Freedom as critique
Foucault beyond anarchism

Abstract

Foucault’s theory of power and subjectification challenges common concepts of freedom in social philosophy and expands them through the concept of ‘freedom as critique’: Freedom can be defined as the capability to critically reflect upon one’s own subjectification, and the conditions of possibility for this critical capacity lie in political and social institutions. The article develops this concept through a critical discussion of the standard response by Foucault interpreters to the standard objection that Foucault’s thinking obscures freedom. The standard response interprets Foucault’s later works, especially The Subject and Power, as a solution to the problem of freedom. It is mistaken, because it conflates different concepts of freedom that are present in Foucault’s work. By differentiating these concepts, this article proposes a new institutionalist approach to solve the problem of freedom that breaks with the partly anarchist underpinnings of Foucault scholarship: As freedom as critique is not given, but itself a result of subjectification, it entails a demand for ‘modal robustness’ and must therefore be institutionalized. This approach helps to draw out the consequences of Foucault’s thinking on freedom for postfoundationalist democratic theory and the general social-philosophical discussion about freedom.

Michel Foucault is known as a philosopher who renewed our thinking of power. His theory of power also makes it possible to develop a new, complex understanding of freedom – by problematizing our social-philosophical thinking about freedom. According to Foucault’s concept of subjectification, power permeates subjects more thoroughly than it is usually described in social philosophy, because power constitutes subjects in the first place. This new thinking of power is also a critique of the common concepts of freedom in social philosophy that either understand freedom as independence from power or do not address the social constitution of subjectivity as a problem of power.

The three common concepts of freedom are negative, reflexive and social freedom.[1] The negative concept of freedom defines freedom as the absence of external interference or domination. But Foucault shows that heteronomy works inside the subject, and therefore there can be no free inner core of the subject, as is presupposed in negative concepts of freedom. The reflexive concept of freedom defines freedom as reason. Yet, according to Foucault, there is no universal reason independent of power, but thinking is always located historically and socially – at the heart of reason beats heteronomy. Finally, the social concept of freedom defines freedom through integration into a good society. But with Foucault, the dark side of integration becomes visible: a paternalistic normalization that leads to conformity. However, Foucault does not propose an alternative notion of freedom that would be compatible with and following from his theory of subjectification. In this article, I develop such a new concept of freedom: freedom as the capability to criticize one’s own subjectification, which is dependent on subjectifying institutions – in short, freedom as critique.

Freedom as critique expands the current debates on freedom by serving as a candidate alongside the concepts of negative, reflexive and social freedom that on the one hand includes plausible insights of these three approaches and corrects their blind spots on the other hand. With social freedom, it shares the view that subjects are socially constituted, that is, that their abilities depend on their socialization through recognition processes and that there is no pre-social core of the subject as assumed by negative freedom and no universal reason as assumed by reflexive freedom. With reflexive freedom, it shares the belief that ultimately only constant critical reflection helps in the process of working on freedom and that therefore the institutionalization of reflection processes is central. With negative freedom, freedom as critique shares the concern regarding the paternalizing effects of a normalization in the name of a theoretically or politically determined good, which can arise from a universalist conception of reason of reflexive freedom and the focus on the social integration of social freedom.[2] If one summarizes the three concepts of freedom in a step-by-step model of development and sublimation, freedom as critique can be understood as a fourth step, that is, as a reaction to the pessimistic, dark side of social freedom.[3] It emphasizes the costs and sufferings of what can be optimistically called socialization, and in this pessimism calls it subjectification. In contrast to all three concepts of freedom, freedom as critique makes it possible to conceive the ‘inner’ unfreedom of subjects as a political problem and thus to debunk the liberal ‘myth of the given’[4] without falling into illiberal, total politicization. Freedom as critique is thereby particularly suited as a normative clarification of postfoundational concepts of democracy.

The path to this new understanding of freedom is not a hermeneutic rereading of Foucault’s works but the critique of a particular branch of their reception: the second phase of social-philosophical debate about freedom in his works. In the first phase of this debate, the problem of freedom was articulated: Social philosophers regarded Foucault as one of their own and criticized him, consequentially, for being a poor social philosopher, in particular for describing subjects as so thoroughly determined by power that freedom and resistance were no longer conceivable. The most prominent authors who criticized Foucault in this way were Ju¨rgen Habermas, Axel Honneth, Nancy Fraser and Charles Taylor, among others.[5] The second phase, which culminated in the 2000s and continues to this day, consists of attempts to defend Foucault against the accusations raised and to solve the problem of freedom by explicating the notion of freedom Foucault puts forward.

While, in the existing literature, there are several systematically different approaches in defending Foucault,[6] which I reconstruct and criticize in Freiheit als Kritik. Sozialphilosophie nach Foucault, [7] one feature prevalent in most approaches is to interpret Foucault’s later works as a solution to the problem of freedom raised in the earlier works. Central to such claims is to refer to the later text The Subject and Power (1983). This is one of the few pieces where Foucault writes in an explicitly social-philosophical idiom in contrast to his normal method and style, the genealogical analysis. Foucault, in this text, states quite clearly that freedom is a presupposition of power. The supporters of Foucault take this stance as a solution to the problem of freedom. I call this interpretation the ‘standard response’, as it is common among social-philosophical Foucault interpreters nowadays.[8] My thesis is that it is mistaken, as it does not solve the right problem of freedom – because there are actually two different kinds of problems. I call these the ‘problem of power determination’ and the ‘problem of subjectification’. I will argue that the standard response only answers to the equally ‘standard objection’, that is, the problem of power determination and not the problem of subjectification. This differentiation already lays the grounds for a solution of the problem of subjectification: It is necessary to account for the ‘modal robustness’ of freedom as critique, that is, the capability to critically reflect upon one’s own subjectification. This entails to take the subjectivating effects of political institutions into account and thereby interpret Foucault not as an anarchist but as a political theorist. This institutionalist interpretation rather than the standard response can give a coherent account of Foucault’s thinking of freedom. The systematic explication of freedom as critique expands the usual concepts of freedom in social philosophy and serves as a normative foundation of pluralistic radical democracy in postfundationalism.

In what follows, I first present the standard objection and the standard response. Second, I elaborate on the problem of subjectification and explore what would be needed to solve it: an account of freedom as critique. Third, I explain that the standard response only accounts for freedom to act differently but not for freedom as critique. Freedom as critique demands a ‘modally robust’ explanation, which is absent in freedom to act differently. I conclude by sketching how the problem can be solved through the institutionalist reading that accounts for modal robustness and therefore is able to explain freedom as critique, and how it enriches postfoundationlist democratic theory as well as the social-philosophical debates on freedom in general.

1. Standard objection and response: ‘The subject and power’ as a solution to the problem of power-determination

In the archaeological phase, the problem of freedom is centred on language and thinking.[9] Foucault shows that the possibilities of thinking are fundamentally limited due to their dependence on a historical a priori, the so-called episteme. All discourses in which thought is expressed are ordered and regulated by their internal rules in relation to the episteme. They are therefore not dependent on thinking of individual subjects or ahistorical reason, but are fundamentally contingent. This social theory entails a problem of freedom because it shows that traditional concepts of freedom are not consistent: Freedom can neither be defined as arbitrary will (as in liberalism), nor as reason (as in Kantianism), nor as historical progress and social integration (as in Hegelianism).

Foucault extends his social theory to the realms of materiality, social institutions and bodies in his genealogical phase.[10] Discipline and Punish (1977) is both a genealogy of the modern prison and a critique of modern capitalism as based on the same social technologies that are used to discipline and subject individuals in the prison. By rejection of the methods of normative social philosophy that would enable to differentiate liberating and repressive aspects of modernity, and rejection of an account of socialization that would explain how individuals develop own agency, Discipline and Punish presents a picture of modern society as total repression. That there is no outside of power is the central critical insight of Foucault’s intervention. However, the standard objection raised against Foucault is that he conceptualizes power as only repressive, and therefore leaves neither a place for freedom and related concepts like resistance and emancipation nor for the articulation of a clear normative position.[11] I call this standard objection the ‘problem of power-determination’ to distinguish it from the ‘problem of subjectification’, which I introduce in the following section. A central advance of my argument will be to show that solving the problem of power-determination is not enough and that the problem of subjectification needs to be solved.

The standard response to the standard objection, that is, the problem of powerdetermination, mobilizes Foucault’s later works in his ethical phase, especially in the conceptual text The Subject and Power (1983).[12] The argument is that Foucault corrects his earlier conceptual mistakes and comes up with a refined theory of power and freedom.[13] In The Subject and Power, Foucault develops an action-theoretical model of power and defines power as ‘relationships between partners’ and ‘an ensemble of actions which induce others and follow from one another’.[14] Foucault is as explicit as one can be regarding the problem of power-determination, which seems to exclude freedom: The definition of power ‘includes an important element: freedom. Power is exercised only over free subjects, and only insofar as they are free’.[15] While most interpreters hold the standard response today, Thomas Lemke’s Foucault’s Analysis of Modern Governmentality: A Critique of Political Reason (2019, first published in German in 1997) gives arguably the best and most exemplary version of this widespread interpretative thesis. Therefore, I refer to Lemke’s text to structure my detailed discussion of Foucault’s text in the presentation of the standard response.

Compared with former publications, a striking feature of The Subject and Power is the action-theoretical and actor-based concept of power. Turning away from his former Nietzschean conception, Foucault describes power as social-ontologically reducible to actions of individual actors. He defines power as a relation between actors where one actor influences the field of action of another actor. This definition is similar to the understanding of analytic social philosophy, for example as in Thomas Wartenberg’s model.[16] It is easy to see how the action-theoretical model solves the problem of powerdetermination because here Foucault explicitly defines power and freedom in a relation of correlation and not in a relation of opposition. The action-theoretical model is built of the following systematic steps.

To start with, Foucault defines power as ‘relationships between partners’ and ‘an ensemble of actions which induce others and follow from one another’.[17] Power ‘is a way in which certain actions modify others’.[18] As action, power ‘is integrated into a disparate field of possibilities brought to bear upon permanent structures’.[19] Power is different from consent, as consent is not modifying other actions but is based on the ‘free’ decision of the other. This means that classic negative freedom as non-interference is the presupposition of consent according to Foucault.[20] While Foucault differentiates actual exercise of power from consent, he states that the field of possibilities in which the exercise occurs can be based on prior consent.[21] To differentiate power from violence, Foucault further specifies the definition: Power does not act directly upon bodies, like violence, but

in effect, what defines a relationship of power is that it is a mode of action which does not act directly and immediately on others. Instead it acts upon their actions: an action upon an action, on existing actions or on those which may arise in the present or the future.[22]

How does one act on the action of others when force is excluded? By acting upon the action environment to indirectly influence their action, because the action environment is both the enabling and restricting factor of the action. Foucault consequently says that power aims to ‘act upon the possibilities of action of other people [... ] to structure the possible field of action of others’.[23]

This demarcation from violence already includes the central definitory step for the solution of the problem of power-determination. Unlike violence, relations of power contain two necessary elements – Foucault says that they ‘are each indispensable if it is really to be a power relationship’[24] and thereby assigns them to the concept of power analytically:

That ‘the other’ (the one over whom power is exercised) be thoroughly recognized and maintained to the very end as a person who acts; and that [secondly, KS], faced with a relationship of power, a whole field of responses, reactions, results, and possible inventions may open up.[25]

Thus, in power relations, the person upon whom power is exercised actually has a whole set of possibilities, and therefore is free. Power, which is ‘a total structure of actions brought to bear upon possible actions [of, KS] an acting subject or acting subjects’.[26] social-ontologically presupposes freedom, as a result. Foucault is very clear:

Power is exercised only over free subjects, and only insofar as they are free. By this we mean individual or collective subjects who are faced with a field of possibilities in which several ways of behaving, several reactions and diverse comportments may be realized.[27]

From this fundamental freedom, which I also call ‘analytic freedom’, as it is analytically inferred from the concept of power, Foucault deduces resistance:

For, if it is true that at the heart power relations and as a permanent condition of their existence there is an insubordination and a certain essential obstinacy on the part of the principles of freedom then there is no relationship of power without the means of escape or possible flight.[28]

And: ‘The relationship between power and freedom’s refusal to submit cannot therefore be separated’.[29] In the standard response, these changes in the theory of power are taken to solve the problem of power-determination, namely the uncertainty about how human beings could be free and resist power. Foucault now assumes that human beings can always resist in relations of power. According to Lemke, this proves the equiprimordiality of power and freedom, which was missing in Foucault’s earlier conception. He sums up the importance of The Subject and Power:

The emphasis placed on freedom as an integral element of power relations holds two important analytical consequences. First, this conceptual shift makes it possible to overcome the negative definition of resistance that marked Foucault’s earlier works by lending substance to his thesis that power and resistance originate in tandem. Second, it presupposes a distinction that does not occur in previous studies, which now assumes great significance: the difference between power and domination.[30]

The standard response emphasizes the importance of this distinction between power and domination to which Lemke is referring. As Foucault rejected such a distinction in his genealogical phase, its introduction is a significant shift. The distinction is important in the standard response as it contains a normative position, and therefore is taken to be a solution to the standard objection that Foucault is normatively confused or contradictory.[31] The unclear status of normativity in Foucault’s critical method is a central part of the problem of power determination and has been raised by numerous critics.[32] The distinction is introduced only briefly in The Subject and Power; Foucault elaborates on it in more detail in a central interview during his ethical phase: The Ethics of the Concern for Self as a Practice of Freedom.[33]

Foucault’s usage of the terms power and domination in his late works on the one hand denominates different physical states of relations and, on the other hand, brings a differentiation between a general and a special case forth. Power is the term that denominates the social-ontological fact of power, always coming together with freedom, as described above. It is a general and fundamental fact of the social, which is why it can neither be disposed nor normatively valued or criticized as such.

That is to say, power relations are rooted deep in the social nexus, not reconstituted ‘above’ society as a supplementary structure whose radical effacement one could perhaps dream of. In any case, to live in society is to live in such a way that action upon other actions is possible – and in fact ongoing. A society without power relations can only be an abstraction.[34]

Because of analytic freedom, these ‘always present’ power relations ‘can be modified, they are not fixed once and for all. [ ... ] These power relations are thus mobile, reversible, and unstable’.[35] To describe the agonistic character of power relations, Foucault calls them ‘strategic’ and analyses their connection to ‘means of combat’ and ‘confrontation strategies’.[36] In difference to former works, Foucault’s usage of the notion ‘strategy’ is now completely transformed into an action-theoretical term, describing the means used by actors in power relations:

One may also speak of a strategy proper to power relations insofar as they constitute modes of action upon possible action, the action of others. One can therefore interpret the mechanisms brought into play in power relations in terms of strategies.[37]

Power relations imply for Foucault, ‘at least in potentia, a strategy of struggle’, which he defines through ‘the objective [ ... ] to act upon an adversary in such a manner as to render the struggle impossible for him’.[38] Foucault deduces this implication from the ‘permanent condition’ of power relations, ‘an insubordination and a certain essential obstinacy on the part of the principles of freedom’.[39] However, if one opponent succeeds to win the battle, that is the end of power: ‘A relationship of confrontation reaches its term, its final moment (and the victory of one of the two adversaries) when stable mechanisms replace the free play of antagonistic reactions’.[40] While consensus and violence are the complete other of power, the stabilization of power is its internal other that ‘result[s] in the limits of power’.[41]

In summary, power is characterized first by its social-ontological generality and second by its mobile, reversible and instable – or liquid – physical state. In difference to power, domination is power’s internal limit or the internal other of power, precisely, a specific case of solidification of power relations:

The analysis of power relations is an extremely complex area; one sometimes encounters what may be called situations or states of domination in which the power relations, instead of being mobile, allowing the various participants to adopt strategies modifying them, remain blocked, frozen. When an individual or social group succeeds in blocking a field of power relations, immobilizing them and preventing any reversibility of movement by economic, political, or military means, one is faced with what may be called a state of domination. In such a state, it is certain that practices of freedom do not exist or exist only unilaterally or are extremely constrained and limited.[42]

Domination, according to Foucault, means the particular case of blocked, frozen resp. solidified power, which prevents reversion and movement. In that sense domination is something one-sided: It is the success of a certain group or individual. In The Subject and Power, Foucault specifies the spatiality and temporality of domination:

Domination is in fact a general structure of power whose ramifications and consequences can sometimes be found descending to the most incalcitrant fibers of society. But at the same time it is a strategic situation more or less taken for granted and consolidated by means of a long-term confrontation between adversaries.’[43]

In this citation, domination is a society-wide structure and not only a local solidification, and its stability is based on its historical development.[44] Freedom and domination are opposites because domination limits the free dynamic of power.

So far, there are two concepts: power (namely free and resistant games of power relations) and domination (namely a situation of solidification that limits freedom and resistance). This definition brings up the question of how power is transformed into domination – the answer being: through government. Foucault defines government as what stands between and connects power and domination:

And between the two, between games of power and states of domination, you have technologies of government – understood, of course, in a very broad sense that includes not only the way institutions are governed but also the way one governs one’s wife and children. The analysis of these techniques is necessary because it is very often through such techniques that states of domination are established and maintained. There are three levels to my analysis of power: strategic relations, techniques of government, and states of domination.[45]

Hence, there are three concepts, which cover a spectrum from generality to specificity and from dynamic to solidified: power (or ‘strategic relations’), government, and domination. As Foucault equates power with freedom, and as freedom is a normative term, the distinction between power and domination is also a normative distinction, and not only a social-ontological one (liquid/solid). The distinction of power and domination, thus, is Foucault’s answer to the constant critique that he lacks a normative orientation. Domination is ‘bad’ because it limits freedom, while ‘free’ games of power are ‘good’:

This is precisely a failure to see that power relations are not something that is bad in itself, that we have to break free of. I do not think that a society can exist without power relations, if by that one means the strategies by which individuals try to direct and control the conduct of others. The problem, then, is not to try to dissolve them in the utopia of completely transparent communication but to acquire the rules of law, the management techniques, and also the morality, the ethos, the practice of the self, that will allow us to play these games of power with as little domination as possible.[46]

In the standard response, the differentiation between power and domination as ‘free’ and ‘unfree’ power is central, as it opens up the possibility for political critique. Thomas Lemke writes:

The theoretical distinction between power and domination holds political significance. Only on the basis of this distinction is it possible to formulate a critique of conditions of domination – and strategies to overturn them – without invoking some idea of a realm ‘beyond power.’ In this way, Foucault retains the premise of ‘omnipresent power’ in social relations while providing a criterion for distinguishing ‘freer’ and ‘less free’ forms to power.[47]

The social-philosophical attraction of Foucault’s thinking lies in his conception of the immanence of power. There is no realm outside of power as in liberal political philosophy, instead power is omnipresent, and freedom cannot be located beyond power. In Foucault’s former works, the conception of immanence did not allow for a differentiation between power that is freer and power that is less free. The standard response is that the distinction between power, government and domination offers this differentiation. It allows for a concept of immanence of power and a normative differentiation between different situations of power: situations in which there is only power (and therefore freedom) in contrast to situations in which power is solidified to domination (and therefore freedom is limited). This entails that while it is not possible to conceptualize a world without power, it is possible to conceptualize a world without government and domination. Because reducing or even transgressing government and domination is the normative vantage point of the standard response, I call it anarchist.

According to the standard response, Foucault’s new social theory solves the problem of freedom: Now that freedom is not contradictory to, but a necessary element of power, it is clear that human beings are not determined by power. Because of this fundamental freedom, resistance is not rendered impossible by the conception of power but is always possible. Humans are not mere objects of the intensification of power through discipline in modernity, but they can and do in fact resist.

My hypothesis is that this standard response is flawed because, merely reacting to the standard objection, it solves the wrong problem. The problem of determination that it solves is not what the debate about freedom in Foucault’s works really is about. The debate is, in fact, about the problem of subjectification. By focusing on the problem of subjectification, it will become clear that the standard response operates with the notion of freedom to act differently, which lacks modal robustness. Freedom to act differently is an alternative term for analytic freedom, indicating that this kind of freedom is rudimentary. Only an institutionalist interpretation of freedom as critique can account for modal robustness and thereby respond adequately to the problem of subjectification. Modal robustness means stability over different possible worlds, for example that there is still freedom in the possible world of the raise of right-wing populism. This reading entails investigating the institutional requirements of freedom and thereby breaks with the anarchist conception of the standard response, which conceives institutions, that is, government and domination, in opposition to freedom

2. The problem of subjectification and freedom as critique

While the problem of determination consists in an interpretation of Foucault’s genealogical works, especially Discipline and Punish as describing a totality of power that determines individuals, the problem of subjectification is present in later works, especially on governmentality. In his history of governmentality, Foucault differentiates his conception of power into three different types: sovereignty, government, and discipline.[48] Discipline is the power that subjectivates in a way that it subjects individuals – hence the problem of determination in a book that only describes this kind of power, Discipline and Punish. Government, on the other hand, is a subtler form of power that is actually based on freedom. In the last section, I reconstructed Foucault’s later socialphilosophical systematization of his concept of power as government, which derives from the material of his genealogical analyses of government. These analyses focus on pastoral power and early liberalism. In both, government is the conduct of the actions of others. Freedom is not only a presupposition of these forms of power but also produced by power itself for its strategic goals.

Foucault and, following him, scholars in the field of governmentality studies show that neoliberal government makes freedom its main strategy.[49] To be a productive neoliberal subject, the subject must be free. Only a free subject can be creative, self-responsible and innovative. Therefore, neoliberal subjectification does not operate by subjugating subjects but by creating incentives to become free and even resistant. The resistance against inefficient structures is taken to be an important driving force for improvement in neoliberalism. This complicates things and leads to the problem of subjectification: Of course, power and freedom are not opposites, but the freedom in neoliberalism is actually a strategy of power itself. Even resistance to power can be a strategy of power. While there is no problem anymore to speak about freedom in the theory of power, it is not clear if this freedom is the freedom one could want. On the contrary, even the free self-constituting activity of the subject is dependent on subjectification through power; freedom is only neoliberal freedom. The analysis of governmentality and the concept of power and freedom as equiprimordial open up the possibility of a totality of repressive neoliberal power (which is supported by neoliberal freedom). So now we have a concept of subjectification that is compatible with freedom, but we still have the justified suspicion that we are more thoroughly subjugated through power than we usually think. The problem of subjectification is that we do not know how free we really are. The analysis of neoliberal government shows that neoliberal freedom is not desirable and does not allow for resistance against neoliberal government, which leads to the question what ‘real’ freedom could be and how it can be conceptualized in the framework of subjectification.

While this problem is very visible in the case of neoliberalism, I do not call it the problem of neoliberalism but the problem of subjectification because it arises from the concept of subjectification, the complex constitution of the subject through the interaction of self-constitution and constitution through power. Following the radical historicism that underlies the theory of power and subjectification, also the technologies of the self, which the subject uses to constitute itself freely, are fundamentally dependent on the regime of subjectification of a given time and place, thus: power. This conception does not allow for a differentiation between freedom as a strategy of power and ‘real’ freedom that would allow for resistance against power. Therefore, the problem of subjectification is an intensification of the problem of determination. The problem of determination was an unrealistic depiction of society as not allowing for freedom at all; but the problem of subjectification is more severe: While freedom exists, it is constituted and subtly repressed by power. Thus, the problem of subjectification lies in the fact that it is unclear how to distinguish between ‘repressive’ and ‘real’ freedom.[50]

How can the problem of subjectification be solved? That is, how can we deal with the reasonable suspicion that our actions, even when we believe to be acting based on free will, are inevitably dependent on power? The term freedom must be differentiated to answer this question. It is necessary to find a concept of freedom that can explain a subject’s activity independent of power, without giving up the hypothesis of subjectification that subjectivity is fundamentally constituted by power.[51]

Foucault himself developed such a concept of freedom in his reflections on the method of genealogy in his later works.[52] I call it freedom as critique, or freedom as the capability to critically reflect on one’s own subjectification. Foucault describes genealogical critique as both an ontology of the present and a history of the present. He considers it to be part of a practice of freedom, as it aims to critically reflect on the topic of subjectification. It renders subtle repression visible and thereby opens the possibility to change one’s identity, ethics, and politics.

More systematically, freedom as critique can be described as a reflection of higher order than the immediate reflection that is initiated by subjectification. This higher-order reflection allows to transgress the immediate reflection. In other words, it is a specific technology of the self that aims at reflecting the potential influence of power on all selftechnologies. While also freedom as critique is dependent on subjectification – absolute freedom or freedom as a fixed status thus is impossible – it is a movement that always aims at transgressing itself and thereby reaches as much independence and distance from subjectification as possible.[53] Freedom as critique is an internalized hermeneutics of suspicion that always critically rechecks everything, including itself. It never stops but adds critical operation on critical operation; it is an operation of nth order, and therefore a practice (and not a status or state) but nevertheless dependent on capabilities.[54]

Through this operation of critique, which is a practice of the self on the self, a subject can transform itself by emancipating itself from the subjectifications that constituted it in the first place. It can reach independence and autonomy vis-‘a-vis the outside that constituted it. Something new is created that cannot be reduced to power. Freedom is an emergent level of operation vis-‘a-vis the subjectifications that constitute the subject – it is inner-psychic emergence. How change happens is not predictable but it is predictable that this sort of inner-psychic emergence becomes more likely due to free subjectifications.[55] The method of genealogical critique aims at creating freedom, which does not mean that it cannot be created through other means of critical reflection. The central argument is that freedom as critique cannot be presupposed in the subject. It can only arise as result of former subjectifications. As a technology of the self, it has to be learned and trained.

The standard response is not blind to the problem of subjectification and the concept of freedom as critique. On the contrary, all interpreters of Foucault acknowledge the complex relationship between power as government and freedom. They also put emphasis on Foucault’s critical method of genealogy and its connection to his concept of freedom. Against this background of the problem of subjectification, they pose ‘the problem of ‘resistant subjectivity’’,[56] and analyse that the answer is related to freedom as critique. However, as the standard response fundamentally draws on The Subject and Power and freedom to act differently, it is incapable of making the argumentative steps that are necessary for answering these questions.

3. Beyond the standard response: Modally robust freedom as critique

The standard response is that the problem of freedom is solved in Foucault’s later works, especially in his concept of power as government in The Subject and Power. By making the distinction between two types of problems and concepts of freedom, which I introduced above – the problem of power-determination in relation to freedom to acdifferently and the problem of subjectification in relation to freedom as critique – it is possible to see why the standard response fails: It only accounts for freedom to act differently, and not for freedom as critique, and therefore cannot solve the problem of subjectification. The crucial difference between the two concepts of freedom is their varying modal robustness: Freedom as critique demands modal robustness of critical capabilities and therefore leads to the question how freedom can be institutionalized. That such an institutionalized reading is in contradiction to the anarchist commitments, that can also be found in Foucault and which are valued by many of his readers, explains why it has not been pursued before.

Foucault derives fundamental freedom in The Subject and Power from the concept of power: It only makes sense to speak about power if the individual over whom power is exercised is ‘as a person who acts’ and thus is free to react to the exercise of power so that ‘a whole field of responses, reactions, results and possible interventions may open up’.[57] I called this concept of freedom ‘analytic freedom’, as it derives social-ontological freedom solely from the concept of power. It accounts for the social-ontological fact that human beings can always act differently, and therefore are not determined by power, more precisely if ‘the determining factors [do not] saturate the whole there is no relationship of power’.[58] Thus, power requires social-ontological freedom to act differently and only changesthe field of possible actions, ‘it incites, it induces, it seduces, it makes easier or more difficult’.[59]

So far, this social-ontological and analytical action-theoretical description of power as necessarily connected with freedom is formalistic and empty. It is analytic insofar as this rudimentary concept of freedom can be derived already fromthe concept of power – and as it does not add anything. Freedom to act differently has significant similarities to the negative concept of freedom as non-interference in the liberal tradition: As an ontological quality it exists quasi pre-socially, that is, subjectification is not part of the concept, and it aims at reducing government in order to maximize freedom. This basic freedom can be attributed even to the disciplined and subjected subjects of Discipline and Punish. This, however, does not already enable them to be politically resistant. In other words: The social-theoretical changes in The Subject and Power solve the problem of power-determination, as Foucault writes explicitly that power presupposes freedom. But an account of analytic freedom — or freedom to act differently — is not already an account of resistant freedom or freedom as critique,[60] which is the concept that characterizes resistant subjectivities.

It is not only a problem of Foucault’s readers to confuse the different concepts of freedom, Foucault does so himself. After the formalistic description of freedom, Foucault is surprisingly charging the concept of freedom with a specific content and connects it to resistance. Because, according to Foucault, the use of power is geared towards determining the boundaries of freedom, freedom must, in turn, be resistant to power: ‘Freedom must resist the exercise of power which finally aims at totally determining it’.[61] From this, Foucault deduces the necessary connection between power and resistance – ‘The relationship between power and freedom’s refusal to submit cannot therefore be separated’.[62]

Foucault already states this connection between power and resistance in History of Sexuality 1 (1978), where he, however, derives it from the Nietzschean conception of power as war and not from conceptual logic and action theory. Foucault seems to come back to this former conception explicitly in The Subject and Power, writing:

Rather than speaking of an essential [‘antagonism’],[63] it would be better to speak of an ‘agonism’ – of a relationship which is at the same time reciprocal incitation and struggle; less of a face-to-face confrontation which paralyzes both sides than permanent provocation.[64]

Here it is not clear whether Foucault’s talk about resistance extends the former, analytical and social-ontological statement that power presupposes freedom to act differently. Foucault seems to hover between deriving resistance from blockade and antagonism (resistance as a concrete reaction to the attempt of complete domination) and defining resistance as a permanent agonistic dynamic. By postulating an antagonism, resistance could be described as one side of it, however, this would fundamentally extend the social-ontological conception because such an antagonism cannot be explained in the action-theoretical model. Therefore, it is more plausible that Foucault derives resistance from a dynamism of permanent agonism. This interpretation is compatible with the action-theoretical and analytical framework but adds a kind of dynamism to freedom and resistance. It is more fitting to Foucault’s concept of power as continuous struggle and seems to be implied in the citation above where Foucault states that it is about agonism, not antagonism. But in this reading, resistance would be as much a social-ontological basic fact as power and freedom – and thereby equally formalistic, empty and not further qualified. These thoughts on resistance do not add anything to the formal and analytic conception of freedom to act differently that has already been established.

But Foucault seems to want more by elaborating on the meaning of resistance. While it was sufficient in the action-theoretical concept of freedom to have a choice of action within a given field of possible actions, the introduction of resistance implies that these possible actions are used to resist the influence of power on the field of possible actions, rather than just following one of the possible actions in a field of action constituted by power. This interpretation – that resistance requires the change of the field of action – is plausible if one takes into account that Foucault’s normative notion of freedom as critique aims at transforming oneself, which also necessitates the transformation of power.

Now, if Foucault’s discussion of resistance is really such a fundamental extension of the concept of freedom that goes beyond the formal social-ontological and analytic conception of freedom, this purely formal conception is not enough to give an account of resistant freedom. If resistance is supposed to be more than just the freedom to act differently, which is analytically presupposed by power, it is necessary to say more about the conditions of this kind of resistance. The problem is that Foucault socialontologically generalizes resistance, as he generalizes analytic freedom, which precludes the possibility to ask social-theoretically (and not genealogically) about the specific conditions (e.g. historic, social or social-ontologically abstractable conditions) of resistance. Foucault does not distinguish between two different modalities of resistance: its general possibility and its probability or actuality.[65] Through his action-theoretical social ontology and the analytic concept of freedom, he can show that there is the general possibility of resistance but not its actuality or probability.

Statements about the probability of freedom can be considered more ‘modally robust’ than statements that state the general possibility of freedom. I call the ‘argument from modal robustness’ the claim that a social-theoretical account of freedom should not end by stating the general possibility of freedom, but should differentiate situations according to the probability of freedom. The argument is inspired by applications of modal logic in contemporary social philosophy, especially the so-called neo-republican tradition with Philipp Pettit as its key figure.[66] While the neo-republican tradition offers a specific account of freedom as non-domination based on a wholly different social ontology[67] than Foucauldian freedom as critique, I argue that the neo-republican use of modal logic without neo-republican social ontology is particularly illuminating for the development of the Foucauldian concept of freedom as critique.

Very roughly, a state is modally robust if it occurs in many different possible worlds, that is, when it is probable. The language of possible worlds helps explain different aspects and conditions of robustness and probability.[68] Neo-republican authors put forward the argument of modal robustness to criticize traditional conceptions of liberal freedom as non-interference. Within this conception, you are free under the rule of a benevolent dictator, given that they choose not to interfere in your life even though they could. In contrast, freedom as non-domination is modally robust: Its core meaning is not actual non-interference, but robust non-interference, that is, the institutional stabilization of freedom as non-interference. To stabilize the absence of arbitrary interference, the rule of law is key, according to neo-republicans, as only the rule of law can guarantee non-interference in relevant possible worlds, for example the possible world of a rightwing government. Accordingly, freedom of non-domination ‘has a built-in rule-of-law requirement.’[69]

This argument from modal robustness applies to the discussion of freedom as critique, albeit here on ‘inner’ and not external limitations of freedom. As in freedom as noninterference, accounting for resistance and the capability for critique through freedom to act differently is not modally robust, but contingent. Resistance is possible, but not probable. The concept cannot say anything about relevant possible worlds and how to ensure resistant subjectivity in such possible worlds. Just like freedom as noninterference does not guarantee that the benevolent dictator does not turn hostile, freedom to act differently does not guarantee the development of critical capabilities. In contrast, freedom as critique is a modally robust concept, that is, it necessitates the stabilization of critical and potentially resistant subjectivity.[70] This demand for robustness stems from the normativity of the concept: In the language of modality, it is a modal desiderata, which describes permissible worlds, according to which we should change modal facts that influence the set of possible worlds.[71] That means we should influence the world in such a way that freedom as critique is ensured in the relevant possible worlds, that is, that resistant subjectivity is stabilized.

Equipped with this vocabulary, it is possible to understand the tensions I analysed regarding the status of resistance in The Subject and Power as stemming from a missing differentiation between the modalities of the concepts of freedom involved. From the discussion of the problem of subjectification and the concept of freedom as critique, it can be concluded that Foucault and his readers mean more by resistance than the mere freedom to act differently: a reflected, directed and intentional resistance as critique with the aim of self-determination. But in this case, resistance is a demanding capability that cannot be generalized, but is only contingent, which is why a modally robust account of freedom as critique is necessary to explain how its general possibility can be transformed into probability and actuality. This allows us to overcome both the standard objection and the standard response by specifying freedom as critique’s modal robustness through the modal facts leading to the probability of critical subjectivity.

These conceptual propositions significantly extend the possibilities of critical social theory following Foucault. In the framework of the standard objection and response, the relationship of practices of government and practices of freedom are not systematically investigated, as government and domination are seen as limiting freedom per se; rather, in this framework, social critique is limited to the optimistic assumption that there will always be critique, often reducing critical research to happily highlighting existing and past resistances.[72] Following freedom of critique, it is possible to conceptualize the social conditions of freedom and resistance, especially the institutions that could follow from Foucault’s normative commitments. My conclusion briefly elaborates on this institutional reading and how it critically expands the three common notions of freedom in social philosophy.

4. Conclusion

Foucault’s theory of power productively complicates the common social-philosophical concepts of freedom by the problem of subjectification, that is, ‘inner’ unfreedom that follows from the social constitution of subjectivity. While the problem of subjectification challenges the three common concepts of freedom – negative, reflexive and social – the Foucauldian notion of freedom as critique is equipped to solve it. In contrast to the standard response of Foucault’s interpreters to the problem of freedom, I argued that it is crucial to account for the modal robustness of freedom as critique.

Modal robustness can be achieved by changing the modal facts, that is, the situation that makes possible worlds more or less likely. In Foucault’s terms, strict modal robustness is a situation of domination (when power is fixed), and relative modal robustness or probability is government. Thus, in contrast to the standard response, which sees freedom as the opposite of government and can therefore be called anarchist, a modally robust notion of freedom as critique entails a normative differentiation of forms of government and their subjectifications. This is because freedom as critique is such a demanding capability that it cannot be presupposed within the subject but rather can only be the result of specific regimes of subjectification that I call ‘free’ or ‘critical subjectification’. And modally robust regimes of critical subjectification are a specific form of government that could be termed ‘free’ or ‘critical government’. This is precisely the opposite of neoliberal government, which uses ‘freedom’ as a means for social control, as governmentality studies showed. To search for such kind of free or critical government means, in other words, to normatively differentiate between political institutions according to their subjectification. This entails a rather un-Foucauldian turn: to connect the theory of subjectification with political theory, understood as a normative thinking about institutions and to break with Foucault’s partly anarchist commitments.

Following this path is promising, as it finally brings clarity to the debate about freedom in Foucault’s works. Furthermore, the institutional reading opens new possibilities to engage Foucault with democratic theory and with the common concepts of freedom in social philosophy. Regarding democratic theory, the Foucauldian notion of freedom as critique can serve to explain how the critique of institutions can be institutionalized, thus systematically presenting the democratic-theoretical consequences and resources of Foucault’s thinking for the first time.[73] Specifically, the concept can serve as a normative basis for postfoundational and radical concepts of democracy. Postfoundationalist democratic theory describes the political as intrinsically contingent and antagonistic, leading to powerful analysis of politics, but this anti-universalism is a challenge for developing a normative account of democratic institutions within this tradition.[74] Freedom as critique, being an institutionalization of anti-universalist critiques of power, is ‘the last universalism’ in postfoundational thinking and allows us to clarify the normative and institutional commitments of postfoundational political theory, as I argued elsewhere.[75]

As a new conceptual candidate, freedom as critique enriches the general socialphilosophical debate about the different concepts of freedom, encapsulating the political dimension of ‘inner’ unfreedom and postfoundationalist scepticism in one conception of freedom for the first time. It offers the democratic-theoretical and social-philosophical desideratum of Foucault’s analysis of power, which can be processed as immanent critique by existing accounts of freedom.

In contrast to a negative conception of freedom, as it is represented by classical political liberalism like Hobbes, Berlin and Rawls[76] as well as by neo-republicanism like Pettit and Lovett,[77] the perspective of freedom as critique emphasizes that internal or psychological unfreedom is a politically relevant problem. While there are many areas of politics in which it is appropriate to understand the politically relevant freedom as negative freedom, the perspective of freedom as critique can shed light on those areas in which negative freedom is not sufficient. These areas of imperceptible domination and normalization are, for example, sexist, racist and capitalist subjectivations. With a negative conception of freedom, these internal types of unfreedom cannot be addressed, which is why an institutional design based on such a conception cannot deal with this unfreedom. This can be corrected by the perspective of freedom as critique, for example through an educational policy that thematizes this type of unfreedom and encourages reflection on it.

In contrast to a positive or reflexive concept of freedom that equates freedom with reason as it occurs in the Kantian tradition, freedom as critique illuminates the normalizing and repressive effects of reason, insofar as it is always impure. While the difference between such a universal moral perspective and the sceptical anti-universalist perspective of freedom as critique seems to be philosophically fundamental, it can be accommodated by contextual accounts of universalism and the concept of immanent critique.[78] Once this step has been taken, the proceduralization of political discourse in deliberative models[79] and the Foucauldian demand for constant reflexive critique of subjectification have great similarities. The concept of freedom as critique complements political theories based on reflection with the problematization of imperceptible normalization through subjectification, which does not appear sufficiently in these perspectives.

Between a social conception of freedom as it is put forward in the Hegelian tradition, but also in traditional republicanism, communitarianism,[80] and freedom as critique, there is a fundamental agreement and a difference: The agreement consists in both social freedom and freedom as critique drawing on a holistic social ontology and conceptualizing subjects as socially constituted.[81] The difference consists in the fact that the tradition of social freedom does not see this social-ontological insight as problem of freedom but tends to equate successful social integration with freedom. In contrast, freedom as critique is fundamentally sceptical of social integration and suspects it of being, sometimes imperceptibly, repressive. Whether the two perspectives are compatible depends on the concrete interpretation of social freedom. Conflict is not fundamentally alien to the perspective of social freedom.[82] If conflict and the potential paternalism of social freedom were brought into the center of the theories on social freedom, there would be no fundamental incompatibility.[83] Integrating the problematization of subjectivation into the theories based on social freedom can correct their optimism about socialization and help describe the repressive normalization they tend to overlook, to make them more realistic, as Katharine McIntyre recently argued in Philosophy and Social Criticism. [84]

In this article, I presented the first step for the development of this new Foucauldian concept of freedom as critique: To explain how the standard objection against Foucault and the standard response of Foucauldians are both committed to an anarchist reading of Foucault, which missed key differentiations of problems and notions of freedom in Foucault’s thought. An institutional account that integrates the argument from modal robustness helps to carefully distinguish the different notions of freedom that are present in Foucault’s texts, albeit often implicitly, which is crucial for any attempt to understand his thinking about power and freedom. That freedom to act differently is not the same as freedom as critique, which demands modal robustness, and that The Subject and Power should not be taken as the end of the debate about freedom in Foucault’s works are central insights that need to be taken into account by the many Foucault commentators who follow the standard response. This clarification about the problems and conceptions of freedom could serve as a new beginning for debates about Foucauldian freedom in postfoundational democracy and social philosophy.

References

[1] Cf. the informative typology in Honneth (2011, 33–118). As in Honneth’s reconstruction, the concepts of reflexive and positive freedom are used synonymously here (2011, 59).

[2] Isaiah Berlin analyses the structural paternalism of liberation theories based on reflexive freedom through which intellectuals empower themselves to philosophically determine the good and the right. At the same time, they explain why not all have attained this knowledge yet, using such an explanation of the unfreedom of others as a justification for ‘liberating’ them by coercive means, see Berlin (2002). Foucault shares this concern and develops his genealogical method to avoid such paternalism.

[3] Cf. Honneth (2011). In this way, the relationship of Honneth’s recognition approach and Foucault can be conceptualized by pointing to the specific similarities and differences of the concepts of freedom, instead of integrating the two schools, as in recent commentary such as McIntyre (2019) and Heidenreich (2015), or displaying them in fundamental contradiction like Kelly (2017).

[4] Cf. Menke (2020, 118–19, 124, 145–46)

[5] See the summary of the criticism in Fillion (2004). Another description of the ‘problem of freedom’ is the ‘Foucault Conundrum’: ‘In fact, the objections that are typically raised against Foucault are so common and so similar – coming from a range of critics including Michael Walzer, Richard Rorty, Ju¨rgen Habermas, Nancy Fraser, and Alasdair MacIntyre – that I have given the dilemma upon which these criticisms are based a name: the Foucault Conundrum. This conundrum has to do, first, with the critical observation that Foucault’s description of our subjection is so thorough that it is not clear how we can speak of freedom in any meaningful sense, since we always find ourselves constrained to a set of power relations. Second, genealogy historicizes both rationality and morality such that it tends to undermine any appeal to normative standards, so it is not clear how Foucault, or for that matter any political theory built on genealogy, can say that we ought to try to free ourselves. As such, it is typically argued, Foucault’s thought leads to a kind of moral inertia’ (Clifford 2001, 14).

[6] Major works of this phase of the debate are Patton (1989), Lemke (2019), Saar (2007) and Allen (2008).

[7] See Schubert (2018); the English translation of the German title is ‘Freedom as Critique. Social Philosophy After Foucault’. This article presents some of the core arguments of the German book and refines them through the detailed discussion of the ‘argument from modal robustness’.

[8] One of the most recent publication that follow the standard response is McIntyre (2019). The dominance of the standard response, that is, the thesis that Foucault corrects his theory of power and freedom in his later works, is shown by the fact that it is reproduced as the only interpretation in a handbook entry on Foucault, cf. Patton (2010). Already the anthology of Moss (1998a) asks how Foucault’s later works answers the problems of the works of the 70s, see Moss (1998b, 2 and 5f.). Some of the many journal articles based on the correction thesis are Ko¨gler (1990), Bevir (1999), Webermann (2000), Schaff (2004), McGushin (2005) and Lepold (2014). Luxon (2008) follows the correction thesis with a focus on Foucault’s parrhesia and Taylor (2003) with a focus on ethics and Foucault’s reading of Kant. Bell (1996) interprets the late works as elaborating on freedom corresponding to the problem of freedom in the earlier works. Kolodny (1996) explains the sceptical method of Foucault’s late ethos to defend him against the accusation of normative confusion. That the correction thesis has lost nothing of its popularity is shown by the most recent monograph based on it (Lynch 2016). However, in relation to the problem of freedom the book offers little new; the thesis of Lynch consists in deriving a normative concept of freedom from a rudimentary ontological concept of freedom (Lynch 2016, 190f.) and is therefore a variation of the systematic core problem that I discuss in this article. In difference to the standard reponse, Oksala (2005) reconstructs Foucault’s notions of freedom throughout the different phases of his work.

[9] According to the common distinction of the three phases ‘archaeological’, ‘genealogical’ and ‘ethical’, the first archaeological phase includes Foucault’s main works of the 60s, its end marked by his inaugural lecture at the Coll‘ege de France from December 1970 (Foucault 1971).

[10] The genealogical phase extends roughly from 1970 until Foucault’s ‘turn to ethics’ at the end of the 70s.

[11] See Lemke (2019, 108–18) for a detailed analysis of these problems, which are evoked by many commentators, for example by Fraser (1989, 17–66), Habermas (1988), Honneth (2000), Taylor (1984) and Dews (1987, 155–70).

[12] The ethical phase includes Foucault’s works on ancient philosophy and ethics, from about 1980 until his death in 1984.

[13] In an alternative view, The Subject and Power is interpreted as a clarification, and not as a correction, of the former works on knowledge and power, which are thereby presented as social-theoretically coherent; paradigmatic for this is Patton (1989). In difference to that, the thesis of The Subject and Power as a self-correction of Foucault entails that the former works were social-theoretically wrong, that is, that the problem of freedom in these works is real.

[14] See Foucault (1983, 217).

[15] See Foucault (1983, 221).

[16] See Wartenberg (1988). For other interpretations that stress the action-theoretical aspect of the concept of power in The Subject and Power, see Fink-Eitel (1992) and Ko¨gler (2005).

[17] See Foucault (1983, 217).

[18] See Foucault (1983, 219).

[19] See Foucault (1983, 219).

[20] See for the notion of power as non-interference Berlin (2002) and Pettit (2011).

[21] See Foucault (1983, 220).

[22] See Foucault (1983, 220).

[23] See Foucault (1983, 221).

[24] See Foucault (1983, 220).

[25] See Foucault (1983, 220).

[26] See Foucault (1983, 220).

[27] See Foucault (1983, 221).

[28] See Foucault (1983, 225).

[29] See Foucault (1983, 221).

[30] See Lemke (2019, 319).

[31] For a different interpretation see Saar (2007), who argues that the distinction of power and domination is more a concession to persistent interviewers than an important systematic element of his theory of power, see also Saar (2007, 256–60).

[32] See exemplary Taylor (1984), and a response in line with the standard response Patton (1989); see also Schubert (2018, 39–61) for a detailed analysis and critique of this discussion.

[33] See Foucault (1997a).

[34] See Foucault (1983, 222–23).

[35] See Foucault (1997a, 292).

[36] See Foucault (1983, 224–25).

[37] See Foucault (1983, 225).

[38] See Foucault (1983, 225).

[39] See Foucault (1983, 225).

[40] See Foucault (1983, 225).

[41] See Foucault (1983, 225).

[42] See Foucault (1997a, 285).

[43] See Foucault (1983, 226).

[44] It is striking that Foucault talks only about ‘economic, political and military’ instruments of domination and thereby limits domination to such phenomena, even though he criticizes Marxism precisely because of its narrow focus on such instruments. Contrary to the standard response, Martin Saar states that Foucault only presented the distinction between power and domination when specifically asked to do so in an interview. Thus, he considers this distinction to be unimportant to Foucault. Domination, according to Saar, is only an unlikely marginal case for Foucault, in which he is not interested. Cf. Saar (2007, 283).

[45] See Foucault (1997a, 299).

[46] See Foucault (1997a, 298).

[47] See Lemke (2019, 322).

[48] I will not go into sovereign power as it is the denominator for a juridical analysis of power in a roughly Hobbesian and Weberian line of thought and not of interest regarding the problem of freedom.

[49] Cf. for the German and anglophone debate of the governmentality studies the description of the approach in Bro¨ckling, Krasmann, and Lemke (2004a) and the overview of the debate in Bro¨ckling, Krasmann, and Lemke (2011a), as well as the other contributions in Bro¨ckling, Krasmann, and Lemke (2004b). Foundational texts are Rose and Miller (1992) and Dean (1999), as well as the contributions in Burchell, Gordon, and Miller (1991), Barry, Osborne, and Rose (1996) and Krasmann and Volkmer (2007). Critical interim balances of the debate can be found in Lemke (2000) and Osborne (2001). A mix of a reflection of methods and current case studies covering different policy areas is offered in Angermu¨ller and van Dyk (2010) and Bro¨ckling, Krasmann, and Lemke (2011b), as well as Vasilache (2014) and Bargetz, Ludwig, and Sauer (2015).

[50] Such a distinction sounds like an attempt to normatively differentiate societies: there are neoliberal societies whose concept of freedom is actually not freedom, but subtle repression, and there are free societies, in which real freedom exists. In contrast to such an Hegelian account, the analysis of subjectification would insist that the problem lies in the relationship between the individual and society. The individual should be able to constitute itself and decide freely to reject certain aspects of power. Therefore, Foucauldian freedom is fundamentally at odds with society and comes from the premise that a ‘free’ society is impossible.

[51] Here my argument fundamentally differs from the method of governmentality studies and Foucault’s own genealogical approach. These approaches affirm that the distinction between neoliberal or repressed freedom and ‘real’ resistant freedom collapses. The blurring of these concepts is the very core of their non-normative critical operation, as Martin Saar (2007) shows. Ulrich Bro¨ckling argues that critique should not be a normative program against neoliberal subjectification, which would be based on a normative concept of freedom, but a tactic without a solid normative foundation. In short, out of the social-theoretical diagnosis that normative distinctions get blurry when seen through a realist conception of power and subjectification, Bro¨ckling infers the methodological imperative for critique to abstain from normative distinctions. Cf. Bro¨ckling (2013, 283–88).

[52] See Foucault (1997b, 1997c).

[53] Cf. Butler (2002).

[54] For a detailed analysis of freedom as critique, see Schubert (2018), especially the systematic summary of the argument on pp. 305–312.

[55] With his concept of subjectification as folding of power, that is, a folding of the outside that constitutes an inside, Deleuze seems to have such emergence in mind. But in contrast to the position I defend, Deleuze does not have a concept of different kinds of foldings or depths, but seems to presume that this emergence always occurs in the same way, cf. Deleuze (2004).

[56] See Lemke (2019, 261). Note that Butler (1997, 10) poses exactly this problem: ‘How can it be that the subject, taken to be the condition for and instrument of agency, is at the same time the effect of subordination, understood as the deprivation of agency? If subordination is the condition of possibility for agency, how might agency be thought in opposition to the forces of subordination?’ However, her psychoanalytic answer to the question faces systematically similar problems as the standard response, that is, it is ontologizing a rudimentary concept of freedom instead of explaining the institutional conditions of possibility of freedom as critique.

[57] See Foucault (1983, 220).

[58] See Foucault (1983, 221).

[59] See Foucault (1983, 220).

[60] Resistant freedom, critical freedom and freedom as critique all refer to the same higher reflexivity vis-‘a-vis analytical freedom. Analytical freedom refers to the choice of options in a given action environment, while freedom as critique refers to the question where the preference for an option comes from, that is, it is located at a meta level. The reflection could then come to the conclusion, that the preferences are based on a subjectification through harmful norms and thereby create the wish to change the preference and the connected lifestyle.

[61] Translated by K.S., German original: ‘Die Freiheit [muss] sich einer Machtausu¨bung widersetzen, die letztlich danach trachtet, vollsta¨ndig u¨ber sie zu bestimmen’ (Foucault 2005, 287). This sentence appears only in the German and French version, not in the English one, cf. Foucault (1994a, 238) and Foucault (1983, 221).

[62] See Foucault (1983, 221).

[63] Foucault wrote this part of the text in French and it was translated by Leslie Sawyer to English. In the English translation, it says ‘essential freedom’, which is a wrong translation of the original ‘‘antagonisme’ essential’, cf. Foucault (1994b, 238).

[64] See Foucault (1983, 222).

[65] This tension between generalization and possibility is visible for instance in these two sentences: ‘For, if it is true that at the heart of power relations and as a permanent condition of their existence there is an insubordination and a certain essential obstinacy on the part of the principles of freedom, then there is no relationship of power without the means of escape or possible flight’ (Foucault 1983, 225). Except for the talk about a possible flight at the end of the citation, this is a social theoretical generalization: resistance is always possible. He continues: ‘Every power relationship implies, at least in potentia, a strategy of struggle’ (Foucault 1983, 225). This is only a generalization of the possibility of resistance, not of resistance itself. Foucault does not notice this difference and seems to treat both statements as meaning the same.

[66] Cf. Pettit (1999), Pettit (2011), List and Valentini (2016).

[67] When seen in the light of Honneth’s systematization of three concepts of freedom (negative, reflexive and social), the notion of freedom as non-domination falls into the first, negative category, as it does not take ‘inner’ unfreedom as a political problem at all, but only external constraints.

[68] This plain connection between probability and modal robustness is a simplification when seen in the light of the complex debate around modality and modal robustness, but sufficient for the present argument, cf. Menzel and Christopher (2018).

[69] See List (2016, 211).

[70] I differentiate here between critique and resistance, as not each act of critique leads to resistance. After all, the critique could reach the conclusion that a certain norm is just fine, or there might be strategic reasons to refrain from resistance. As this difference is absent in Foucault and his commentators, I mostly use the concepts of critique and resistance interchangeably.

[71] See List (2016, 208).

[72] See, for example, Hechler and Philipps (2008).

[73] For former discussions of Foucault and democracy see Connolly (1998), Walter (2008) and Rosenthal (2018).

[74] Cf. Marchart (2007), Mouffe (2008) and Wingenbach (2011).

[75] See Schubert (2020a, forthcoming). The notion of freedom as critique is also helpful to explain the normative position of an intersectional critique of privileges, which is based on a particularist political ontology and a perspectivist epistemology, and is nevertheless rooted in the universalist project of equal freedom. Cf. Schubert (2020b).

[76] Cf. Hobbes (2000), Berlin (2002) and Rawls (1999, 176–80).

[77] Cf. Pettit (1999), Pettit (2012), Pettit (2011) and Lovett (2010). From the perspective of classical political theory, the self-marking of these theorists as ‘republican’ is rather misleading, because they use a fundamentally negative concept of freedom and could be better described as a specific kind of liberals.

[78] Cf. Allen (2008).

[79] Cf. Habermas (1995).

[80] Cf. Honneth (2011), Rousseau (2004), Arendt (1963), Arendt (1970) and Taylor (1992).

[81] Cf. Taylor (1995).

[82] Cf. Honneth (2003).

[83] See, for example, for a rereading of Honneth from Ranci‘ere’s perspective, Deranty (2003), see for approaches between Honneth and (among others) Bankovsky (2012).

[84] Cf. McIntyre (2019).