1 Introduction

As dominant liberal conceptions of the relationship between rights and freedom maintain, freedom is a property of the individual human subject and rights are a fundamental mechanism for protecting that freedom—whether it be the freedom to speak, to associate, to practise a certain religion, and so forth [36]. Rights, according to these kinds of accounts, are protective of a certain zone of permitted or valorised conduct and they function either as, for example, ‘side-constraints’ upon the actions of others [48] or as ‘trumps’ over governmental or community goals [10]. In such accounts, of course, the emphasis is placed predominantly upon the forms of illegitimate power against which rights protect the individual and her freedom, whether that power be figured as the incipient trespasses of other subjects or the overweening attentions of the state. Such accounts famously do not themselves take much account of the multiple ways in which rights also function as forms of power, often delimiting the courses of action that a putative rights-holder can take and affecting the manner of that right’s exercise, indeed often in the very name of freedom itself.

Of course, there is a sizeable critical literature which does address itself exactly to these kinds of questions, most notably from within the radical traditions of Marxism and critical legal theory, in which rights are viewed in terms of the relations of production, consumption and exploitation that they institute and reproduce between legal subjects [for a recent survey of the literature, see [2]. For various reasons, discussions of the work of Michel Foucault have not figured as prominently in critical engagements with rights and with the politics of rights. Here I do not propose to enter into debates surrounding Foucault’s engagement with, or failure to engage with, law as an object of study [32, 35], nor with the emergent literature on Foucault’s various deployments of rights, indeed even of human rights, in his later work [29–31, 37–39, 44, 46, 52, 53, 63, 64]. Rather, what I want to do in this paper is to articulate and defend the view that Foucault’s work, both on rights specifically (of which, admittedly, there is not much) and on power relations more broadly (of which, famously, there is much), reveals the political ambivalence of rights. For Foucault (and for some of the post-Foucaultian scholars whose work I shall draw upon in what follows), rights are both political tools for the contestation and alteration of mechanisms of power and simultaneously mechanisms of capture and inscription; both disciplinary and governmental apparatuses, that is, which conduct the behaviour and go to constitute the very identities of those who deploy them. Far from being unproblematic tools for the protection of the subject’s freedom, rights emerge in this account as conflicted and ambivalent mechanisms which subjectify and regulate the would-be subject of rights even as they claim to protect that subject or to enlarge the domain within which that subject moves.

In the first part of this paper I develop a Foucaultian account of rights along precisely these lines. Whilst developing this account by reference to Foucault’s direct engagements with rights (and with power relations more broadly) I hope also to illustrate it by reference to several different examples drawn from areas as diverse as the disciplinary constitution of gender and cultural identity via rights through to the constitution of the identity of ‘the refugee’ in law. In the final part of the paper I make a return to the idea of freedom in Foucault’s work and link the view of rights as ambivalent mechanisms developed in the first part of the paper to a certain conception of freedom in his work. Let me start now, however, by sketching the contours of what I have just called the political ambivalence of rights in Foucault’s work. This discussion will take as its starting point Foucault’s own, and to my mind by no means straight-forward, deployment of rights discourse in his later work.

2 The Ambivalence of Rights

Whilst for certain commentators the Foucault of the 1970s arrives at some fairly ‘dismal conclusions with regard to the potential of rights as a language of political contestation or resistance’ [45, p. 162], for yet others his position on rights amounts to something much more extreme: the ‘denial of any potential political value’ to rights per se [35, p. 63]. And yet, as several critics have observed [see for example: [54, 55], Foucault’s oft-remarked ‘problem with regard to the language of rights’ [52, p. 43], a problem (indeed, a set of inter-linked and related problems) amply discussed in his genealogies of discipline, bio-power and governmentality in work of the 1970s, by no means prevents him from employing the language of rights himself in a range of very disparate contexts somewhat later in his career. As is by now quite well-observed, Foucault makes increasing resort to the vocabulary of rights in his philosophical, journalistic and political interventions in the late 1970s and early 1980s. There, for example, he affirms a ‘right to compensation, care and damages’ [24, p. 374], a right to asylum [21, p. 427], a right to suicide [24, p. 380], a right to choose one’s own sexuality [quoted in [38, p. 30], and indeed several other rights [53, p. 97 (n1); 53, pp. 269–270].

Of course, from a perspective which interprets the use of any given political discourse in terms of either acceptance or rejection of that discourse, Foucault’s invocation of rights are admittedly difficult to make sense of in light of his critiques of sovereign right and of the discourse of rights sustained in his earlier work [see for example: [17, pp. 221–222; lectures 1 and 2 of [26]. On my reading, Foucault’s own subsequent deployments of rights discourse bear a complicated and conflicted relationship to standard liberal articulations of rights—his is a difficult and tactical attempt to inhabit, deploy and yet critically depart from (and deform) the orthodox idioms of ‘rights talk’ (a task which Judith Butler describes in a related context in terms of ‘being implicated in that which one opposes,… [that is, a] turning of power against itself to produce alternative modalities of power’ [[7], p. 241]). Indeed, as Foucault reminds us of discourse, it ‘transmits and produces power; it reinforces it, but also undermines and exposes it, renders it fragile and makes it possible to thwart it’ [13, p. 101]; that is, the space of discourse represents not a closed systematicity but rather an unstable (and promising) ‘dispersion’ [12, p. 72], and his critical re-deployment of rights discourse is precisely an attempt to take up this promise in a range of different ways and directions.

There is unfortunately not space here fully to develop this understanding of a critical and creative affirmation of rights in Foucault’s work [for more, see 31], but what I do want to elaborate on in the present context is one aspect of this enterprise, namely the theorisation of the political ambivalence of rights. And, to further refine my topic, the particular dimension of this ambivalence that I am going to discuss is the way in which rights function both to allow and to disallow agency, to open up possibilities but at the same time to circumscribe and to channel those possibilities in particular ways. The critical point, of course, is that this structural tension and irreducible ambivalence of rights—again, in the terms introduced above, their being simultaneously protections of freedom against power and forms of power which work to forestall freedom themselves—is elided in standard liberal accounts in both their idolatrous and pragmatic, maximalist and putatively minimalist guises. The present account hence intends neither to celebrate nor condemn rights per se but rather, via Foucault, critically to open up a debate concerning the limits, possibilities and in-built constraints of rights as a modality of politics. That there are other—pre-existing and as-yet-unimagined—ways of conducting politics and contesting the contemporary political de-formations of sovereignty, capitalism, and so forth, and that these forms of politics take place at some remove from (indeed, even tension with) rights, hardly needs reiterating. This paper does not counsel a war-weary return to rights simply with a better appreciation of their ambivalences (as if rights were the only political tools available to us today). Rather, it aims to excavate what I have been calling the ambivalent dimension of rights as part of a wider critical project of appreciating what rights contain, politically—in both senses of that phrase.

Let me turn now to some contemporary work in the philosophy of rights which does foreground this dual, politically ambivalent dimension of rights, in order to set up my argument. In a chapter of his recent book, Rights, entitled ‘Rights as Conduits’, the philosopher Duncan Ivison describes the work of both Marx and Foucault on rights in terms of this topographical-mechanical metaphor of the ‘conduit’. He writes of rights:

But they can also, I shall argue, be understood as conduits, that is, as modes for distributing capabilities and forms of power and influence and thus shaping behaviour as much as constraining it. They key idea here is that rights are often implicated in various relations of power as much as they are a means of criticizing them…. So rights are implicated in relations of power not only because power must be exercised to enforce them… but also because rights themselves represent a distinctive relation of power. [37, p. 180]

What might it mean for rights to represent, as Ivison puts it here, a ‘distinctive relation of power’?Footnote 1 I want to take up this question from the perspective of what Foucault famously calls disciplinary power and governmentality; that is, to try to show how for Foucault rights function both as disciplinary mechanisms and as elements which enfold the individual rights-holder or rights-claimant into wider apparatuses of governmentality, and which thereby regulate populations via the juridical freedoms sustained by rights. But before I do so, and in order to establish what we might call the first (and ideologically privileged) dimension of the ambivalence of rights (namely, the way in which rights can protect or defend freedom, or enlarge the domain of possible actions of a subject), I want briefly to discuss how for Foucault rights can indeed function as political tools both to protect certain behaviours of the subject but also to establish the possibility of other behaviours or to contest political or legal arrangements. Of course, and as we shall see in a moment, this is not the only thing, or perhaps not even the most politically important or interesting thing, that rights do, but pace those commentators discussed above, in whose estimation Foucault utterly rejects rights, it is clear both from a reading of Foucault’s own occasional invocations of rights, and of his genealogical work of the 1970s, that rights can be used to protect, and expand the domain of action of, subjects. Obviously, Foucault’s own usage of ‘rights talk’ functions in this way, whether that usage is intended to protect conventional or pre-existing relations or to argue for the establishment of new, as-yet-unaccepted relations, such as would be entailed, for example, by his advocacy of a ‘right to suicide’ [24] or his well-known articulation of the ‘rights of the governed’ in the context of international affairs [20].

We might then think of rights as ‘instruments’ or ‘tools’ in the service of either protecting such relations or of (re)producing them. And this is precisely how Foucault discusses rights in his lecture course at the Collège de France in 1976, ‘Society Must be Defended’. Describing the emergence of what he calls ‘the first historico-political discourse on society’ [26, p. 49], Foucault discusses how the dissident speaker of this discourse ‘cannot, and is in fact not trying to, occupy the position of the jurist or the philosopher, or in other words the position of a universal, totalizing, or neutral subject’ [26, p. 52; see also 15, p. 126] but rather

speaks the discourse of right, asserts a right and demands a right. But what he [sic] is demanding and asserting is ‘his’ rights – he says: ‘We have a right.’ These are singular rights,… both grounded in history and decentred from a juridical universality [26, p. 52].

The subject who speaks thus wields ‘a truth-weapon and a singular right’ [26, p. 54]. Foucault traces examples of this ‘historico-political discourse’ in both England and France from the seventeenth century onwards. As he puts it:

In England it was one of the instruments used in bourgeois, petit-bourgeois – and sometimes popular – struggles and polemics against the absolute monarchy, and it was a tool for political organization. It was also an aristocratic discourse directed against that same monarchy [26, pp. 49-50].

But, of course, and turning now to the other dimension of the political ambivalence of rights, namely the ways in which—whether that be on a disciplinary or a governmental scale—rights function to position, constrain and conduct those who deploy them, it is clear that the would-be rights claimant is not in a simple position of mastery or instrumentalisation vis a vis rights. Rather, rights form part of disciplinary and governmental networks which affect, even constitute, that subject. Famously, for Foucault,

the individual is not, in other words, power’s opposite number; the individual is one of power’s first effects. The individual is in fact a power-effect, and at the same time, and to the extent that he [sic] is a power-effect, the individual is a relay: power passes through the individuals it has constituted [26, p. 30].

According to such an understanding, then, we must not ask ‘Who is the subject of rights?’, if such a question presupposes a stable, intentional, effective subject who stands instrumentally before rights, who ontologically pre-exists them, and whose knowable interests or freedom is thus protected by them.Footnote 2 Rather, we must ask, with Foucault, ‘Who is the subject-effect of rights?’, or, in a more classically Foucaultian idiom perhaps, ‘How do rights subject those who rely upon them?’,Footnote 3 ‘What kinds of subject do rights regimes rely upon, bring into being and iteratively (re)produce?’, and ‘Whose subjective behaviour is affected by rights and who is correlatively excluded from the domain of rights?’.

The political theorist Wendy Brown has profitably taken up some of these questions. In the specific context of a discussion of identity-based rights (she is writing here of gender, but her insights are more broadly applicable), Brown brings attention to what she calls the paradoxical dimension of rights discourses:

The paradox within this problem is this: the more highly specified rights are as rights for women, the more likely they are to build that fence insofar as they are more likely to encode a definition of women premised upon our subordination in the transhistorical discourse of liberal jurisprudence. Yet the opposite is also true… [for] the more gender-neutral or gender-blind a particular right (or any law or public policy) is, the more likely it is to enhance the privilege of men and eclipse the needs of the women as subordinates [[4], p. 232, emphasis added].

As she goes on to remark, the second element of the paradox (we might call this element the false abstraction or generality of rights) is the one thematised by Marx—and, in terms of feminist jurisprudence, pursued by Catharine MacKinnon [42]—whilst the first element ‘might be understood as the problem that Foucault painted most masterfully in his formulation of the regulatory powers of identity and of rights based on identity. To have a right as [for example] a woman is not to be free of being designated and subordinated by gender’ [4, pp 231–232, emphasis in original]. Rather, in this context, rights precisely become the vehicle for the imposition and regulated performance of gender. A right, as Brown puts it elsewhere in a claim advanced as a correction to Foucault, is not simply a mechanism that converges with disciplinary power but is itself, ‘from the beginning a potentially disciplinary practice’ [3, p. 99].

As I mentioned above, the disciplinary determination of identity in and through regimes of rights is not a phenomenon limited to the sphere of gender relations. In the context of a discussion of what he calls a ‘reluctant critique of legal identity politics’, Richard T Ford draws attention to way in which assertions of racial or cultural difference in and as right become the occasion for the legal and bureaucratic instantiation of very particular, often reactionary and exclusive, cultural norms of what it means to be a member of that group. He counsels:

The point here is that the discourse of racial difference can take on a life quite independent of the good intentions of those advancing cultural identity rights. The nature of rights discourse is that anyone can assert a right and have it tested in court. But the ill effects of the codification of bad definitions of group culture and identity will not be limited to the litigant asserting the right: they will instead be deployed to regulate all members of the group [11, p. 56].

One need not necessarily subscribe to the voluntaristic and normative language of good and bad intentions, good and bad definitions of culture, and so forth, in the above discussion, in order to agree with Ford’s contention that the invocation of rights discourse can redound upon the would-be claimant (and claimants to-come) in ways which profoundly challenge the directionality of the liberal rights narrative of a subject standing before the law. Here, to the contrary, it is the law which configures that subject. Ford again, this time writing about the case of Regents of University of California v Bakke (the United States Supreme Court decision concerning affirmative action programs for university admission, which in part determined that ‘only by highlighting their own distinctiveness could minority students justify their presence in the universities that admitted them [by linking admission to normative standards of ‘cultural diversity’ within institutions]’ [11, p. 46, emphasis added]:

[B]y altering the character of the institutional treatment of race, it also altered the incentives surrounding racial identity and thereby altered performance of racial identity, at least among those directly affected by the institutions [11, p. 46, emphasis added].

For Ford, then, rights regimes can specify and hence juridically entrench a very particular racial script which would-be claimants must adhere to and iteratively (re)produce, to the detriment of themselves and others (or, at any rate, to the exclusion of other possible ways of performing that identity). As Brown herself puts it, this disciplinary function of rights—whereby rights mandate particular forms of identification, belonging, disposition, comportment, and so forth—‘may subject us to intense forms of bureaucratic domination and regulatory power even at the moment that we assert them in our own defense’ [3, p. 121, n 41].

As a final example of such ‘bureaucratic domination and regulatory power’, consider in light of this identity-producing function of rights discourse the question of the refugee seeking the right of asylum. In few areas of rights discourse is the imperative to narrate one’s own experience so juridically and institutionally entrenched, and ethically and politically fraught, as in the injunction directed at the refugee claimant to divulge their personal story of persecution and statelessness. This process might perhaps appear as a neutral exercise in simply fitting one’s pre-existing factual reality into a relatively stable and accommodating (or otherwise) set of legal criteria. Deborah Anker writes in this vein that ‘[t]he refugee definition does not fix a refugee claimant’s individual or group identity. Rather, it emphasizes the persecutor’s perception of the refugee claimant’s social status or opinion… [and] it does not force a choice of one particular ground of persecution, as claims can be based on any combination of the five grounds [of race, religion, nationality, membership of a particular social group, or political opinion contained in the 1951 Refugee Convention]’ [1, p. 153, emphasis in original]. But other commentators are more attuned to the ineluctable narrativity of the refugee determination process and the multiple constraints embedded within it—the ways in which the process solicits and conditions the types of stories told by refugee claimants, to whom and for what purpose:

The temporal and ethical elements of the refugee’s self-narrative are nowhere more apparent than in the retelling of the story which the determination process requires. From the moment they arrive in a country of refuge, a refugee must begin the process of telling and retelling their story – to the authorities, to their legal representative, to torture and trauma services, to welfare agencies, in written and oral form, on each occasion translated, summarised, reworked and massaged by the recipient, and all the time creating an ever growing bureaucratic record of their experiences before and during flight [[65], p. 12].

Often such a constraint is framed in terms of a distortion or an alienation of a refugee’s original, authentic and unmediated experience. ‘“[W]hy do legal frameworks distort the experiences of refugee women?”’, asks Natalie Oswin [51, p. 355]? But what if that very experience (understood as an iterative artefact, produced and reproduced, reconstituted in reflection) is itself partly the result of the constraints of refugee law, rather than a pre-existing substance which either succeeds or fails in finding itself reflected adequately in law? For Cynthia Hardy, that is, ‘the refugee subject is a product of the processes of determination that lead to his or her classification as well as the broader discourses that impinge on and overlap with refugee discourse. There is, then, no autonomous subject [who enunciates a prior experience]: a refugee only exists insofar as he or she is named and recognized by others’ [34, pp. 476–467]. Bluntly, then, we might say with Roger Zetter that the refugee is ‘one who conforms to institutional requirements’ [66, p. 51].Footnote 4 And now with Foucault, we can see how these institutional requirements of refugee law and its ancillary bureaucratic apparatuses function according to a logic very similar to that form of avowal described by him as being integral to modern methods of political individualisation: the compulsory extraction (read: production) of a subject’s truth through the mechanism of the confession [see especially: 13].Footnote 5 Such an extraction-production of course takes place in the context of entrenched power relations and is anything but a neutral process [16, p. 39]. As Foucault puts it in the first volume of the History of Sexuality:

For a long time, the individual was vouched for by the reference of others and the demonstration of his [sic] ties to the commonweal (family, allegiance, protection); then [in modernity] he was authenticated by the discourse of truth he was able or obliged to pronounce concerning himself. The truthful confession was inscribed at the heart of procedures of individualization by power [[13], pp. 58-9].

So, from the foregoing discussion of rights to gendered and cultural identity, and the right to asylum in the form of claims for refugee protection, we can see how the assertion of the subject of right is simultaneously a subjection. For Foucault, the term assujettissement (translated in the below passage as ‘subjection’) comprises this dual meaning of being constituted as a subject, or of occupying a particular subject position, at the ‘cost’Footnote 6 of a (routinely disavowed) subjugation. Rights are disciplinary vehicles of precisely this relation of assujettissement, or subjection, in that our assumption of the valorized and putatively autonomous position of rights-claimant or rights-holder is simultaneously our entry into regimes of power-knowledge which bind us to particular truths, ways of thinking and acting and being.

This disciplinary production of the individual subject of rights is paralleled—actually, as Foucault might say, it exists ‘orthogonally’ to [26, p. 253]—a range of mechanisms at the level not of the individual to be disciplined but rather at the level of the population to be governed. To conceive of rights as an incident of wider mechanisms of governmentality in this way might at first blush seem contradictory; or, at any rate, it may seem to run counter to a predominant interpretation of Foucault which opposes the political technology of sovereignty (with its correlative juridical subjects of right) to the emergent biopolitical technologies of modernity, which pointedly took as their object not the sovereign subject of right but rather the population.Footnote 7 As Foucault makes clear, however, the relation of these different technologies is a matter not of historical succession but rather of co-implication in the present:

So we should not see things as the replacement of a society of sovereignty by a society of discipline, and then of a society of discipline by a society, say, of government. In fact we have a triangle: sovereignty, discipline, and governmental management, which has population as its main target and apparatuses of security as its essential mechanism [[27], p. 107-8].

My question here, then, is how rights might figure in the governmental management of the population?Footnote 8 What are the governmentality-effects of rights? A common understanding might be, again, to pose rights in opposition to mechanisms of governmentality and thereby to insist upon the former’s critical potential vis-à-vis the latter. ‘The vocabulary of right and practices of rights claims,’ insists Rosemary Coombe, ‘continue to afford new resources and opportunities for articulations at and of governmentality’s limit(s) and thus spaces of politics, critical insight and possible transformation’. But, she goes on to counsel, ‘[n]ot every assertion or activity couched in the vocabulary of rights articulates the space of governmentality’s limit; we must be continually attentive to the ways in which rights achieved entrench their own regimes of governable regimes and subjects’ [9, pp. 285, 287]. Here again we see the ambivalence or the dual character of rights—that is, the ability of rights not only to perform critical or contestatory roles but also, and at the same time, to entrench and extend regimes of governmentality. Having argued in the above passages that rights perform a disciplinary function in the sense understood by Foucault, I want in what follows to briefly sketch one way in which rights regimes perform this function of transmitting relations of governmentality that are simultaneously enabling and disabling. The aim is not to provide an exhaustive or even necessarily a detailed account but simply to begin to outline how rights function within the domain of governmentality.Footnote 9

For Foucault, the transition from a concern with the disciplinary production of subjectivity to a model centred upon the supposedly self-constitutive capacities of the ethical subject in work of the early 1980s was achieved via the analytic of governmentality. This latter body of work—conducted largely in Foucault’s lecture courses of 1978 and 1979 at the Collège de France: Security, Territory, Population and The Birth of Biopolitics [27, 28]—began to emphasize the multiple ways in which subjects, situated within regimes of power/knowledge, were led to work upon themselves in the exercise of their autonomy. As Thomas Lemke puts it, the ‘problematic of government [simultaneously of oneself and others]’ is the ‘missing link’ which articulates Foucault’s analysis of ‘technologies of domination’ with ‘technologies of the self’ [41, p. 50]. However, this change of emphasis from the disciplinary production of subjectivity to the more diffuse notion of self-regulation via technologies of the self did not signify a lack of concern with the way in which subjects were conducted by others. Crucially, the autonomy of the governable subject, the free subject, was not unconditional but was rather—in Nikolas Rose’s apt phrase—a ‘regulated autonomy’ conditioned to align with governmental imperatives. For my purposes here, these ‘new mechanisms to link the calculations and actions of a heterogeneous array of organisations into political objectives governing them at a distance through the instrumentalisation of regulated autonomy’ [59, p. 57]; that is, these emergent assemblages of governmentality, include within their various tactics the deployment of rights. Let me return to Foucault’s lecture courses of the late 1970s in order to illustrate this claim. This discussion will raise the question of freedom which I will then conclude by problematising, or at least reframing, in the final section of this paper.

In the opening lectures of Security, Territory, Population Foucault begins to explain the differences in operation between systems of legality, disciplinary mechanisms, and apparatuses of security. That is, between law, discipline and governmentality. Whereas law, on Foucault’s account here, operates according to a negative logic of the permitted and the forbidden which prohibits certain behaviours (‘a system of legality always focuses with greatest precision on what is to be prevented’), discipline functions in a more productive register by specifying the obligatory (‘[a] good discipline tells you what you must do at every moment’) [27, p. 46]. Both intervene either at a formally abstract level (law) or at an embodied material level (discipline) with individual behaviour. By contrast, governmentality concerns itself directly neither with the juridical nor with the disciplinary subject per se; rather, apparatuses of security function at the level of the population, ‘standing back sufficiently,’ [27, p. 46], as Foucault says, to perceive the ‘natural’, immanent regularities of that which is to be governed. Population comes to be conceived as having a life and a specific density of its own, to which the techniques of security must adapt themselves and upon which they must begin to operate obliquely, at a distance: ‘It is a set of elements in which we can note constants and regularities even in accidents… and with regard to which we can identify a number of modifiable variables on which it depends’ [27, p. 74].

One of the examples Foucault gives to explain the difference between an interventionist and individualising disciplinary approach and a predominantly laissez-faire and population-centred security approach is the question of grain shortages in the seventeenth and eighteenth centuries. Scarcity and the economic and political crises it could engender are used here as a means to elucidate the difference between discipline and governmentality. For Foucault, ‘the disciplinary police of grain… is in actual fact centripetal. It isolates, it concentrates, it encloses, it is protectionist, and it focuses essentially on action on the market or on the space of the market and what surrounds it.’ In short: ‘[D]iscipline regulates everything,’ such that even ‘the smallest things… must not be abandoned to themselves’. On the other hand, for apparatuses of security:

laisser-faire is indispensable at a certain level: allowing prices to rise, allowing scarcity to develop, and letting people go hungry so as to prevent something else happening, namely the introduction of the general scourge of scarcity…. The function of security is to rely on details that are not valued as good or evil in themselves, that are taken to be necessary, inevitable processes, as natural processes in the broad sense, and it relies on these details, which are what they are, but which are not considered to be pertinent in themselves, in order to obtain something that is considered to be pertinent in itself because situated at the level of the population [27, p. 45].

And key to enabling the circuits and circulation of people, things, capital and commodities which emerge as pertinent regularities within the domain of the population is the notion of freedom. Freedom, as Foucault puts it, ‘is nothing else but the correlative of the deployment of apparatuses of security’ [27, p. 48]:

An apparatus of security, in any case the one I have spoken about, cannot operate well except on condition that it is given freedom… : no longer the exemptions and privileges attached to a person, but the possibility of movement, change of place, and processes of circulation of both people and things. I think it is this freedom of circulation, in the broad sense of the term, it is in terms of this option of circulation, that we should understand the word freedom, and understand it as one of the facets, aspects, or dimensions of the deployment of apparatuses of security [[27], pp. 48-9].

Freedom here is hence not opposed to government but is one of its central modalities of operation. However, whilst Foucault insists in the quotation above from Security, Territory, Population that ‘exemptions and privileges attached to a person’ are not what is important in terms of this operative freedom, and whilst again in The Birth of Biopolitics he maintains once more that ‘the freedom that the physiocrats and Adam Smith talk about is much more the spontaneity, the internal and intrinsic mechanisms of economic processes than a juridical freedom of the individual recognized as such’ [28, p. 61], it is clear that this distinction between a juridical freedom of the individual and the freedom of the market or the population can only serve an initial, heuristic function. As Foucault was himself at pains to point out, ‘our [Occidental] political rationality’ is characterized by ‘a constant correlation between an increasing individualization and the reinforcement of [a] totality’ [22, p. 417; also [23, p. 325]. The task is thus to uncover how what might not be pertinent ‘as such’, namely the juridical level of the individual, is nevertheless correlated with what is more biopolitically pertinent, namely the enabling of productive flows and exchanges of people, things, commodities, etc. at the level of the population. Individual juridical freedoms of movement, of property, of association, of speech, and so forth, are plainly functional to what Foucault calls, several pages later in The Birth of Biopolitics, the governmental production of freedom:

Broadly speaking, in the liberal regime, in the liberal art of government, freedom of behaviour is entailed, called for, needed, and serves as a regulator, but it also has to be produced and organized. So, freedom in the regime of liberalism is not a given, it is not a ready-made region which has to be respected, or if it is, it is so only partially, regionally, in this or that case, etcetera. Freedom is something which is constantly produced. Liberalism is not acceptance of freedom; it proposes to manufacture it constantly, to arouse it and produce it, with, of course, [the system] of constraints and the problems of cost raised by this production [28, p. 65].

Just as Foucault argues that sexuality is a hinge or a nodal point articulating disciplinary and biopolitical technologies of rule [26, pp. 251–252], so too do rights function both as a juridical guarantee and as an apparatus of security attuned not so much to the rights of the individual (although taking this as its starting point) but rather to the circulations and transactions engendered at the level of the population to be governed. In societies of the governmentalized type, populations are governed best—fluidly, aptly, according to their ‘natural’ propensities—when its constituent members are equipped with rights which enable commerce (in both a broad and a narrow) sense between them. ‘[T]he right of individuals,’ writes Foucault, ‘is an element that has become indispensable to governmentality itself. Henceforth, a condition of governing well is that freedom, or certain forms of freedom, are really respected’ [27, p. 353]. In this sense, via rights, individuals can indeed be ‘governed through their freedom’ [60, p. 201].

3 Conclusion: Freedom and the Work of Rights

If, as I have argued in this paper, rights cannot unproblematically be opposed to power but must rather be understood both as protective and as destructive of a certain freedom, both as enabling and as disabling of conduct, then this amounts to what I have been calling the political ambivalence of rights. What, then, does this mean for the freedom of the subject? Are subjects of rights doomed endlessly to find themselves reinscribed disciplinarily or conducted in a governmentalized sense by apparatuses of security, their very ‘freedom’ the vehicle for their rule?

Of course, Foucault refuses the presuppositions inherent to such a juridical formulation of subjective freedom. As he puts it in the extract from The Birth of Biopolitics, quoted above, ‘freedom is not a given… it is not a ready-made region which has to be respected’ [28, p. 65] Freedom for Foucault is not a property of the individual subject [50, p. 188], and there are for him consequently ‘no [subjective] spaces of primal liberty between the meshes of [power’s] network’ [14, p. 142]. And yet it does not follow that Foucault, refusing this traditional conception of freedom (namely, as the freedom of the subject), thereby refuses to talk of freedom per se. Indeed, in his later work ‘freedom’ comes to designate the condition of possibility for ethical and for power relations. ‘Freedom is the ontological condition of ethics,’ he writes in ‘The Ethics of the Concern for Self as a Practice of Freedom’ [19, p. 284]. And in ‘The Subject and Power’ he asserts: ‘[F]reedom may well appear as the condition for the exercise of power (at the same time its precondition, since freedom must exist for power to be exerted, and also its permanent support, since without the possibility of recalcitrance power would be equivalent to a physical determination’ [25, p. 342]. Foucaultian freedom in these later renditions does not refer to an ineluctable property of human nature or human essence. Rather, as Johanna Oksala explains, it refers to ‘the ontological contingency of the present,’ understood as ‘the moment of the unexpected as opposed to the normalized, the unforeseen as opposed to the determined’ [50, p. 188]. Freedom as precondition of, or as condition of possibility for, power is nothing other than the constitutive instability and possibility of reversibility of power itself, of power’s always potentially being otherwise, of its never being ultimately determined. Here freedom is neither counterposed to power (as in the standard liberal formulations I have been critiquing) but nor is it reducible to an ‘internal modality of power relations’ [56, p. 29] (as in the understanding of freedom as an aspect of governmental rationalities, drawn upon in the preceding section). Rather, as Foucault stresses in both the formulations extracted above, freedom is the necessary precondition of power. Where there is power, for Foucault, there must also always—and precisely to the extent that there is power (and not an immoveable state of utter domination, [25, p. 342; 19, p. 283])—be freedom: freedom, that is, understood as the possibility of movement, contestation, resistance [13, p. 95]. Freedom can neither be located in the subject nor in the surrounding structures or networks of power, but is brought to light fleetingly in their agonal and dissensual interaction.

And, to conclude now by relating this concept of freedom back to my primary theme of rights, this bringing to light, this exposure and actualisation of freedom, is (or, better: can be) the ambivalent work of rights. Rights are both the instantiation of disciplinary or governmental regimes but are simultaneously, but not necessarily in the exact same measure, possible tools for their reform, contestation or rupture. In that (often foreclosed, frequently attenuated) possibility resides, I would argue, the Foucaultian connection between freedom and rights.