prosperity. Poverty names unsatisfied – indeed, as yet unknown – needs, and it symbolizes a market without frontiers.
Poverty is also morally indispensable: its role is to serve the ‘pedagogy of freedom’ and constantly illustrate what the wrong use of liberty entails. As the stigma of moral error, poverty must remain present in order to mobilize people’s better qualities. As such, it performs an important economic function and, at the same time, serves to encourage moral perfection.19
Thus, the view of poverty is not to be separated from the strategic significance of work within the overall problematic of liberalism. Not only does it have a negative connection to poverty, it also has a constitutive relationship with wealth, functioning as a hinge between them. Work represents the magic key for solving the problem of poverty. Work and wealth form a progressive circle; each one points to the other, and they reinforce each other mutually. In this light, the solution to the problem of poverty involves eliminating all obstacles preventing access to a free labour market, such that the impoverished population may be integrated into the cycle of production and all members of society come to be property owners.20
That said, nineteenth-century events showed that such trust in the wonders of work was unwarranted. Liberalism’s promise to dissolve poverty into work did not go into fulfilment. Not only did industrial capitalism and the free market not absorb old forms of poverty, they created new, hitherto unknown forms of impoverishment, different in intensity, scope and distribution. ‘Pauperism’ – mass poverty – took the place of poverty as an individual destiny. Working proved unable to realize the hopes placed in it; it did not function as a general principle of order, but became a factor of disorder. Instead of solving the question of poverty, the liberal coupling of poverty and work itself turned into a problem.21
This problem was not just a matter of the gap between formal equality and material inequality, that is, the difference between legal and social reality. The difficulty did not follow from the inequality of the poor so much as their equality. The economic and political rules of the liberal regime turned the poor into a new kind of subject – one that could not be dealt with through old-fashioned paternalism or exclusion from society. These morally and legally autonomous subjects could not be managed by either patriarchal or repressive means: although poor, they were equals among equals. This combination of politico-legal equality and economic inequality could only function so long as the latter could still be viewed as the result of individual behaviour and on the basis of theoretical equality.22
Mass impoverishment led such an equation to fall apart more and more. If the close connection between poverty and work was constitutive for liberal thinking, yet the free labour market proved incapable of eliminating poverty – and, indeed, brought forth more of it – then a problem already held on the level of formal equality: inasmuch as work serves as a mechanism for absorbing poverty, what is more self-evident than calling for a right to work in order to get rid of poverty once and for all? Yet such a right contradicts the right to ownership (of the means of production). It demands, for its material implementation, that the state intervene. In other words, it brings forth the very danger that liberalism means to counter in the first place.23
In the course of the nineteenth century, the close bond between poverty and work, which once had defined the liberal art of government, underwent relativization again and again. To counter revolutionary demands that society be radically reorganized, it was urgent to separate the question of poverty from legal demands for work and to view things in a manner that avoided reference to politics and economy. The original, economico-legal interpretation of the connection between poverty and work was modified by working new elements into the equation: on the one hand, by intensifying and reformulating moral categories, and, on the other, by taking up medico-hygienic considerations. This theoretical elaboration made it possible to steer mass poverty away from the economic system. Indeed: the new line of interpretation finally managed to transform the potential for disorder and destabilization into a means of achieving order.24
Liberal security politics remained indebted to the calculus of responsibility until the end of the 1800s, but political rationality about managing poverty had already changed during the first half of the nineteenth century. The studies by Foucault’s associates discussed in the following concentrate on two elements that signalled a corrective and complement to the strategy of classic liberalism: social economy and paternalism.25
The emergence of mass poverty following broad-scale migration from the countryside and industrialization gave rise to social economy, in contradistinction to political economy. During the early nineteenth century, Malthus in Great Britain and Sismondi in France had already responded to political economy’s manifest failure to explain how poverty and free markets could exist simultaneously. Social economy expanded the moral analysis of poverty’s causes with a further dimension: ‘milieu’. Social economy’s achievement, in operative terms, is to distinguish between reasons for, and conditions of, poverty. It continues to fault the carelessness of workers, but this general disposition can meet with more or less favourable conditions; it can be augmented or lessened.26
That said, it was not just a matter of making additions to standing analyses. The discourse of social economy also presented a new view of the poor’s recklessness: a matter of variable disposition admitting correction came to represent an immutable trait. From now on, (self-) neglect defined workers’ social identity. Individual pedagogics gave way to ‘social pathology’.
This shift in the objectivation of workers meant changing social security practices. If a lack of circumspection and responsibility characterizes workers in general, then charity cannot occur in passing; it must be instituted for good. At the same time, it must be assured that the institution not stand subject to political regulation or legal fixation; it should remain a ‘private’ matter. The liberal response to the problem is ‘patronage’. Its significance lies in the break it introduces in the continuum extending from formal equality, on the one hand, to actual inequality, on the other. A symmetrical conception of rights yields to an asymmetrical order: assigning comprehensive obligations to the employer – the ‘patron’ – and denying the possibility of free will to employees. Radicalizing the principle of responsibility, on one end of the spectrum, corresponds to dismantling it, at the other end.
Inasmuch as they counted as ‘irresponsible’, workers were not equal to their employers, nor could they be. On the contrary, they required guardianship and surveillance on the part of a ‘patron’, whose economic potency also represented a moral value. In order to fulfil the need for security in the political sphere while meeting the call for private regulation, employers became public figures functioning as ‘mainstays of authority’ in the general framework of ‘paternalism’: bearers of a public mission that unfolds optimally when political government promotes private initiatives. In consequence, legislative and juridical authority shifted, blurring the traditional liberal distinction between the social-private and the state-public spheres. The dividing line between public and private interest vanished. Under paternalism, economy and politics fused; as a result, the politics of social security met up with the core problem of economics: production.27
According to François Ewald, ‘patronage’s ‘politics of production’ managed human beings on the industrial model.28 Such practices displayed two complementary aspects. First, the insecure position of employees was exploited to ensure their dependency (through workers’ colonies, childcare, savings banks, and so on). Second, it was not enough for workers to be tied to the company: they had to be made productive (through education, discipline, instilling good habits, and so on). In other words, the institution of patronage combined the demands of the economy and political domination in a technology that made both terms reinforce each other. Its politics of security aimed at exploitation and subjugation simultaneously. The task was to transform an abstract wage-relation into a concrete, individualized form that would reinforce the standing political and economic order.29
The patronage economy points to a philosophy of social distinction and hierarchy miles apart from classical liberalism. The employer must ensure the reproduction of the workforce as a race – instead of promoting workers’ emancipation through savings and property (however modest). This heralds a change in the programme of security politics, too. It is no longer an individual matter, but the duty of one class towards another. Class division becomes an elementary component of liberal government. Charity, which once obeyed the law of brotherhood, now follows a logic of interdependence; in the process, the formation of classes becomes an important part of liberalism’s strategy for security. If the events of 1848 and 1871 exposed the danger of the undisciplined and pauperized urban classes, then integrating the working class into the political body and the emergence of a ‘proletariat’ represented an essential element of class struggle on the part of the bourgeoisie.30
Constituting class in this way required a central line of division within the spectrum of those without property, separating an economically determined proletariat from a socially determined plebs. According to Giovanna Procacci, social economy ‘invented’ a new social figure: the ‘pauper’. In this discursive context, pauperism is not defined by economic categories; it does not refer to a certain level or intensity of poverty so much as a form of destitution that poses a social danger. Pauperism names the spectrum of the mob, a collective and mainly urban phenomenon: an element of the population threatening it from within – calling forth a sense of fluidity, indeterminacy and so on. In this manner, social economy managed to distinguish between pauperism and poverty. The separation enabled the elimination of the former to be the corollary of preserving the latter.
From this perspective, poverty represents a social fact of capitalist, industrial society, whereas pauperism represents a hyper-natural form of life that is ultimately primitive. The logic may be summed up as follows:
On the basis of an analysis of the instinctive antisocial tendencies of the individual, society comes to be presented as inevitable restraint: freedom and equality, innate tendencies which can find expression in their pure state only in ‘savage’ society, and there encounter only natural limits and obstacles, are unavoidably frustrated and repressed in civilized society. … Thus, if it is true that humanity is spontaneously social, this means that it tends instinctively towards an uncivilized society based on natural appetites.31
For social economy, the ever-present danger that the social bond will dissolve and lead back to the state of nature is especially present in classes where poverty, ignorance and isolation prevail. All the same, poverty must not disappear: inborn tendencies should not be repressed, for they are the precondition for producing wealth; instead, they should be channelled and pacified by means of a ‘social regime’. The concept of pauperism does not aim at eliminating inequality, but at eliminating difference. Pauperism is ‘dangerous’ because it represents ways of living that do not fit into the project of socialization. The centrepiece of this strategy involves creating a subject different from the productive and economic subject: a subject aware of his or her duties and responsibilities, a civil (bürgerliches) and political subject. It does not mean combating poverty – which bears the stigma of inequality – but pauperism: a complex of behavioural characteristics underscoring social differences.32
In the discourse of social economy, pauperism does not arise from an economic structure, but a moral disposition. The strategy is to define poverty as ‘the fault of the poor’, which makes it possible to moralize society as a whole by means of techniques for managing poverty. In consequence, the problem of inequality in a society of equals is depoliticized: inequality no longer results from a different use of liberty; instead, it represents a difference in sociability. This strategy aims to depoliticize conflict by establishing a ‘normal’ relationship between wealth and poverty.33
When early liberalism sought to define the category of responsibility, the strict separation between law and morality played a role that was just as important as the principle of regulating social processes. Although paternalism made far-reaching changes to classically liberal designs and introduced a central asymmetry to the relationship between employers and employees, it still rested on the principle of responsibility and the distinction between law and morality. In the second half of the nineteenth century, a development occurred that broke fundamentally with such preconditions.
Over the course of the nineteenth century, liberalism’s constitutive distinction between law and morality was abandoned time and again. Hereby, the nature of the law and rights also changed. Law became normative: it no longer made prohibitions, but prescribed what should be done. It incorporated more and more ideas that had been developed in the context of the human sciences; conversely, the norms of the human sciences acquired a legal character.34 Law was no longer limited to averting danger and providing protection in order to coordinate and secure individual liberties; it became social in nature, suspending the notion of individual responsibility and legal accountability – which now were replaced with ‘social responsibility’. Legal claims and moral determinations of guilt came to be voiced less against individuals than directed towards ‘society as a whole’. The mode of regulation switched from the principle of individual responsibility to that of social risk.
At the beginning of the nineteenth century, ‘accountability’ and ‘guilt’ had formed the central elements of a well-ordered and harmonious society. In turn, the rise of industrial capitalism and its economic and social consequences challenged the idea of collective regulation via the principle of individual responsibility. The renunciation of classically liberal theory seemed to imply that a socialist-revolutionary solution was a viable option – or at least that state intervention in economic processes should occur. Such a course was precisely what bourgeois-conservative forces wanted to avoid.35
The way out of this dilemma – as François Ewald and Daniel Defert have shown – involved applying a technology, initially developed and tested in the private sector, to the regulation of society: insurance technology. Breaking with both the legal category and the principle of responsibility, it opened a path beyond the dead end of antagonism between individual and state. The insurance system replaced the confrontational, legal claims of particulars with a social bond tying everyone together. It did not address determinate social classes, but the continuum of a population varying in levels of risk according to age, sex, profession and the like. Such risks do not constitute class borders so much as they overlap with them. Instead of placing labour and capital in opposition, insurance protects both groups. As such, insurance technology does not represent the first step towards socialism so much as it guards against it: ‘Social insurance is also an insurance against revolutions’.36
In light of the problem posed by industrial accidents, François Ewald has examined how the dispositive of responsibility was rearranged along lines of social risk. The accident has a function similar to poverty for liberal rationality. Both represent natural misfortunes that form an indispensable component of the liberal universe. Like poverty, the accident – which is possible anywhere and at any time – symbolizes the principle of equality. Inasmuch as it necessitates preventive measures, it serves as a motor for individual perfection and moral progress. According to liberal logic, accidents justify claims for support just as little as poverty does; instead, they represent a possibility that must always be taken into account, against which provisions are to be made. As an event, the accident cannot be blamed on anyone; thus, damages may be demanded only in the case of manifest guilt. From this perspective, accidents and poverty are matters of individual fate and not social phenomena.37
That said, this mechanism for evaluating and assigning guilt proved problematic when a certain type of accident occurred. The reality of industrial accidents made the limitations of individual responsibility clearer than ever. Liberal thinking had been certain that guilty action represents the exception; in contrafactual terms, accidents do not happen when individual, rational actions occur. As industrialization proceeded, however, it became clear that industrial societies cause accidents as a rule: harm results from everyday operations, not just from instances of malfunction. This circumstance led to a problem within liberal law: social misfortunes justifiably called for financial damages, yet no one could be blamed, strictly speaking. Individual accountability was the precondition of, and basis for, liability, since no material compensation could be made without individual fault. But industrial accidents are not defined by a subjective quality; on the contrary, they display a particular ‘objectivity’ in two respects:
1. The first lies in the regular nature of modern accidents. Industrial accidents are not exceptional events so much as normal occurrences, and they follow statistical patterns. They are defined not by being unusual, but by their empirical regularity.
2. The other trait of industrial accidents is that they are a product of life in community – more the result of human interdependency than the consequence of interactions between people and machines. Industrial accidents are the ‘price of progress’: what regularly happens in activities that ultimately redound to universal benefit.38
In light of the ‘objective quality’ of industrial accidents, subjective accountability increasingly came to count as inadequate. The interdependency of actions escaping individual volition moved to the fore. Whereas liberal logic had been based on personal liability – for better or worse – the second half of the nineteenth century witnessed the rise of a perspective that invalidated this position by focusing on social distribution. Society and its obligations – and not the individual and personal responsibility – became the point of reference. The matter concerned the distribution of goods and ills following a distinct mode of rationality that derived neither from human misconduct nor from natural coincidence (or divine providence).
This social logic meant completely rearranging the liberal conception of justice – which held that natural distribution is just per se. The liberal idea of natural justice is closely tied to the principle of individual responsibility; as such, there is no need for socio-legal correctives. But inasmuch as individual responsibility is suspended, natural justice proves obsolete. In the liberal world, inequalities admitted justification – if only to the extent that the wealthy had acquired their possessions individually and the poor bore personal responsibility for their condition. But when the distribution of ills came to be viewed in relative independence from the good or bad behaviour of particulars and now appeared to follow social patterns, the liberal principle of justice posed a problem:
At the beginning of the nineteenth century, nature had still served as the point of reference for the question of justice: the ‘natural’ arrangement of goods and ills counted as just. There was nothing to correct, except in a case that counted as unusual: namely, when misfortune suffered by one person was occasioned by the fault of another. But now, justice does not appeal to nature so much as society: it is objectivated as a fact; as the totality of collective goods, it provides the basis for distributing these same goods among individuals. Law becomes social and corrective: the task is to reestablish balances that have been troubled, to reduce inequalities in the distribution of social burdens. The concept of risk – which expresses the singular reality of an ill prevailing independent of individual conduct, in keeping with ‘the way things are’ – and, more generally, the technology of insurance makes it possible to conceive and organize this new challenge in juridical terms.39
Insurance technology proceeds by devaluating the principle of individual responsibility and transforming the law at a profound level. As Ewald shows, legal mechanisms are increasingly ‘colonized’ by a calculus of risk that works with other criteria of regulation and evaluation; as such, it implements a different mode of individualization.
‘Juridical reason’ starts with the reality of accidents and seeks, in turn, to determine their cause. It aims to uncover (potentially) guilty conduct – which provides the sole basis for compensating damages. Calculus based on the notion of responsibility necessarily presumes that no accident would have happened without culpable behaviour. Juridical reason proceeds in keeping with a moral view of the world that represents the other side of the liberal model of harmony: the judge assumes there would not have been an accident, if those involved had not acted as they did – that the world would be orderly, if only human beings acted as they should.
In contrast, ‘insurantial reason’ adopts an entirely different outlook based on determination, and nothing else. Instead of starting out with a contrafactual assumption, it considers the facticity of events. The index is not subjective will, but objective probability: whatever we might wish, a precisely calculable number of workplace injuries, traffic accidents, deaths and illnesses occur annually, and these figures are repeated with remarkable regularity year on year. For juridical reason, the accident represents the exception in a world that counts as fundamentally harmonious. Insurance shows that the exception obeys a fixed rule and identifies empirical patterns.
By the same token, the difference between law and insurance is evident in the different forms of individualization they advance. The juridical scheme of responsibility sets the accident at a remove and separates victims from perpetrators. The accident involves two persons: it is a matter of one party’s fault vis-à-vis another. Insurance works with a wholly different scheme – one that does not involve confrontation with another individual but reciprocity in a collective. Members of an insurance society represent risk factors for each other: they pose dangers for others and, at the same time, stand exposed to the dangers that others pose. But for all that, the general principle of risk does not mean that every person is exposed to – or poses – the same risk. Disregarding concrete persons, insurance individualizes risk. It is no longer oriented on an abstract norm, but on the whole of the population: average individuality, on the basis of which concrete individual risks admit calculation. The contract of sovereignty, which is legally defined, stands opposed to the contract of insurance, which replaces legal categories with an index of social interdependency.40
The precondition for setting aside the law and discovering society as a specific object with its own, distinct rules was the emergence of a new form of knowledge: sociology.
Sociology objectivates society on the basis of principles fundamentally different from liberal ideas. Whereas liberals understood society as a (voluntary) association of individuals – with the whole comprising the sum of its parts, to which it can always be boiled down – society now comes to represent a reality ‘independent’ of its components. Sociological objectivation inverts the liberal relationship between sum and parts: society is not the association of individual subjects, but a subject itself. It is characterized by its own laws and patterns that impose themselves on individuals as a matter of alien and external fact.
The sociological objectivation of society provides two decisive elements to insurance technology: the constancy of probabilities and the concept of solidarity.41
Ewald reconstructs an important epistemological break that occurred in the human sciences during the nineteenth century. In the early 1800s, the idea still prevailed that human affairs follow universal laws. Irregularities in the social cosmos counted as exceptions to a general rule, the result of disruptions. Newtonian physics provided the model for the human sciences, and investigations of human phenomena employed methods borrowed from the natural sciences.
Adolphe Quetelet established his ‘social physics’ to demonstrate continuity between nature and society. He did so by linking statistics with probability calculation. Statistics had defined the population as a sum of individuals – a passive complex possessing no traits as such. Against this backdrop, probability permitted an entirely new conception of the relations between the whole and its parts, offering a way out from the dead end of the state and individual by means of a new logic different from both political rights and economic interests.42
Ewald enlists Quetelet’s studies to trace the emergence of a new method of investigation. Although the object of ‘social physics’ is still the human being, an epistemological shift occurs in order to examine the individual, now approached by way of a ‘detour’. For Quetelet, the individual comes into view only in light of the mass: one must understand people in vast numbers in order to know them individually. The art of probability calculus plays out non-knowledge against itself, thereby making it the basis for analysis. In this framework, ‘law’ and ‘cause’ no longer designate objective or effective laws in the Newtonian sense; instead, they stand for empirical regularity: probabilities, tendencies and inclinations.43
Probabilistic logic breaks with all notions of substance. Quetelet posits an ‘average human being’, but no corresponding individual actually exists; instead, the term refers to the type of person encountered in society. ‘He’ is not a model – the original – whose more or less exact copies are actual human beings, but the latters’ shared point of reference. Quetelet offers an objectivating scheme that does away with all metaphysical references to abiding human nature; instead, his definition proceeds from the immanence of ‘man’s’ social existence. This mode of analysis does not start from individual specimens, but from the group to which they belong. Sociology is an instrument for understanding a given population in terms of intrinsic features instead of external factors.44
The sociological mode of objectivation means that ‘human being’ no longer counts as a universal; instead of looking for a totality of qualities to be found in everyone, its definition holds in terms of social existence. The theory of the average human being offers a way to understand the register of identities and inequalities on the basis of real differences. At the same time, it shows how a collective identity is possible despite – indeed, through – these same differences. Sociology extracts a principle of cohesion from social divisions by replacing the antagonistic language of law with the homogeneous discourse of statistics.45
As Jacques Donzelot has shown, the incorporation of statistics and probability into examinations of social realities was not the only blow dealt to the sovereignty of the law. A second front opened, too. By the end of the nineteenth century, the concept of solidarity had spread far and wide. Especially thanks to the works of Émile Durkheim, its foremost theorist, it implicitly subjected to critique the validity of legal categories for analysing social phenomena.
In the course of the nineteenth century, the juridico-political conception of (popular) sovereignty – once thought to offer the key to solving political and social ills – came to pose more and more of a problem. Questions arose about the declared sovereignty of citizens at the polls and their factual subjugation in the factories. The idea of sovereignty did not constitute a factor of order, or a principle of coherence, so much as it produced division. On the one hand, the matter involved fears of despotism no longer based on a monarch’s whims but on those of the people – a tyranny of the majority. On the other, the notion of sovereignty extended an invitation to reclaim a form of rule that the ‘bourgeois’ republic had betrayed by excluding the economic sphere from processes of political will-formation. Under these conditions, the decisive question was the principle for steering governmental activity: what the state’s contribution should be, and where its borders lie.46
According to Donzelot, the ‘theory of solidarity’ provided an answer. He situates it in a field of tension with a liberal-conservative pole at one end and a revolutionary-Marxist pole at the other. Steering clear of rival alternatives – revolutionary fraternity or liberal charity – the ‘theory of solidarity’ that Durkheim presents in La Division du travail social points to a third way; in this framework, the individual is neither the starting point for society nor positioned against it. This theory offers the strategic ‘advantage’ of substituting ‘mute’ social constraint for ‘talkative’ political voluntarism. As such, society neither represents the product of free will, nor is it constituted by originary and natural associations. Society is not an institution serving the interest of particulars, nor can it be. Instead, society is a material state, where the prevailing patterns and rules always already exceed individual horizons inasmuch as they concern the lives of all its members.
For Durkheim, society defies instrumentalization because it follows its own laws. From the inception, its elementary forms have developed to yield increasingly ramified divisions of labour and complex modes of action. But in the process, social cohesion does not diminish; it switches from one form of solidarity to another. Underneath – if not counter to – legal contracts, mutual dependence grows as society evolves. Accordingly, ‘organic solidarity’ proves something more basic than formal arrangements between members of society; the law represents a derivative category. In brief: the fundamental law of society is not legal in nature; not sovereignty, but solidarity constitutes the social.47
The concept of solidarity made it possible to replace calls for sovereignty with belief in progress; indeed, sovereignty is not only dangerous, but illusory inasmuch as it fails to account for the many-layered reality of social relations. As such, the sphere where sovereignty applies and intervenes had to be restricted. To preserve society from the dead end of sovereign whims, choosing rulers must be separated from governmental practices. Political sovereignty was redefined strictly in terms of selecting those who would hold power; concrete governance turned into a technical affair limited to questions about the means for achieving social advances.48
Simultaneously, the notion of solidarity has the merit of specifying the need for, and limits of, the state. It provides criteria that are compatible with a democratic definition of the state and, at the same time, resolves the grievous ambiguities of the sovereignty concept on which both liberal and socialist solutions rely. On the one hand, solidarity founds the social role of the state; on the other, it imposes limits on intervention by obligating it to respect individual liberties and initiatives, which are necessary for overall social progress. As such, solidarity is not the first step towards socialism, but the best means to avoid the subversive reorganization of society.49
Studies of ‘insurance society’ make the theoretical profile of an alalytics of government clear. The social is not to be viewed only in legal or military categories. Contractual relations do not constitute it, nor does it represent the fragile and constantly imperilled result of warlike confrontations. Rather, the ‘invention of the social’ goes along with the emergence of new regulatory techniques. According to the works discussed here, modes of analysis limited to taking note of ‘crises’ and ‘contradictions’ and/or viewing the social state in terms of expanded rights do not go far enough.
First, whether the establishment of social rights represents a crisis of classic liberal governmentality or stands in opposition to it, ‘society’ constitutes itself in spite – if not, in fact, by way – of such crisis and contradiction. On the one hand, one may rightly speak of the social state as the ‘victory’ of the working class. Still, it is not enough to view social measures and their legal consolidation simply as the result of political struggles, for the former also change the conditions and stakes of the latter:
The point is not to deny any connection between these measures and political struggles, but to detect what there is about them that cannot simply be traced back to the simple logic of collision, victory or defeat – to find out what, specifically, these measures introduce to the field of social relations.50
The social state represents the result of social struggles, yet its strategic significance extends in the opposite direction: an active ‘politics of depoliticization’ which is supposed to make these very struggles unnecessary. The social does not stand as the product or instrument of the political so much as it entails a complete change of this sphere. In historical terms, the emergence of the social proceeded as the gradual subsumption of the political, silencing the struggles here – rubrics such as ‘victory’ or ‘defeat’ came to seem increasingly inadequate.
Second, by the same token, social insurance does not concern a simple, quantitative increase of rights – adding new (social) rights to old (liberal) rights. François Ewald, in particular, has demonstrated that the order of social rights changed the very nature of the law:
Liberal rationality objectivated the legal subject as freedom: the famous principle of free will. In the order of social rights, the subject becomes juridically competent by the simple fact of being alive. If social rights are indeterminate in principle, they can never be anything but so many expressions of a fundamental right to life. … The liberal state should secure the possibility of exercising civil rights; it is supposed to demand nothing more of citizens than what is necessary for the common interest. If one can demand social rights from society, in turn society can impose forms of conduct it deems socially appropriate.51
Indeed, the ‘government of the social’ operates by institutionalizing two steps that seem incompatible yet stand in a complementary relation. The corollary of protecting the individual from society is ‘defend[ing] society’52 from the individual who threatens it. At its core, the social draws a dividing line between two forms of population and two ways of treating them: the normal on the one hand, and the abnormal on the other. Its invention cannot be separated from discovering ‘perverse’, ‘degenerate’ and ‘extremist’ elements whose existence depends on a norm that defines them and a rule to which they provide the exception.53
The ‘defence of society’ represents the flipside of ‘insurance society’, the result of a changed scheme of objectivation. Inasmuch as society no longer emerges from the lives of individuals, but stands at their source, it is a subject that must defend itself. The means for doing so is a particular kind of anthropology, which simultaneously offers a symptomatology, pathology and therapeutic method. Such ‘social anthropology’ no longer takes the individual subject as its reference, but society; subjects are no longer measured with regard to human nature, but by their degree of sociability – their ‘second nature’. Inadequately socialized individuals not only count as unreliable and unpredictable loners, but pose a danger to the continued existence of society.54
In his 1976 lecture series, Foucault examines how politico-military discourse got transformed into a racist-biological discourse.55 From the seventeenth century on, the model of war provided the matrix for describing the course of history and the reality of power relations. It did not take long for the expression ‘race’ to surface in this ‘politico-historical discourse’. That said, the term did not have a fixed, biologico-social meaning yet; initially, it designated a historico-political division: the idea that society splits into warring camps: two antagonistic social groups exist and live alongside each other without having mixed, since they differ through geographical provenance, language and/or religious customs (see the section ‘The War Hypothesis’ in Chapter 6).
The particularity of such ‘racist discourse’ is to view political relations in military-martial terms alone. Accordingly, the political bond tying the two groups to each other is arbitrary; their connection is not a matter of social contract and rational encounters, but the result of military force. War – victories won or defeats suffered – structures this political unit. They constitute a whole ‘because of the differences, dissymmetries, and barriers created by privileges, customs and rights, the distribution of wealth, or the way in which power is exercised’.56 As such, this discourse of racial war must always start from a plurality of races, not a single one: one race is defined in opposition to another; antagonism constitutes it. Race theory is relational, then; the point of departure is not given biologically or socially; it represents a power relationship that provides the basis for biological or social distinctions in the first place.57
In the course of the nineteenth century, Foucault contends, two important ‘transcriptions’58 of ‘race’ and ‘racism’ occurred, changing a semantic field that was initially political. First, talk of the ‘war of races’ underwent an ‘openly biological transcription’59 that, even before Darwin, based its claims on materialist anatomy and physiology. This historico-biological race theory coded social conflicts as ‘struggles for survival’ and analysed them along the lines of an evolutionary scheme. The second (social) transformation was to understand the ‘clash between two races’ as class struggle and analyse it as a dialectic. The beginning of the nineteenth century witnessed the emergence of revolutionary speech that increasingly replaced the race defined in political terms with the theme of economico-social classes.60
Foucault’s thesis holds that these ‘reformulations’ of the problematic of racial struggle – which was originally political – factored into an evolving biologico-social discourse from the end of the nineteenth century on. Such racism (the first time the word held the meaning it has today) enlisted elements of its biological version to formulate a response to the challenge of social revolution. What was historically novel and theoretically innovative about such biologico-social discourse is that it turned the original idea of social war against itself:
What we see as a polarity, as a binary rift within society, is not a clash between two distinct races. It is the splitting of a single race into a superrace and a subrace. To put it a different way, it is the reappearance, within a single race, of the past of that race. In a word, the obverse and the underside of the race reappears within it.
… The discourse of race struggle – which … was essentially an instrument used in the struggles waged by decentred camps – will be recentred and will become the discourse of power itself. It will become the discourse of a battle that has to be waged not between races, but by a race that is portrayed as the one true race, the race that holds power and is entitled to define the norm, and against those who deviate from that norm, against those who pose a threat to the biological heritage.61
The problem of a plurality of races shifts to the singular of race – which is no longer threatened from without, but from within. The notion of a binary society divided into two races yields to the idea of a monistic, biological entity menaced by heterogeneous elements. At the same time, this heterogeneity no longer structures a basic line of separation running through society. Instead, it establishes distinctions that are matters of accident more than substance: ‘foreigners’, ‘deviants’, ‘others’ – i.e. the (side-) products of society itself – are what threaten it. Society brings forth, from within, the dangers that constantly imperil it. The process gives rise to ‘a racism that society will direct against itself, against its own elements and its own products’.62
As the old theme of racial war evolved, it tended towards ‘biologicization of the political’.63 The historico-political theme of war – with its battles, victories and defeats – gave way to an evolutionary-biological model of the struggle for survival. It was no longer a matter of struggle in a political sense; the problematic of struggle assumed an exclusively biological meaning: the differentiation of species, survival of the fittest, preservation through adaptation, the danger of degeneration and so on.64
On this basis, Foucault concludes that the racism that emerged at the end of the nineteenth century did not form part of a more comprehensive strategy of domination, which might then be implemented and instrumentalized, so much as it afforded a new ‘technology of power’:65
It is no longer: ‘We have to defend ourselves against society’, but ‘We have to defend society against all the biological threats posed by the other race, the subrace, the counterrace that we are, despite ourselves, bringing into existence’. At this point, the racist thematic is no longer a moment in the struggle between one social group and another; it will promote the global strategy of social conservatisms.66
In other words, Foucault views racism as a form of ‘government of the social’ – a distinct technology of government enabling a biological entity to be steered by seemingly neutral means, beyond politico-legal and economic criteria. Racism defines a political strategy of depoliticizing and dedramatizing social conflicts by pinning them to the natural world – i.e. the laws and constraints that prevail there. Just as living in nature demands respect for certain exigencies and acknowledging that radical change is pointless, so too does life in society: who would presume to change the laws of nature and not come to realize that doing so leads to self-destruction?
To be sure, none of this means that a political element is missing from the analysis. However, the politics that follow abandon rigid adherence to received structures and voluntaristic impulses for seeking change. Conceiving society as a biological whole presupposes the institution of a centralized instance to steer, lead and guide it in order to guarantee its ‘purity’ – one strong enough to counter ‘hostile elements’ from within and without: the modern state. For Foucault, racism guides the rationality of state actions, materializes in its apparatuses and concrete policies, and finds articulation as ‘State racism’.67 Not only does racism form a central, functional element within power relations; it comes to be the principle organizing the state. The historico-political discourse of races had concerned the state and its apparatuses, which it denounced as instruments one group used to dominate others, and the state’s laws, whose partisan nature it unmasked. In contrast, the discourse of race represents a weapon in the hands of state sovereignty – which once had been the target of this same weapon:
The theme of the counterhistory of races was … that the State was necessarily unjust. It is now inverted into its opposite: the State is no longer an instrument that one race uses against another: the State is, and must be, the protector of the integrity, the superiority, and the purity of the race. The idea of racial purity, with all its monistic, Statist, and biological implications: … racism is born at the point when the theme of racial purity replaces that of race struggle.68
The reorientation in civil law away from the liberal principle of responsibility towards the concept of social risk corresponds to a radical transformation of penal law. Foucault’s ‘hypothesis’69 is that the insurance model for social relations gave rise to the idea of punishment, serving to ‘defend society’ and preserve it from the inevitable risks attending life in community. In the framework of the new penal rationality, sentencing received a new meaning. It no longer sought to punish a legal subject who had broken the law as a matter of personal initiative or to reintegrate this party into the law-abiding community; instead, it aimed at reducing, as much as possible, the risk of criminality that the individual posed and, if need be, ‘neutralizing’ him.70
Such a goal presupposes another mode of objectivating the lawbreaker. In this framework, the lawbreaker is not a universal legal subject of the law, but a human being – albeit a particular kind of human being defined by inhumanity: a ‘criminal’. The criminal is not a rational entity whose life follows a course determined by free will, but a person whose psychic and moral constitution is abnormal. And because the ‘criminal’ is not normal, it makes no sense to look for his motives, interests or intentions. The reason for lawbreaking does not stem from (mis)calculation on the part of the perpetrator; it follows from a pathological and defective personality. Homo criminalis represents a new type, wholly different from the homo penalis of classical criminal justice.71
Changing the way the lawbreaker is objectivated presupposes a new organization of knowledge. Pasquale Pasquino has reconstructed this scheme by juxtaposing liberal legal theory and ‘social legal theory’. Liberal penology is based on the principle of free will – a trait possessed by every legal subject. As such, it cannot constitute the object of specific knowledge (‘criminology’). ‘General anthropology’ accounts for the deeds of homo penalis: ‘he’ is no different from homo œconomicus inasmuch as the actions of both follow rational calculations. Anyone can commit a crime – the postulate of freedom includes freedom to break the law, too. According to the penal rationality of liberalism, homo penalis dwells within all of us as a latent potential; violations of the law represent ‘his’ realization. Because human action is rationally motivated and follows calculations of pleasure and pain, punishment must aim at the will of potential perpetrators. It operates by exercising an effect on the mind, calibrating crime and punishment; calculating the punishment risked deters the subject from carrying out the misdeed. Finding the appropriate sanction for a given violation involves identifying a disadvantage that will strip misconduct of its appeal once and for all. This theory of punishment is dynamic; it works with the principle of balance to neutralize interest in breaking the law.72
In the latter half of the nineteenth century, this form of penal rationality garnered more and more criticism. For both practical and theoretical reasons, the ‘school of social law’ declared that cancelling out criminal volition was outdated. Pasquino sums up the argument:
Beccaria’s theory of punishment as an instrument ad deterrendum counterweighting the interest in the committing of crimes is false, both theoretically and practically: practically, because the statistics of crimes and criminals simply continue to rise; theoretically, because the criminal does not think like a normal and honest person such as Beccaria – indeed, we may say that he or she does not think at all. The criminal cannot be a homo penalis because he is not a Man.73
According to the line of reasoning of the ‘school of social law’, the actions of ‘criminals’ do not result from the general nature of human beings (who are prone to miscalculation), but express a specific kind of evil. The ‘criminal’ does not qualify as homo penalis precisely because he is not a rational human being. The figure of homo criminalis takes the stage, shifting penal rationality towards ‘social defence’; here, the axis of culpabilization no longer involves right or wrong calculations, but conformity or non-conformity with regard to social life. The new legal theory is not based on lawbreaking as an anthropological or psychological fact, but on understanding the lawbreaker as part of a separate species and ‘social race’.74
Pasquino shows how the invention of this new ‘juridical figure’ changed the justificatory principle of punishment and the nature of the law:
1. Nature of the law. In the new framework, law no longer precedes society (defined as the direct expression of members’ free wills). It becomes a variable, secondary category. The law does not constitute the rules of communal life; it simply codifies them. The theory of the social contract as the legal foundation of society is inverted: henceforth, law displays a derivative quality – as one manifestation, among others, of a historically mutable society. It is impossible to found society on law, because every human group in history brings forth its own rules; society is not the result of, but the basis for, the law.
2. Justificatory principle of punishment. Punishment’s basis and point of reference shift from the rights of subjects to the norms of society. Inasmuch as law must adjust to evolutionary dictates, society provides the sole instance for legitimating punishment. Penal laws represent mutable codifications of a given society’s vital interests. The principle of immanence prevails. Law and rights are no longer measured by some standard of natural justice – so that the positivity of their rules will conform to its prescriptions. Instead, the point is to correct the injustices of nature. Law becomes interventionistic and serves as a social corrective; in contrast, nature offers a negative point of reference.
When ‘defending society’ stands at issue, the operative principles are self-referential; society provides not only the source of the law, but of the lawbreaker, as well. Criminal law and criminals stand locked together in circular fashion. Penal law, which serves society, reacts to a threat that emerges from society itself: society is constantly producing the same antisocial tendencies that threaten its existence and demand corrective intervention. Just as each society brings forth its own, historically specific culture, it brings forth its own kind of criminals – and the law corresponding to them.75
For all that, nature does not simply vanish. If anything, society absorbs and incorporates nature. Simultaneously, the ‘society’ that law is implemented to defend displays its own, ‘social naturalness’ differing markedly from ‘natural naturalness’. Pasquino remarks the central paradox of the school of social law: ‘if it is claimed that criminality is a phenomenon with a social aetiology, how is it possible to say that the criminal has an asocial nature?’.76
For Foucault, modern racism offers a response to this very question; it dissolves the paradox of the social origin of antisocial conduct. The ‘solution’ that racism represents entails, first, making divisions within the social sphere as a biological continuum (that is, a given population, or the human species as a whole) enabling underclasses to be distinguished and arranged in a hierarchy along racial lines. This fragmentation of the biological field enables qualifications of good and bad, higher and lower, and ascendant and descending races; at the same time, a line of division ‘between what must live and what must die’77 is established. The splitting and differentiation of the biological sphere serves to define the general realm of operation for ‘biopolitics’ and structure the modes of its concrete interventions.78
Still, racism is not content to draw a line separating two sides; it also provides a new principle of articulation between these realms. The point of departure is a further reformulation of the traditional theme of racial war. Racism abandons the old, negative relation between the life of some and the death of others (‘If you want to live, the other must die’) and makes the terms function in an unprecedented manner. The core of this relation is no longer either-or; instead, racism produces a positive and productive relationship between a ‘more’ and a ‘less’ (‘The more you kill, the more death you cause, the more you will live in turn’).79
And with that, racism makes it possible to establish a relationship between the life of some and the death of others that is not military-political, but biologico-medical. It establishes an organic connection between the fact of ‘degeneration’ and the health of the species: to the extent that inferior races vanish and abnormal individuals are eliminated, their share of the species as a whole diminishes and, in consequence, the population/species becomes stronger, more vigorous, more alive and so on:
This is not, then, a military, warlike, or political relationship, but a biological relationship. And the reason this mechanism can come into play is that the enemies who have to be done away with are not adversaries in the political sense of the term; they are threats, either external or internal, to the population and for the population. In the biopower system, in other words, killing or the imperative to kill is acceptable only if it results not in a victory over political adversaries, but in the elimination of the biological threat to and the improvement of the species or race.80
For Foucault, the strategic achievement of racism is threefold: first, it enables breaks within a (biological) continuum; second, it formulates a positive relationship between death and life; third, it lays the foundation for constituting individuals and classes in terms of a specific race. Identifying a collective subject of ‘criminals’ presupposes that the individual is not a single case, but part of a social collectivity, a group of persons with the same traits. Racism ‘normalizes’ violations of the law by constructing lawbreakers as an abnormal component of society and bringing this component back into the social body as its internal border. In this framework, the lawbreaker no longer counts as a monster or creature standing outside of society, but an unnatural element of social nature.81
For penal law that prevailed into the nineteenth century, the central problem was whether an individual qualified as mad or a delinquent. Liberal law, based on the principle of a free will, set up an elementary division: a mentally ill person was not a delinquent; his deed was no crime, but the symptom of an illness. Condemning violations of the law presupposed legal subjects with a free will.
During the second half of the nineteenth century, this border became increasingly porous and the categories started to mix. Now, the delinquent individual was not only a lawbreaker, but also – and above all – a criminal characterized by a certain psychologico-moral disposition. More and more, lawbreakers came to count as mentally ill, and medico-psychiatric categories replaced the notion of free will:
So then, the old dichotomy in the Civil Code, which defined the subject being either delinquent or mad, is eliminated. As a result there remain two possibilities, being slightly sick and really delinquent, or being somewhat delinquent but really sick. The delinquent is unable to escape his pathology. … Pathology has become a general form of social regulation. There is no longer anything outside medicine.82
This change in penal law towards ‘defending society’ was made possible by way of a concept corresponding to the notion of risk in civil law: danger. The idea of liability without fault corresponds to that of accountability without freedom. The concept of danger displays immense flexibility. ‘Dangerousness’ is a paradoxical notion, since it simultaneously designates an abstract possibility and a concrete, psychic capacity: on the one hand, it affirms a certain quality particular to an individual; on the other, it symbolizes uncertainty, since proof of dangerousness can only be provided after the fact. It represents a hypothesis concerning the greater or lesser probability of connection between symptoms in the present and future action. As such, it is universally applicable: everything potentially poses a danger or a risk.83
And so, the category of danger could apply, beyond the category of criminals, to other marginal groups as well. What tied together material need, vagabondage, criminality, perversion and so on was the danger they represent for society. This concept served as a unified analytical instrument for various social problems and, at the same time, made it possible to treat dangerous individuals differently, in keeping with the intensity and nature of the menace they represented. Danger provided not only the instrument for diagnosing ‘antisocial tendencies’, but also the key to solving them, and opened an array of treatments: psychiatric therapy, medico-hygienic intervention, charitable or preventative detention and so on.84
The notion of dangerousness implies that society is constituted as a body with inherent dangers and maladies. The social significance of this medical complex derives from objectivating human life in community as a biological entity. The ‘normalization’ of society presupposes its permanent pathologization, making medicine into a ‘political intervention-technique’85 that serves the interests of control and regulation: ‘Modern medicine is a social medicine whose basis is a certain technology of the social body’.86
The eighteenth century had already witnessed a liberal ‘health politics’ (‘noso-politique’) of the population, which concerned the well-being of the social body in general and took distance from the ‘medical police’ of old. Health became the duty of the individual and the goal of intervention on the part of the state. An economic imperative stood behind the commandment to be healthy: physical health was important inasmuch as it furnished labour power, and state interest turned to the conditions for reproducing this resource. The capitalist economy relied on establishing a certain politics of the body to ensure that it remained productive. The biological data of the population came to represent an indispensable element for ‘economic regulation’ that necessitated, over and above the subjugation of bodies, the permanent improvement of their utility.87
The precondition for such liberal politics of health involved shifting the problematic of health with respect to other techniques of social support. Foucault’s thesis is that autonomizing poverty in economic terms corresponds to the development of a medico-administrative body of knowledge, at the centre of which stands the health of the individual and the welfare of the population. This medico-administrative complex has various aspects:
1. The family represents a decisive switching point for the medicalization of individuals. ‘Noso-politics’ produces the family–children complex, ordering the relationship between parents and their offspring in a new fashion. It is no longer enough to ensure a certain number of progeny; the children must be guided and supervised – they must have a certain physical constitution and demonstrate certain abilities. In this context, childhood comes to represent an independent period of life, separate from the world of adults and oriented towards an end goal. Raising children becomes an important goal of new policies, and the family the focus of medical and moral campaigns. As the hinge between the general goal of a healthy social body and care for individuals, it gives rise to a ‘private’ ethics of good health as a relationship of mutual obligation between parents and children.88
2. The second element of this ‘noso-politics’ is the privilege granted to hygiene – that is, the way medicine serves the purposes of ‘social hygiene’. Inasmuch as medicine elaborates certain prescriptions, it institutes rules for individuals to follow in order to ensure their own welfare; at the same time, this means creating the basis for medical intervention in social sectors where the rules are not respected. As such, medicine comes to play a more and more important role within a politics that seeks to investigate and improve the social body in order to keep it ‘healthy’. Foucault assumes that the development of medico-administrative knowledge, which ‘treats’ society in terms of health and sickness, represents the germinal condition for the emergence of sociology in the nineteenth century.89
Modern medicine constitutes a ‘biopolitical strategy’.90 Its operations are not limited to identifying and healing illness; above all, they aim at prophylaxis and prevention. Everything qualifies as a potential object of medicine insofar as anything may promote or imperil collective health. The field of medical intervention extends from water quality to nutrition, city planning and sexual practices – no longer restricted to the ill or illnesses, it is situated on the plane of ‘life’ itself. Medicine becomes an instrument of universal diagnosis and therapy; it proves indispensable to the extent that the ills it ‘treats’ constitute integral components of the naturalness of the social body – ills that can be combated only through improved, more precise and more ‘holistic’ medical knowledge:
What is diabolical about the present situation is that whenever we want to refer to a realm outside of medicine we find that it has already been medicalized. And when one wishes to object to medicine’s deficiencies, its drawbacks and its harmful effects, this is done in the name of a more complete, more refined and widespread medical knowledge.91