Agency, power, modernity: A manifesto for social theory Open Access

Over the course of an intellectual arc lasting approximately 40 years, European and Anglophone social theory has articulated an image of the person as an actor enmeshed in a series of situations, connections, hierarchies, patterns, meanings, and historical conjunctures.¹ In so doing we have rearticulated the antinomies of classical social theory (individual/collective; Gemeinschaft/Gesellschaft; use value/exchange value) in new and elaborated form. We have shown, in great detail and with significant aplomb, that persons make history, but not in circumstances of their own choosing. This elaboration has been intellectually progressive, whatever its politics. Particular attention to the melding of body and mind in action, the construal of the self via cultural meanings, and the navigation of a hyper-differentiated world of (late, perhaps liquid) modernity has made it possible for a sophisticated understanding of human sociality and historicity to emerge.

Particularly in and through the disciplines of sociology and anthropology, whose practitioners tend to be uncomfortable replacing people possessed of concrete bodies and idiosyncratic meanings with abstract decision-makers or voters with preferences, social theorists have gravitated towards questions concerning the intersection of the objective world beyond the individual and the subjective inner life of that same individual. In this image, persons, guided by certain habits and using certain repertoires and understandings, step onto the stage of social life to act, and in so doing ignite the contingencies of history and society that caused Max Weber to identify the social sciences as sciences of concrete reality (Weber, 1949). Furthermore, we are (and should be) obsessed in social theory with the causes of social change and dynamism, and it should surprise no-one that sociological research coming out of the Continent and the Anglophone world sublimates the question of free will descended from the Western philosophical tradition. And so, as it has developed its multifarious language for the study of the behaviour of human individuals in interrelation with others and their various environments (natural, built, economic, symbolic …), social theory has found itself inflected by a concern with ‘agency’. Yet here, despite all the intellectual progress, we have tripped up.

What, really, is the relationship of action by these individuals – or their group or organisational quasi-equivalents – to agency? In social theory, we frequently suggest that sociology is the science of social action, while also suggesting that individuals (possessed of subjectivities, bodies, and habits), or organisations (possessed of collective decision-making processes and variable capacity to take advantage of political opportunities), have something called agency that upsets the causal determinism implicitly evoked when the goal of social-scientific explanation is articulated. There is difficulty here. We have, in social theory, found ourselves spinning our wheels, unable to articulate an account of agency outside of our intuitions about certain key actions by individuals and groups that we (are right to) find paradigmatically interesting – for example, demands for justice by social movements who seek to win concessions from modern states, the subversion of oppressive social norms by creative individuals, or the counter-assertion of lived reality over and against dominant discourses and prevailing folk explanations of inequality, ill health, and social strife.

Adams (2011) identified this tendency as the recurrent use of a ‘heroic’ notion of agency, which combines a romantic concept of will with a Kantian ambition for individual autonomy. She despairs at its use as an ‘intellectual bludgeon’ – it is always possible to ask ‘What about agency?’ in almost any intellectual setting in human science, and such questions can be interpreted in myriad contradictory ways while simultaneously attesting to the speaker's earnest subscription to normative aspirations for the self-determination of individuals and groups. And so, this concept of agency informs our sociohistorical investigations, repeatedly if sometimes implicitly parsing the messy social world into ‘structure’ and ‘agency’. A series of meanings then obtain, via informal convention, associating both ‘action’ and ‘agency’ together with the will to resist large structures, and the possession, by persons, of the ability to choose to act otherwise, or choose not to act at all. This takes place despite – or alongside – the appreciation of well-known critiques (Alexander, 1995; Emirbayer, 1997) of the classics of ‘structure and agency’ theoretical writing (e.g. Giddens, 1984; Sewell, 1992). In a different space of argument, the intriguing issue of non-human agency has been repeatedly raised and discussed; but the relationship between these discussions and the central problems of social theory that such discussions disavow is far from clear.

The overall result is a certain amount of cacophony. We have found, furthermore, that these informal conventions around the meaning of agency encourage reversion to a set of modern antinomies that have a gendered bias and partially encode the relations of power that they propose to analyse and criticise. Confronted with these intellectual problems, the authors of this paper and the participants in their research group have found it useful to ‘begin again’ with Aristotle, and in particular, with a rather stark distinction between action and agency, even if the two are ultimately shown to be related. In this portentously titled manifesto we (1) set out the distinction between action and agency, discuss its implications for our thinking, and introduce some new vocabulary, (2) ground our arrival at this problematic via a brief, critical discussion of the theorisation of power in social theory, (3) articulate a creative synthesis of certain texts we have found especially useful to read in the light of our research group's emerging concerns, (4) reveal what can be seen in the arc of modernity by using the vocabulary we propose, and, finally, (5) point the way towards further areas of study, given the (necessarily tendentious and overbold) claims contained herein. Throughout, our language is slightly coloured, and our claims overreach. This is precisely our intention, for our goal is to provoke argument and dialogue, and thus enter into a new research programme for the twenty-first century. After all, progressive research programmes in the human sciences only emerge through honest contention.

Consider action. Individual humans have projects-in-the-world; as Arendt (2013, pp. 57, 55) writes, people leave the ‘multiplication of one's own position’ within the space of the intimate to engage the larger world constituted by others and others’ perspectives, and they do so ‘because they want[ed] something of their own or something they had in common with others to be more permanent than their earthly lives’. Individually, persons aim to educate themselves, to earn a position, to gain recognition; collectively, they form associations, rebuild the environment, constitute institutions, forge or remake legal, social, and economic relations. In so doing, as Arendt stresses, such persons-involved-in-projects (or ‘sharers of a common world’) are tuned into the past and the future, as well as the present, and thus have a series of ends-in-view, combined with interpretations of the situations they are in and the means at their disposal. Having these, they engage in a constant navigation of means and ends in the pursuit of their projects. They act in time. This navigation of the world is action; understanding of the relationship between means, ends, and the projects possessed by actors emerges from interpretation. The projects of actors can be fantastical, informed by myth or misguided theory or both, and even when they are not, their pursuit is consistently a matter of conjuring in the present an interpretation of how to bring one future about rather than another. Arendt (2013, pp. 230, 188), in emphasising the process character of action and its interpretation, noted the complex relationship between the rendering of action as it is happening, and the full meaning of action as accessed by the historian:

This account of a minded engagement with the future, constructed by means of anything from pure fantasy to careful means-tested assessment of probabilities, is what we talk about when we talk about action. Action is a process that unfolds in the world, through the reconciliation of our individual and collective environments. But action as a human universal should not be confused with agency. For, while action is always incomplete and subject to a constant competition of mutually implicating representations, agency is different.

To have agency-in-the-world is not just to be a human with a project but rather to produce, through the successful execution of one's actions as one has willed them to be interpreted, a relatively high probability of bringing the project to fulfilment. To have agency is to have capacity to move and shake the world in alignment with the story one tells both about one's own actions and the way they appear in the world, to others. These stories, into alignment with which the world is brought by those that access agency, are both forming of, and formed by, one's projects. Agency, then, is enhanced by engaging and recruiting other humans (as well as non-humans) into a given actor's projects. This means that while all actors have projects and action is a universal feature of social life, agency is variable, and thus requires both analytical thinking and empirical research.

This means that agency is not a metaphysical substance or participant in a duality that mirrors determination and free will. This was the significant error of Giddensian sociology (Giddens, 1979, 1984) – to write onto ‘structure’ and ‘agency’ the Kantian dualism of determined causes and free moral experience, and thus to rewrite the interpretation of Karl Marx and Max Weber as if sociology was not a science of the concrete and the historical after all. In contrast, instead of thinking of agency as a kind of mysterious and unknowable substance without a traceable lineage, let us instead consider what it means to be an agent as Freeland (1985) has characterised it in giving an account of ‘Aristotelian actions’. As Freeland (1985, p. 399) notes, Aristotle (2012, 3.1.1111a, 3–6) describes ‘six “particular circumstances” of actions: (1) agent (tis); (2) action, or “what” (ti); (3) patient (peri ti en tini prattei); (4) instrument (tini); (5) purpose (heneka tinos); and (6) manner (pos)’. To understand what it means to be an agent in Freeland's (and Aristotle's) sense is not exhausted by knowing that this person is acting in such a way at such a time (these are factors 1, 2, and 6). Rather, it is necessary to know other decisive particulars in the circumstances of the action; namely: ‘toward what object in the agent's self-concept’ (peri ti en tini prattei [condition (3)]), ‘with what’ (tini [condition (4)]) the action is done, and for what purpose [condition (5)]).

The ‘structure and agency’ model misses precisely the element of agency wrapped up in conditions (3) through (5). Agency, for Giddens and others following the Kantian schematism, is a cognitive affair, the space of free action and intersubjective cooperation or struggle, while ‘structure’ is the material and cultural (e.g. resources and schemas) basis for this, and the space of causal connections. For Freeland's Aristotle, on the other hand, what might ordinarily be analysed today as ‘structure’ is actually integral to agency itself: the condition, yes, of engaging in practice (2) as a person (1) in a certain way (6), but also bearing in mind a certain understanding of toward whom one's action is directed (3), by means of something (4), and with a purpose (5) that is not exclusively of one's own design.

Thus, although Aristotle does not have the theory of intersubjective relations we pursue here,² he does provide an intellectually useful – indeed we would go so far as to say necessary for contemporary social theory – account of what it means to act as an agent in the fulfilment of a project. For only by challenging the now-dominant sense of agency in social theory, something we believe is done best on the Aristotelian grounds of a power in the Aristotelian sense of dunamis,³ is it possible to understand why a picture in which agency is capacity to act just about has things backwards. Rather, in conversation with Aristotle and Freeland's account of his view of action, we show that agency is action on behalf of a project, wherein a project might be one's own but more often is ultimately under the guidance of a more original author of the project, in which case the agent is not the source of the action, but rather the one who carries out the action for the sake of the project which may originate with another.

To be clear: Aristotle, too, has an understanding of ‘the actor’ as a source of choice, and thus of action as a human universal; a return to Aristotle is not a giving up of the individual, or even of the individual's capacity for extraordinary and heroic resistance. Action and the ability not to act, for instance in the refusal to move oneself when given a direction, crucially rely on ‘choice’ and stand at the very heart of Aristotle's understanding of human being. Indeed, Aristotle provides a functional definitional of the human being precisely as ‘the source of choice’, while choice itself is defined as a ‘desirous thinking and/or a thoughtful desire’ (2012, 6.2.1139b,3-4). So action in this sense is a trans-historical and cross-cultural property of human existence that maintains a degree of independence from power relations; the source of the romance with action which social theory conducts is indeed quite real. But this context independence most certainly does not carry over to agency. For the dunamis not to act as directed does not, if we follow Aristotle, involve a kind of agency in which the human mind is principal and the body is agent.⁴ Instead, in keeping with Aristotle's metaphysics of action, we understand the actor's chosen failure to act in accord with expectations or directions not as an expression of agency on the part of the actor who so chooses, but rather the fusion of desire and thinking in the actor's choice not to act.

Agency, quite differently, consists in the fusion of desire and thinking in the actor's choice to align their knowledge-how and their power-to-make with the project or projects of another. This is to think of agency in relational terms.⁵ Viewed from this perspective, this means that agency is to be found in the social processes of direction, delegation, and organisation, and thus accrues to the one who delegates and not the one who is directed (Adams, 2011). When the actor acts (‘chooses’) to align their action with the project of another, agency accrues to the latter.

Agency is the ability to send someone else as your agent – a ‘company man’ in the colonies (Adams, 1996), a general who will execute a civilian commander's orders (Feaver, 2005), or a lobbyist who represents one's interests in Washington. For a person to have agency, then, that person must be able to make someone else (or, perhaps, something else) an agent-in-the-world for their project. So, to be human is to be able to act with a project, but to exercise agency is to send someone or something to act for you. For this reason, the question of action and agency – of human beings acting within an environment, and, in the name of their projects, attempting to extend themselves across space and time – leads us to the vocabulary for conceptualising power that we will propose here.

Viewed from the perspective of this distinction between action and agency, the study of social organisation can be understood as the study of varying solutions to the problems created by the frailty of human persons, who find themselves bound in time and space, and whose immediate environment, no matter how enhanced by technology, cannot be infinitely aggrandised (Arendt, 2013, p. 188). All actors have projects, and to achieve them, they must do so in space and time; to extend their capacity to pursue projects in space and time, it is necessary for actors to engage people and things, and come up with propitious combinations of the two, so as to achieve ends-in-view and invent new ends altogether.

The utopias of critical theory, liberal political philosophy, and positive economic theory can be understood, in part, as debates about how to accomplish this organisation of persons in a way that is just (or perhaps merely preferable), and which enhances or guarantees something like the good life for those involved in it. In many cases, this assimilation of competing projects to some shared project in the service of the good life in common emanates, in part, from interpretations of the line of thinking that flows from Aristotle through Hegel to Marx, and in particular, finds the resources for critique in Aristotle – as do we here.⁶However, this tradition of thinking about deliberation, identity, and consensus as the route to the organisation of humans into groups, so as to accomplish more together than they could alone, constitutes a social theory of solidarity, and solidarity cannot be the only, or even the central, concept for understanding agency. For, if we make solidarity central, we tend to disavow both hierarchies as a central format of social organisation and the experience of ambivalence vis-a-vis those who direct action as central to humans’ subjective engagement with the world. We must avoid this disavowal, and doing so will involve a painful break from the putatively Aristotelian foundations many of us are familiar with from these traditions.

For, in contrast to the tradition of solidarity, we require an account of the efficacy of hierarchical relations in enhancing the likelihood of achieving one's projects – including the very projects of the good life in common so dear to the communitarian tradition. It is commonly recognised in social theory today that modernity has a ‘dark side’ that manifests in the utility of bureaucratic and technocratic rationality for the fulfilment of projects of genocide, slavery, and labour exploitation – we know about the others of modernity, and their overproduction and overexploitation (see e.g. Alexander, 2013; Go, 2016; Sala-Molins, 2006). What is not often explicitly connected to this argument about modernity is the Aristotelian account of action and agency presented above. Therein, the agency that accrues to those on whose behalf actors act is inextricable from the phenomenon of a joint project; joint projects, too, involve power. That is, one achieves one's projects by giving direction, finding allies to whom parts of the project can be delegated, and thus securing a series of actors as one's agents. If this is the case – and the consistency with which hierarchy attends human action suggests that it is – then we can, in fact, bring these two insights together in the following way.

We can hypothesise that one interpretation of the world that will consistently emerge from human persons pursuing projects is a division of persons in the world into three types: rectors (the persons with the project), actors (those that act on behalf of the rectors and/or on behalf or ‘in the name of’ the project), and others (those who are outside the intersubjectivity of the articulation of a project, and who may be obstacles to it). We may then ask investigators to reconstruct the meaningful variations in this mode of interpreting the world and its enactment, and follow carefully the implications of these variations in meaning for power and position.⁷

This is an interpretation from a rector's perspective, and if rector succeeds in accruing agency, the interpretation will become that of the rector's actors-turned-agents, and perhaps for others as well. But this is a contested matter. For, relations between rectors, actors, and others are not only links in a social chain, itself the site of struggle (though they are certainly that); they are also formed out of representations of such relations (see Reed, 2011 for an account of forming causes). This is the very crux of the issue, for as we shall see, the framework within which the struggle over power relations and their symbolic representation takes place is itself a tremendous site of struggle, one which, in its own small way, critical social theory seeks to engage.

Furthermore, chains of rector–actor–other relationships criss-cross and overlap, leading to a complex web of agency relationships, in which agency flows and, in some cases pools or accrues on, a specific node or set of nodes, and thus accrues to the persons or organisations who occupy said nodes. For agency chains can unfold like an accordion into much more than a dyad or triad. And thus, if agency is variously dispersed among persons in society, this is because those persons differ in the positions they occupy within multiple overlapping chains of rector–actor–other relationships.

We propose, then, that studying the nature of such overlapping chains is how we might best study social power. For example, supply chains for commodity production are long chains of economic power in which buyers like Wal-Mart are increasingly ultimate rectors, setting the very terms upon which factory-owners-as-rectors can negotiate with their workers-as-actors; the precise conditions that enabled this accrual of agency to such buyers constitute the object of study for a history of capitalism. A second example: to study the relationship between police and population is to study the distribution of the use of violence and its legitimation across different rectors and actors. The relationship of police officer to citizen is (sometimes) one of rector to actor, where orders are given (and sometimes followed), though of course, the actor has different projects. But police officers are themselves actors in the name of their chief, most immediately, and, more diffusely, of ‘the public’, a complex and ambivalent signifier that presumably refers to a large aggregation of persons who share a project called ‘the maintenance of law and order in society’. Nonetheless, citizens can have some claim on police as well, and thus can hope to escape the directions of police, or reverse the relationship of officer-rector and citizen-actor (via those notoriously expensive actors-turned-agents we call lawyers). Or, at least some citizens do. In the liberal democracies of the West, there are also persons who enter into a relationship with the police not as actors but as others. This is not a relationship characterised by a hierarchical game of recognition, claims-making, and symbolic struggle, but rather one of violent and radical physical and symbolic exclusion. Persons who have been categorised as outside of the social order, as exceptions requiring exceptional policing, are not interpolated or hailed as citizens, but rather excluded as bodies and morally humiliated as others (Fassin, 2013). For the others of the police, an individual police officer can indeed become a sort of sovereign, possessed not only of the power, but in some complicated way the right, to take life or let live (Agamben, 1993; Foucault, 1990; Reed, 2017).

The insight that power dyads are in fact triads of rectors, actors, and others has been central to sociology at least since Karl Marx theorised the industrial reserve army as other, placing tremendous pressure on the rector–actor relationship of capitalist and worker. But beyond this foundational example, sociology has in fact discovered tremendous variation in how a dyad of rector and actor – or a web of rectors with their respective actors acting on their behalf in various ways – relates to, and is affected by, others. For example, a structural account of heteronormativity (found, for example, in Gayle Rubin [1975], and, modified by the concept of performativity, in Judith Butler [1999]) notes the importance of the abject, queer other to the constitution of male-female relations. Men and women exclude queer persons, and disavow queer desires, and thus constitute the sex/gender order in terms of men and women with desires for their opposites. However, symbolic exclusion is only one type of otherness to rector–actor dyads. The outsider or other can also arrive on the scene as the third person to mediate a dispute, an enforcer regulating a labour contract, or a public whistleblower (Shapiro, 2005; Simmel, 1950).

How, when, and why are these positions transformed? When can a radically excluded other ‘get in the game’ as yet another actor? When do rectors abdicate their powers and responsibilities? When do actors radically change their interpretations of hated others? When do actors abdicate any chance to even be ‘in the game’ – giving up on loyalty and voice, and choosing instead exit – or assault? And how do the representations of who is rector, who is actor, and who is other, condition how people do politics, as these representations circulate through society?

Our contention is that power should be understood as we have here described the accrual of agency. Persons have projects of various kinds, and they act to achieve them, often by recruiting allies as agents. But taking on actors as agents is always complicated, not only because the world is in flux and any agency chain will take place amidst myriad other agency chains, but also because human actors, as humans, have their own projects and meanings (Eigensinn, see Krause 2014, 64–68), and thus only with difficulty and contingency can be made into agents. The tension that surrounds power relations is always in part a result of how rector gets actor to give up previous projects and take up rector's own. Further tension derives from the fact that outside of any set of rectors and actor lies a vast and varied world of others.

This is the picture of the world in which we propose to launch a research project into what Michel Foucault once called criticism ‘as a historical investigation into the events that have led us to constitute ourselves and recognise ourselves as subjects of what we are doing, thinking, saying’ (Foucault, 1984, p. 46). But our tools are somewhat different than his were; they derive, in particular, from the vast efflorescence of social theory within which we ourselves were trained, in which the interventions of Foucault were a significant, but surely not the only, constituent feature of the debate.

Consider the debate on power. Hearn (2012) narrates an arc of ‘American debates’ from the 1950s to the 1970s concerning elites, pluralism, and power, followed by a second arc of what he terms ‘epistemological approaches’, including the figures of Pierre Bourdieu and Michel Foucault, each themselves now founders of discursivity for social theory. The key link between the two arcs is Steven Lukes, who, Hearn writes, turned ‘the problem of power into a problem of knowing’ (2012, p. 80). This identification of Lukes’ Power: A Radical View as a lynchpin accords with the judgment of Clegg (1989), who sees Lukes’ book as the last in a longer arc of theorisations of power beginning with Thomas Hobbes, and ending with Lukes’ Gramscian study of consciousness and real interests. Clegg identifies Foucault, post-structuralism, and science studies as initiating, then, a return to Machiavelli's emphasis on bounded strategy, which he finds more appropriate for the ‘postmodern’ world. Yet Clegg, too, finds much to appreciate in Lukes’ treatment.

The continuing intellectual relevance of Lukes rests, in our view, with the way that he articulated an understanding of the importance of cultural representation and consciousness for the functioning of power in society, yet maintained nonetheless an interest in the basic Hobbesian question of ‘the power of A over B’.⁸ Various locutions of the dyad then became the subject of debate (‘A causing B to do something that B would not otherwise do’; ‘A causing B to do something that is against B's interests’). These debates are well-known: is ‘interest’ a value-laden term? Is this actually a discussion of domination as a subtype of power? But consider instead a much less commented upon part of the debate – the ambivalent and complex uptake and disavowal, in social and political theory inflected by the cultural turn, of the ‘power of A over B’ language, and, in particular, its realist/strategic roots (Selg, 2016).

It might surprise some of our readers to remember, upon returning to Clegg's text, his deep appreciation of the work of the rational-choice theorist James Coleman (Clegg, 1989, pp. 213–216). In a move that imitates Foucault's own close reading of the economist Gary Becker (Foucault, 2008), Clegg returns to Coleman shortly after making the evocative comment that, in the study of power relations, ‘Realism would have been greatly facilitated in the past if some conception of structural power had metaphorically skewed and made uneven and fissured that level table on which conceptual billiard balls from Hobbes to Dahl have moved so easily’ (Clegg, 1989, 209). His appreciation of the rational choice tradition derives from the way in which Coleman's analysis does not lose sight of the relation that obtains between those with more power and those with less (‘the power of A over B’), even if the analyses, in rational choice theory, of the reasons why this relation obtains tend towards the ideological. Coleman remembers that, in modern organisations, the relational organisation of agency is one of the most important modalities of power to study. And so, Clegg's careful merging of (1) his appreciation of the Dahl–Coleman tradition, (2) Foucauldian insights, and (3) his own work in the sociology of organisations, allows him to arrive at a relational concept of agency as part of his project on power. This has deep affinities with our own rector–actor–other language:

Clegg's ‘central paradox of power’ is something in which we are also, in our research group, particularly interested. Clegg (1989, p. 201) explains that ‘the power of an agency [e.g. an organization or individual – in our terms, a rector] is increased in principle by that agency delegating authority; the delegation of authority can only proceed by rules; rules necessarily entail discretion and discretion potentially empowers delegates’. Clegg is both interested in delegation to non-human agents, and in the way in which delegation to other humans creates opportunities for resistance. He thus articulates a relational concept of agency focused on what is, for us, the complex relationship between rector and actor, subject to all sorts of (mis)recognitions, partial understandings, ambivalences, and resentments. We too imagine that allies of a project might include computers, scallops, and bacterium as potential agents. But we would also note that scallops are not known to show an affinity for a politics of racial resentment, which would appear to be rather important for understanding the arc of political modernity.

Where Clegg wrote ‘the delegation of authority can only proceed by rules’, we would write the delegation of authority depends in the last instance upon signification. Indeed, for all of his appreciation of Foucault and Latour, Clegg's own semiotic analysis was rather reductive. Concerned with norms and rules as the primary source of meaning, his image of circuits of power pictures a modern world dominated by efficient organisations, and he subjects, in his theoretical apparatus, social integration to systems integration, with the latter understood as a conduit of technological change. We attempt, herein, something a bit different.

We propose to expand upon the relational account of agency by asking how rectors, actors, and others are represented, and about the struggle over the meanings that condition those very representations. By drawing on Aristotle's account of the situated character of agency – situated, that is, in complex constellations of particulars both known and unknown to those who enact such agency- – we can pursue a hermeneutics of modern political semiosis. This launches us into the question of consciousness, articulated by Lukes and by the two dominant models of power – Bourdieusian and Foucauldian – in the last 40 years of social theory. For if Lukes was the end of one line of thinking that began with Robert Dahl (or Thomas Hobbes), he was the beginning of another – the cultural turn in social science. Here we very briefly discuss the theoretical root metaphors (Pepper, 1942) that organise these new discursive spaces.

Field theory provides a theoretical imagery of social struggles for and over power as divided into different zones of activity, each constituted by a mutually held (and also struggled-over) illusio. Different fields, so constituted, achieve a certain degree of (relative) autonomy from the economy and from each other, and interact through the metafield of power. Thus, power is that which emerges out of the struggles for position by persons who (1) act within certain fields, and (2) try (albeit sometimes unconsciously) to make the field they (variably) dominate more autonomous. And so, the relationship of all fields to each other constitutes the ultimate structure of power in a given society. This is also a structure of authority, because fields are themselves dependent upon the dispositions of the actors within them, which depend upon expectations, inscribed habits, and so on – fields are simultaneously social and symbolic and simultaneously objective and subjective; they thus reveal the distinction between power and authority as artificial. It is fields all the way down (see Breiger, 2000).

The image of the dispositif emerges from the work on power and discipline that fully effloresced in the oeuvre of Michel Foucault, but which also manifested in other theorists of rationalisation, discipline, and the self (especially Max Weber and Norbert Elias); allied to this is the research programme launched by Foucault's concept of governmentality. This approach models power as, in part, anonymous, capillary, and productive, and perhaps increasingly so. The image depends upon an understanding of knowledge and power as themselves inextricably intertwined in a politics of truth. Instead of characterising certain persons or roles as having power over other persons or roles, this work proposed to see persons as subject to various, layered regimes of discipline, transmitted through texts and practices, and ultimately producing self-governance in modern subjects. These regimes of discipline often come together unintentionally, and, in producing different kinds of people, can serve various ends – tactics are autonomous from strategy.

We are, at this point in the development in social theory, well aware of the vital differences between Bourdieusian and Foucauldian approaches. Let us pause for a moment and place ourselves outside both of these magnificent systems of signs. For, when we review the critical literature on how these two images have been mobilised to study power in social life, we see an interesting phenomenon emerge. On the one hand, we have criticisms of the tendency of work done under the sign of Foucault or Bourdieu to anticipate how the interpretations of actors will derive from the positions of actors. Critics note that these are research programmes on culture and power that come to expect moral, aesthetic, and decision-making judgment to hew closely to the ‘structure of society’, though their pictures of that structure are indeed different. To be sure, both Bourdieu and Foucault are correct that, in part, power works via the slanting of interpretation that occurs almost without notice, in a naturalised world; there are great intellectual gains to be made when one can predict with a certain accuracy the emergence of interpretations that self-serve, legitimise rule, justify violence or, in their very practice of interpretation, instantiate certain subjects.

And yet, there is always a risk that the expectation that interpretation will be slanted in a particular way will become a truism – the reality of the model will overcome the model of reality, with the result that our ability to see power in its dispersion and reversals (the explicit Foucauldian goal, after all!) will be impoverished. This is the essence of the Alexandrian critique of Bourdieu and Foucault (Alexander, 1995; Alexander & Smith, 2003). Foucault tried to avoid this problem, but we must ask our reader: did he (and his followers) succeed? Is not the consistent implication of work on governmentality that the meanings assigned to the experience of being human will serve the need for control inherent in the social systems in which such humans find themselves? Hence Neil Brenner's insight – Foucault offered a ‘new functionalism’ (Brenner, 1994). Thus, one problem with these images of power in society concerned their tendency to underestimate the importance and autonomy of the interpretation of meaning for power itself, and in particular the contingencies in the organisation of agency introduced by variations in regimes of interpretation and their performance. Hence we can say that Bourdieusianism and Foucauldianism are ‘not cultural enough’.

And yet. Consider that an equally clear and compelling critique has been advanced that the focus on symbolic violence bequeaths an overall tendency – perhaps derived from the political self-perception of the developed West in the trentes glorieuses – towards a focus on consensus, slanted common sense, and Gramscian hegemony. In this view, what Hearn calls the epistemological approaches to power are too cultural (an accusation, of course, that Bourdieu himself levelled at Foucault, see Bourdieu, 1993, p. 179). This critique holds that these perspectives place a singular cultural engine at the heart of what is a much more varied phenomenon, losing in the process the insights of the sociology of organisations, and those terms from the instrumental accounts of power that dominate comparative-historical sociology – the network, the project, the ambition to rule, the command of the instruments of violence. Focused as they are on reproduction, the epistemologists of power forget that power is useful to have, so as to accomplish specific projects. Power can enhance the instrumentalities of modern life via the reorganisation of networks of influence and alliance, and the use of coercion. Where, in Bourdieu, do we get a serious theorisation of logistics? Where, in Foucauldian sociology, do we find an understanding of the mundane utility of enhanced organisational networks and the legitimation of command? Perhaps these lacunae partially account for the appeal, in the last fifteen years, of the counter-movement against critical social theory, evident in such exclamations as ‘What if explanations resorting automatically to power, society, discourse had outlived their usefulness and deteriorated to the point of now feeding the most gullible sort of critique?’ (Latour, 2004, pp. 229–230).

We have, then, a paradox of sorts: our dominant models of power and culture, or power and knowledge, are somehow both over- and under-focused on representation. Faced with this difficulty, it would appear that Jameson’s (2013) imperative is called for: ‘Always historicize’. And this is precisely what Hearn (2012) does in his recent book. Noting that for large portions of academic practice, this understanding of power as cryptic domination has itself become a kind of ‘truth’ or ‘doxa’, he is attentive to the way in which Lukes, Bourdieu and Foucault have articulated ‘not just theories of power, but ultimately theories of modern power’ (2012, p. 22).

Hearn notes the generational tensions that inspired Lukes, Foucault and Bourdieu, and then goes further in his analysis of this doxa: ‘This phase itself was a cyclical expression of deeper ideational tensions that are constitutive of the modern age’ (2012, p. 22), by which he means a tendency towards suspicion concerning authority as merely a mask for power. To be sure, the hermeneutics of suspicion is an aspect of the modern intellectual project,⁹ and this is clearly a structuring aspect of the thought of Lukes, Foucault, and Bourdieu as well. But we would go even further. For everything that has been gained by these approaches, they remain, in our view, within a circuit of thought that is determined by the French Weberian understanding of modernity, in which the (relatively sudden) deracination of the experience of the sacred admits science, the civil service, and the rationalisation of capitalist exploitation and its legitimation (or discipline, intersecting with, if not reducible to, the needs of factory owners).¹⁰

The tremendous revolution accomplished by the epistemological approaches to power, and the empirical knowledge their terms have enabled researchers to generate, are both undeniable, but we wish to open debate onto new territory, spurred by the problems of our own era, with particular attention to political trends and forces well outside the vision and division of the world provided by the power conversation in its current incarnation. Sometimes one can move backwards to go forwards.

We propose, then, a double turn: First, an Aristotelian response to the contradictory and deeply unsatisfying inner logic of rational choice theory, which sublates the core insights of the instrumentalist traditions of social thought. We seek not to avoid the insights of the late modern incarnations of the tradition of Thrasymachus, but rather to engage and transform them. Second, a reconsideration of ‘power in modernity’ based on a reading of of feminist political theory's encounter with Aristotle and Rousseau.

(a) Principals and agents: James Coleman's Foundations of Social Theory

Economics, rational choice theory in sociology and political science, and legal theory are the home of principal-agent theory, and it is within this intellectual space that we find the claim made most clearly that, rather than a metaphysical principle, ‘agency’ is really a question of whether a person or group has the ability – given the social setting, organisation, and so on – to send an agent and bind that person or group to work on the rector's behalf. By reading symptomatically, we can see how in rational choice theory actors with their own interests are turned into ‘agents’ and the myriad ways in which they attempt to resist being transformed. In this abstract model, rector (‘principal’) tasks actor (‘agent’) with conducting rector's business, representing rector, or working for rector. Often this relationship is one in which actor – removed, perhaps, in space and time from rector, has more information and expertise than rector. Furthermore, actor also has interests that do not entirely align with rector's interests. Hence the core questions of principal-agent theory: how, via monitoring, punishment, or the manipulation of preferences, is rector able to control actor to the point that actor will actively seek to achieve rector's ends?

In rational choice theory, a favourite topic of theorisation is one in which there are multiple rectors (‘principals’). In corruption, for example, a second rector peels actor away from his original rector. There are also ‘others’ in rational choice theory (though they would be unrecognisable to post-structuralist social theorists). These ‘third parties’ to the rector–actor relationship come in many forms. The other may be a regulator, tasked with enforcing a set of rules for how rector and actor negotiate (but, who, then, is monitoring the monitor? Does this other act on behalf of his own rector, and what, really, is that rector's project?). Or, the other may be the injured party who, hurt by actor, seeks recompense from rector. Finally, the other may be the world of the unknown, or the world of the unknown combined with an enemy – who may or may not be rational.

This frame has found a variety of applications in social-scientific research. In political science, voters are imagined to be a collective rector, with low information, inclined to punish or reward politicians as their actors by reelecting them or voting them out (Miller, 2005; Morrow, 1994). In the bellicist theory of state formation in comparative-historical sociology, rulers of state or proto-state organisations are rectors, and their staffs and the elites they coopt are their actors. These rector–actor dyads then confront (1) recalcitrant populations that have resources the rectors need (manpower, money in the form of taxes), and (2) other state organisations ruled by other rectors and other actors, with various outcomes depending upon the mix of initial conditions that obtain in a given case (Downing, 1992). Rational choice and instrumentalist theories of action, then, grasp that agency accrues to rectors when actors act on their behalf. Rectors who pursue projects such as building a state, growing a business, or forcing a graduate school programme up in the rankings need actors to become their agents. If you have (good) agents, you have agency. We would, finally, note how amenable the images of multiple principals for a given agent, as well as extended chains of delegation to agents’ agents, and agents’ agents’ agents’ (and so on …), fit the process of ‘time–space compression’ identified by Harvey (2015) and recently subject to critical examination by Rosa (2017).¹¹

However, because the account of the person in rational choice theory is, by hypothesis, empty (preferences are merely revealed empirically), the understanding of how the dyad rector–actor works is un-substantive (Adams, 2010). Via just-so stories about everything, rational choice loses its grip on the explanatory ‘why’ (the Aristotelian ‘that for the sake of which’ each thing is [to ti ēn einai, later essentia, and thus ‘essence’]) relevant to practices of delegation and motivation that constitute agency in the relational perspective we are pursuing here. Previously in this article, we noted and criticised the way in which the social theory of agency can efface the difference between the context-independence of the metaphysics of action and the historically situated power relations at stake in agency. When rational choice theory interprets conflicts between the ‘interests’ of principals and agents, it ends up telling us little about the why of right. How should we understand the circumstances under which actors give up the right to direct their action? Why do certain observers, at certain times and places, view certain instances of rector–actor relationships as legitimate, and others as illegitimate?

Even Coleman, who carefully examines not only transfers of the right to direct one's own action, but also transfers of the right to transfer that right, openly admits that there is no philosophy of right in rational choice theory – in his social theory, right is as right does empirically. This disavowal of the meaningful basis for right initially appears as providing certain theoretical payoffs; ultimately, it indicates how much Coleman operates from within the constitutive fantasy of the modern age. Coleman presents a clean distinction between ‘authority’ and ‘power’: Authority is a transferred right to direct action, sometimes augmented by the transfer of the right to transfer the right to direct action, and is recognised as such. Meanwhile, power is external to recognition: it arrives in the form of the possession of goods that enable one to make an advantageous exchange for the right to direct another's actions. Rector has power in so far as rector can buy more actors, exchanging something they want or need for something he needs; he gains authority when the social group as a whole recognises as legitimate the transfer, as part of one of these exchanges, of the right to direct actor's action within certain bounds.

However, Coleman's residual category of ‘consensus’ as to which transfers of authority are viewed as legitimate is a monster in hiding. The purely analytic, timeless distinction between power and authority cannot be sustained. Consider: what makes tipping a waitress a more legitimate transaction concerning the right to direct the use of another's body than tipping a prostitute? As MacKinnon (1993) notes, a determination of the legitimacy of one form of servitude – one ‘service profession’ – as opposed to another, is contingent upon a legal interpretation of whether or not that service practice is voluntary. In the context of the United States, MacKinnon (1993, p. 23) writes: ‘Courts have been far less concerned with whether the condition was voluntarily entered and far more with whether the subsequent service was involuntary.’ This makes clear that, leaving aside making a normative judgment for oneself, even an empirical exploration of beliefs about legitimacy requires the investigator to develop an account of the meanings of actions. To correct this problem at the core of Coleman's account of principals and agents, we would require an account of how and why actors more or less voluntarily act in ways they would never actually choose save for the circumstances in which they happen to find themselves.

Classically, such actions are what Aristotle (2012) analyses in Nicomachean Ethics, III, 1 as ‘mixed’ actions, those that ‘no-one would choose’ simply in themselves, but which are undertaken under duress. The sense of legitimacy that Coleman reaches for via his ad hoc category of ‘consensus’ is expressly to be found in Aristotle's account of how we judge such actions to be voluntary, meaning that we hold those who undertake them responsible for them, even though we recognise that they are not strictly speaking chosen, which would normally be the criterion of our judgment (that is why they are ‘mixed’) (2012, 1110a,1–25).

The essential kernel of this account for our purposes is that, in coming to answer that these mixed actions are to be considered voluntary, Aristotle establishes the test for the conditions under which an action ‘no one would choose’ is nevertheless voluntary. This test, in turn, grounds the sort of ‘consensus’ to which Coleman is illicitly appealing in presuming that the status quo implies consent. Aristotle, on the contrary, describes the test we apply in achieving consensus or failing to do so. Namely, he says, we ask whether or not ‘the moving principle is in the one who acts’; if yes, then the action is voluntary, even if not simply chosen. Aristotle goes on to show why this grounds ‘our’ judgment of those who act voluntarily, but under duress, in this way.

Standards of judgment of this sort vary – for this reason, they must be studied empirically, and reconstructed hermeneutically – but what they share is the meaningful social recognition of the other actors in the network of actors and rectors that makes legitimate the turning of the actor into an agent. Such recognition (the ‘sharing of a standard’; Greek: suggnome) is offered by various observers (we could say: rectors, actors and perhaps even for some matters, others) to actors who have, under certain circumstances, made themselves agents and thus allowed rectors to accrue agency, but not given up responsibility for their actions. Note, then, that the question of responsibility for one's actions is fundamentally approached via the distinction between action and agency. We may pity or pardon – or in extreme circumstances even praise – a person who commits a crime, for instance if we determine that they have endured this outcome that is unworthy of choice for the sake of something ‘great or beautiful’, or again, if they have been subjected to ‘more than a human being can bear’. Thus, individual actions and our responsibility for them cannot be assessed without critical analysis of the projects for which persons find themselves acting, and thus of the chains of power and their representation in which persons find themselves embedded.

To pursue this critical analysis, we now take up one of the most profound retellings of the transition to modernity, one positioned precisely against the modernist story of the social contract, either liberal or authoritarian, and one that overcomes the tensions of Marxist-Feminism as well. We are referring to Carole Pateman's classic The Sexual Contract.

(b) Homosociality and inter-rector recognition: Carole Pateman

The Sexual Contract is well known for (1) its critique of Marxist-feminism's analogy of the domestic relationship between husband and wife to the relationship between owner and worker, and (2) its critique of liberal feminism's adoption of contractarian language in its criticisms of marriage inequality. The basis for this double move is Pateman's retelling of the social or civil contract of liberal modernity as dependent upon the sexual contract and thus upon a homosocial contract between men as husbands (Pateman, 1988). In Pateman's model of contract, an agreement does obtain – but it is an agreement between a group of rectors (men) to control a set of actors who are also unknowable and dangerous others (women) – and thus the sexual contract is the basis for the perceived equality, or horizontal relations between, multiple male rectors in modern liberal societies. Pateman thus theorises the agreements that can obtain between rectors in a society of many rectors, the dependence of modern liberal consent on power and control, and the accrual of agency to men rather than women in liberal modernity.

She begins her argument by noting how John Locke's political philosophy emerged as a counterargument to Robert Filmer's argument for the authority of the King as productive father, and Filmer's location of the ultimate justification for authority in the progenitive capacities of men as fathers. For Locke, by contrast, when sons come of age, they are no longer subject to their fathers’ authority, and as such can ‘contract in’ to the civil/political order. This civil/political order distinguishes – fiercely – between ‘public’ and ‘private’, and relegates women to the private sphere as daughters, then wives, and then mothers.

What Pateman sees in Locke's philosophy and in the development of modern liberal democracies over the eighteenth and nineteenth centuries, then, is a liberal order that manages and distributes men's access to women's bodies, on the one hand, and produces a social agreement to control women's bodies and desires, on the other. The sexual contract gives men sex-right to women, and ultimately defines masculinity and femininity in terms of possession of this right (men), on the one hand, and the availability of the body of those without the right (women) to those who possess the right, on the other. The rough and ready equality of liberal society for men is founded on this distribution of sex-right. To make this argument, Pateman retells the origin myth of the social contract; then, she traces the complex ways in which her retelling manifests in the modern history of the liberal West. We retell her retelling in our own language.

The liberal order (economic, social, and political) rejects a framework wherein there is a single rector (the King-father), a series of actors (men-brothers) who act on his behalf or in his name, and a wide set of others that those actors control both in the name of rector, and to meet rector's needs. That is, in the pre-modern order the King has, in effect, access to all women to meet his sexual needs, and thus in this order, rector has the ultimate claim on all others-who-are-not-actors. The liberal order decapitates the King-father, and replaces the order centred around the King with a new order, in which those who were merely actors before become rectors themselves.

How can it be that there are many rectors, without there being social breakdown and endless conflict – given that said rectors will inevitably have a plurality of projects and self-understandings? Pateman shows quite clearly how the threat of both social and sexual chaos was frequently levelled at Lockeans to dispute their understanding of right. The answer to this challenge is bourgeois patriarchy: order is maintained in liberal pluralistic orders because an agreement is made between men upon contracting into the new order to respect the marriage contract. That is, men act out a mutual understanding that each rector (man) in the new social contract will have total sexual access to one actor/other (woman), and will also protect and control the bodies and desires of that woman. In return, each rector will recognise each other rector's freedom to do what he wants in his private life – that is, men will not intervene in other men's relations of domination over their wives.

This founding of the supposedly free, consensual social contract on the domination of the sexual contract is made possible by the blurring of the line between ‘actor’ and ‘other’ as these subject-positions apply to women in society. This blurring is evident in the marriage contract. Women enter into, in some sense freely, a marriage contract by accepting an offer of marriage; in so far as this is like a capitalist labour contract, they are actors – unequal actors, but actors still. However, once they have consented to the marriage contract, women are stripped of this right to consent and make contracts, because the marriage contract, unlike the civil contract, is understood to be between persons who are pictured as naturally unequal. Qua a treaty to manage relations between one man and one woman, the marriage contract articulates the belief that there is an ‘inherent’ cause for the husband, as more rational and more human, to be dominant, and thus to ‘objectify’ his wife. This blurring of woman as actor/other is also evident in the labour relation wherein the housewife works, but not for a wage. Finally, woman is other because, in the constitutive fantasy of the liberal social contract, she is unable to control her own desires, subject to emotion rather than reason, likely to pursue adultery if not constrained, and ultimately possessed of a threatening body connected to nature rather than humanity. In a world in which this depiction of femininity is ubiquitous, order is understood to emanate from an agreement among men to respect the sexual contract (and not sleep with each other's wives).

For Pateman, then, the newly liberated male citizen rector of modernity (1) embraces openly the objectification of certain persons into not fully human actor/others, and (2) draws recognition from other rectors who also are free to dominate ‘their own’ inferiors – wherein ‘their own’ is defined by the version of the family normative in society. If we understand sexual activity as a project into which certain persons recruit other persons as allies and aids, we see then that the liberal project of self-determination by the citizenry is intertwined with the projects of specific men to secure and dominate sexual partners. To think of husband-as-rector with access to wife-as-actor-and-embodied-other is to invite an investigation into the politics of gendered representation contained in the institution of marriage. On the one hand, the wife publicly represents the husband if she takes his name and then enters ‘society’ with it – so (female) actor stands for (male) rector. But it is also the case, as Gilman (1892) makes claustrophobically clear in ‘The Yellow Wallpaper’, that, in an androcentric world, the husband is the ‘public face’ of the family, effectively representing the larger family unit, and acting ‘on behalf’ of the family as himself ‘merely’ its actor.

It must be said that it is a powerful moral trick – and this we take to be Pateman's ultimate point about patriarchy and modernity – to convince the world that one is acting as servant to a larger entity that one rules oneself. Liberal thinkers like to relegate to the traditional past of Kings and Lords the arrogance of making personal desires matters of public concern, but for Pateman that is also what the modern public/private divide does for men who are citizen in public and Kings in their own households (see also Arendt, 2013, p. 28; Patterson, 2008). For every patriarch, a miniature Kingdom of rectorship is offered, providing energy and care for him to make his public efforts, and a sexual partner for his private desires – a truth borne out by the idiom ‘A man's home is his castle.’

Pateman sharply exposes this sleight of hand in the representation of power relations; its obfuscation occurs via the relegation of Kingly power to a putatively irrelevant ‘private’ realm, gendered as feminine, abjected as the site of ‘mere’ sensory and nutritive life, and subalterned to the ‘public’ realm, understood as the true and only site of rational activity and ethical–political life. Her account, however, does not take on the classical source of this modern myth-making, for which is required a feminist engagement with Aristotle's metaphysical biology and moral psychology. From such a study, as seen with the help of Cynthia Freeland and Charlotte Witt, we come to see human agency is the activation (energeia) of a potency (dunamis) in a choice that belongs to the actor's volition but is dependent on the fulfilment of conditions sponsored by its fellows. By appropriating this Aristotelian resource, we hope to find an interpretation of action-for-a-project-in-the-world as a nexus involving rectors and others, as well as actors through whom agency is enacted but in whom it does not accrue. But, as Witt (2003, p. 545) notes, the search for practicable sources for appropriation in Aristotle raises ‘questions concerning method in the history of philosophy’ more widely relevant for Pateman's project in rewriting the contractarian tradition. Pateman, we might say, clearly believes that as feminists we should, as Witt puts it, ‘always regard [canonical] thought with suspicion’. Practising the hermeneutics of suspicion makes a great deal of sense, given all the implications of their ontologies and epistemologies that the classical theorists are obscuring and avoiding. Yet, with Witt and her partial rehabilitation of Aristotle's metaphysical biology in the service of a more promising moral psychology and fuller account of action, we also ask ‘What makes an idea or theory useful or interesting for feminism and hence worthy of appropriation by feminists?’

With both this feminist appropriation of Aristotelian action theory and Witt's critical understanding of what such feminist appropriation is and why it might have value, we now see Pateman's argument about the relationship between liberal political modernity and gendered power as one efflorescence of the instability which, in multiple forms, articulates itself at the heart of the modern. Political modernity is a metaproject that preserves actors as the sources of actions out of which social and political reality is created, while denying the persistence of agency relations between those actors and the rectors through whom and by whom the actors’ capacity to act is in fact generated. As Pateman caustically clarifies, liberal political modernity claims, as one of its ‘promissory notes’ (Wittrock, 2000), to make more room for various individuals’ own projects. Yet, as a metaproject organising relationships between rectors, actors, and others, it takes the form of a diamond, or, really, two triangles, the bottom one inverted and the top one right side up. At its centre – where the two triangles meet – is the (homosocial) relationship of citizen-to-citizen, (metaphorically brother-to-brother). Having decapitated the state, the new liberal state is understood to emerge from the mass of male citizens coming together, each a rector with a voice and a vote, to delegate their various public projects to representatives, who will occupy the location of sovereignty that the King, as ultimate rector, once did. In this way, modern political agency accrues to individual citizens as rectors, who make politicians-as-actors into their agents, recallable from the office of rectorship if they do not behave as intended. This is the top triangle. The bottom triangle is the sexual contract, which constitutes the homosociality of male citizens through the public/private split and the sexual domination of women.

However, this liberal myth of the state has a problem, because what it replaces – the world of the King and father-rulers generally – was a world that conveniently combined, in the constitutive imagination, the order obtained in state-society relations with the order obtained in sexual relations, in a single vast chain of being descending from the King's two bodies to the lowliest unmarried female servant. The solution to this problem in the liberal myth occurs in the format of domination embodied in the marriage contract – that contract which is not a contract – which creates, for each rector-citizen, a private zone for the reproduction of life managed on the terms not of the contractual society of mutually recognised rectors, but rather via the patriarchal family. The two recognisably modern power relations – of multiple male rectors selecting a political leader to act on their behalf, and of multiple male rectors agreeing among themselves that each will rule his own private home – support each other, as feminist histories of domestic labour, violence and coverture laws have shown repeatedly (Cott, 2002).

We are prepared, now, to offer an initial set of working hypotheses about how to understand the modern world – derived from the insights of Clegg and Hearn, and from a critical interpretation (guided by Freeland and Witt) of the texts of Aristotle, Coleman and Pateman – and compare these to the language of social theory as inherited from the triad of Lukes, Bourdieu, and Foucault. Most immediately, it will be clear to the reader that we have in a certain sense adopted (co-opted!) the cultural turn, and specifically, the view of social relations as always already symbolic. We have also adopted Lukes’ focus not only on consciousness, but on the somewhat formally rendered relationship of control/agency: ‘A has power over B when … ’ but made that formal rendering triadic rather than dyadic (indeed, as we shall see, rather than a ‘fable of the Bs’, we would ask Selg [2016] to consider the fable of the Bs and the Cs, differentially dominated by A). And yet, we have also claimed that there is something we can see in the transition to modernity that the spaces of discourse dominated by Lukes, Bourdieu and Foucault left in the shadows. Here we begin to sketch this out, though we hope, in so doing, to send a signal requesting help with the project. We start with a note concerning intellectual history and then move to two empirical locations of study as exemplars of what is revealed by the perspective we are developing.

In his classic study of medieval political theology, Ernst Kantorowicz traced the logic of the ‘King's Two Bodies’ as a conceptual frame for comprehending state power. Derived from an earlier understanding of Christ-like Kingship, the King's Two Bodies flourished in late-medieval England as a juristic rendering of both the relationship of individual Kings to the succession of Kings and the existence and persistence of the state. The King's second body, which was understood to exist in time, but potentially ‘forever’ (it was semipiternal), could justify war and the taxes required to pay for it. So, we can say of the pre-modern world: the King's Two Bodies, variously symbolised, provided a frame for thinking, interpreting, and conducting politics, and thus for giving meaning to agency and power.

There appears to be a wide consensus that, somewhere on the road to modernity, and especially in the logics of justification for the eighteenth-century revolutions in North America, France, and Saint Domingue, the representation of the world of power and politics in terms of the King's Two Bodies was disrupted. And yet … this recognition has not been thematised in our dominant images of power in modernity, but rather minimised.

Both Pierre Bourdieu specifically, and, more generally, accounts of modern societies in terms of differentiation, track the evolution from the ‘King's household’ to the bureaucratic machine as different models of the state instantiated in practice. In this view, the emergence of ‘the people’ as the supposed location of sovereignty is but the veil for what is really the emergence of the anonymous logic of the state and its rules and regulations – the emergence of the tyranny of the written (and printed) word, or ‘rule by nobody’ (Arendt, 2013). Bourdieu's admiration for Kantorowicz is well-known; his understanding of the utility of the metaphor of the house matches quite precisely the tracing of rectors and actors in the medieval order:

But then Bourdieu turns to education and bureaucracy; mimicking, in new language, the Weberian hypothesis of the rationalisation and differentiation of different value spheres, and the puncturing of the ‘sacred canopy’ as the driving force behind the creation of modern state power. ‘The house logic contains contradictions that lead to a superseding of dynastic thinking’ (Bourdieu, 2014, p. 258), Bourdieu suddenly announces. He means that this logic comes into contradiction with something important that is happening in the world:

As such, in Bourdieu, another system appears on the scene to take place of the King's system, and it is a system that is immanent to the world, both an avatar of and subject to a logic of (proto)industrial efficiency whose reality is insurmountable. And so, the triumph of the logic of the bureaucratic state over that of the King's household becomes a kind of foregone conclusion, a historical preface to his own account of field, habitus, and capital.

Bourdieu's supple restating of a Weberian insight masks the shift in registers that constitutes his project: from politics to education. The point is not that Bourdieu is incorrect about the rise of the bureaucratic state. It is that, in the focus on understanding that rise as the rise of a space understood as a field, with the attending logics of struggle and exchange (Bourdieu, 2014, p. 20), something has been missed. For surely, even if we somehow wanted to dismiss elections as but a ruse, a ritual of affirmed power, Bourdieu's own dictum would still apply: ‘A social fiction is not fictitious. Hegel already said that illusion is not illusory’ (Bourdieu, 2014, p. 28). So, how could we investigate the non-illusory illusion of ‘the people’? How can we grasp the stunning and spectacular mythologies of ‘the people’ that constitute modern politics itself? It does not appear that field theory has a lot to offer us with regards to what replaces the King, once his head has been separated from his two bodies and the problematic of law as responsive to ‘the people’ as a strange and multitudinous rector has been introduced (Walzer, 1993).

In the opening pages of Discipline and Punish, Foucault evocatively cites Kantorowicz, calls for a history of the two bodies of the condemned man, and hints that the energies of a previous era of sovereignty are somehow redistributed in the modern. However, the vast majority of the Foucauldian literature has not attended well to this problem, hewing rather closely to the concern with discipline, the production of knowledge and subjects of knowledge, and governmentality – neoliberal or otherwise. Nonetheless, this problem – of what ‘remains’ after the king – has been subject to one extended interpretation in terms of biopower and discipline. If we trace a line from Foucault (Foucault, 1990, pp. 135–159, 1991, pp. 3–31, 2003, pp. 239–263), to Agamben (Agamben, 1998, pp. 8–12, 15–29, 91–103, 119–153), and finally to Eric Santner, we find an intricate reading of ‘the people and the people’ in modernity, which is understood to consist of (1) the invention of the disciplined modern subject with a docile body and (2) the emergence of the population, as an object of inquiry, manipulation, and control, for its fertility, productivity, health, and so on (Santner, 2011). Santner wishes to understand the order of biopower that follows the irruptive event of the French Revolution as one in which the ‘missing flesh’ of the King's sacred body is the repeated, traumatic point of fantasy for a modern era that cannot come to terms with itself.

Yet here too, something is missed. Despite evocative readings of the terror and fantasy that accompanied the shift to post-Kingship society, Santner takes the diffuseness of biopower in the modern era for granted. He thus accepts that the body of the unnamed citizen or prisoner, disciplined by soul/psyche, is the location for relatively anonymous discipline. Nowhere is there an analysis of who is pictured, in the commonsense world of popular culture, as being ‘in charge’. Who manages to represent the body politic in the double sense of elected representative and figuration of the people? This is an issue endlessly debated in modern liberal democracies (Nelson, 2014). Of course, there have been many ventures, under the signs of Foucault and Agamben, to comprehend modern bodies as variously gendered, raced, and classed. But it is notable, in this regard, how radically silent Foucault's texts are on the fundamental shift in politics qua politics constituted by the French Revolution, something which is absolutely central to the feminist histories of the birth of modern France.

If we open ourselves to the language of rector, actor, and other, what becomes available? We would have, of course, a natural way to follow the far-flung commodity chains of global capitalism; the high-speed financial transfers reliant on algorithms as ‘agents’; the migrations of persons across borders dependent upon networks variously constituted; the stunning reordering of the environment and emotion created by climate change and the nuclear threat; in sum, a route to grasping those hybrids which are the basis of any claim that we have never been modern, but also have themselves intensified in the so-called modern age (Latour, 2002). But we would also be able to comprehend what really happened to the King's Two Bodies – not its abstract negation but its transformation. The ‘modern order’, in other words, like the feudal order (supposedly) before it, relies on a political semiosis experienced as sacred and transcendent, containing the constitutive rules of the game and an overarching set of representations, which can be used and re-used to represent the world and the struggles within it. The fantastical myths of political modernity also have something to say about who is rector, who is actor, and who is other, and these myths help arrange how power functions in society.

We begin with the French Revolution, not because it is necessarily the centre of the modern but because, perhaps even since Kant's memorable assessment (Feher, 1990, pp. 201–218), it has served as a ‘model system’ for the sociohistorical comprehension of radical social change, and indeed for the transition to modernity (Guggenheim & Krause, 2012). Is it possible to approach an interpretation of that event that comprehends it, as Durkheim intended to do, as a transformation of the sacred rather than its deracination? How should we understand its politics? Given the regicide – the literal and metaphorical destruction King's and the Queen's Two Bodies – it would appear to be a promising route of analysis. Indeed, we have built our own understanding of rector, actor, and other in the French Revolution in part via a reading of Joan Landes’ classic Women and the public sphere in the age of the French Revolution (1996). Whereas both Foucault and Bourdieu, despite all of their differences, imagine the modern as the inauguration of a diffuse, abstract bureaucratic anonymity, Landes instead is quite precisely concerned with ‘who is taken to be in charge’, and how access to decision-making is distributed (for instance, who is understood to ‘hold power’) before, during, and after the revolution. She is deeply interested in who is taken to be rector, who is taken to be actor, and who is othered. It is on these terms that her interpretation of both modern society and modern philosophy and is made.

Landes shows that the interpretation of power as ‘influence’ by those engaged in the struggles for power in France at the close of the eighteenth century is a key point of analysis for understanding the onset of political modernity. She carefully reconstructs a history of the salon as a route to power in a world where the King's household or court was the public space of power and the King was the only truly public person; in such a world, to accrue agency and bring one's projects to fruition, one had to have influence. And influence could be accessed in the salons. Run by aristocratic women, often divided into tiers wherein higher floors were more exclusive, the salons were spaces of appearance and performance. To move and shake a world that required the assent of the royal body – an assent often achieved by influencing the ministers who would both influence and then act on behalf of the King – one needed, in the salons, to match wits with the high and mighty, carry off a turn of phrase or subtle insult, or even stage a play. Only by such a route could one ‘be pragmatic’ and accomplish projects – for example, drain swamps, seek improvements in farming, or achieve changes in law. Influence through charm, grace, and superior aesthetic cultivation (often baroque, but not always, for the salons were also spaces of experimentation and indeed enlightenment) was the route to power; to accrue agency in the world of the King's two bodies (‘L’état, c’est moi’) required a certain kind of cultivated persuasion.

Neither Robespierre nor Rousseau could stand it for one minute. The sociality, the performance, the required keeping up with fashion – all were, to them, the intolerable marks of a horrifyingly corrupt society that had to be overturned by force. And so, Landes shows, the resentment of the ‘powerlessness’ of ‘the people’ in the ancien régime was inextricably bound up, in the revolutionary imagination, with the perceived link between aristocracy, ‘influence’, and femininity. The revolutionaries, then, created for themselves (and via delegation to Jean Louis David), an aesthetic of masculine warriors bound in homosociality to the defense of each other and the republic. The liberal order was born as a band of Roman brothers, liberated from the irritating (terrifying?) influence of women, an influence which they perceived as having grown pathologically in a world in which there was, symbolically, only one true male body.

Thus, in Landes, the birth of the modern is about the spectacular representation of the world of many rectors and how it will work (and not, then, anonymity or bureaucracy). She shows, via her history of the Society of Revolutionary Republican Women (Landes, 1996, pp. 139–151), what Pateman articulated for political philosophy. In particular, the revolution and its aftermath – the emergence, over the tortuous nineteenth century, of a bourgeois republic in France – is fundamentally premised on the reconstitution of rule, via the imagination of politics as constituted by glorious achievements of a band of brothers. In this imaginary of rectors, actors, and others, a sharp line is drawn between the sphere of the domestic and the sphere of public politics, with women in the former as actors (subject to their rector husbands), and men in the latter, engaged with each other as co-rectors in the precarious project of democracy. In their self-conception and spectacular representation in novels and films, the new masters of the modern world specifically proposed to replace corrupt influence (coded as feminine, emotional, aesthetic) with the public conduct of power via deliberation (coded as masculine, reasoned, ascetic). And of course, they took as their others not only the dangerous bodies of women, but the enslavable and colonisable bodies of Africans, as recent historians of French philosophy and of the revolution have shown in considerable detail. As Mbembe writes:

Thus, in our interpretation of Landes, the shift to the modern is a shift in the representation of chains of power. The ‘social question’ is a question of whether white men without sufficient property will be allowed into the game of delegating representatives to ‘hold power’. (Somers, 1992). The question of technology, nature, and the modern is the question of how and when non-humans can become parts of these chains of rectors and actors, enabling their increasingly far-flung dispersion and their compression of time and space. But at its core, the modern reconfigures the accrual of agency to rectors by refiguring the representation of the chains that grant agency in the first place. Viewed from the perspective on power presented here, in other words, the transition to modernity takes place via the transition from the King's Two Bodies to the People's Two Bodies – that sacred semiosis that draws together the universalism of the Declaration of the Rights of Man and Citizen with the ‘dark side of the light’.

Recall that in Pateman's theory of modern patriarchy, the representation of woman as uncontrollable other contributed to the rendering of relations between men and women via the marriage contract, which then underwrites the social contract as understood in liberal modernity. Similarly, according to Charles W. Mills, a racial contract underwrites the white experience of modernity. But what is this game of liberal recognition? To what are the critical theorists of the modern (sexual, racial) contract referring?

We posit that liberal political modernity is a particular instantiation, in practice, of the mythology of ‘the people’. This practice took elections and thus ‘politicking’ as one of its primary sites for working out this mythology of sacred and profane, civil and anti-civil (Alexander, 2006). In particular, in the process of electing representatives, liberal political modernity instantiated the practice of the executive's second body, coded as emerging from the ‘people’ as sovereign, where ‘the people’ was understood as myriad rectors voting to delegate to the modern democratic executive the role of esteemed actor who is the agent of the people. Thus, the paradoxes of liberal modernity consist not only in the way it is underwritten by the sexual and the racial contract, but also in the ambiguity of delegation to a powerful executive that takes place within the space of those who, via the ‘social question’ accede to the status of citizen rectors.

The ambiguities that attend this aspect of the modern production of agency relations were much commented upon, from a legal perspective, by the political theologian Carl Schmitt. But Schmitt's distaste for democratic delegation led his insights to be mixed together with a misreading of liberal modernity common on the German right in his generation (see Bourdieu, 2005) – that liberal modernity had emptied the political world of meaning and authenticity. Quite to the contrary, liberal modernity meaningfully invests the energy of actual persons, and their labour, work, and action into cultural creation of the executive's second body. This is quite evident in the ‘civil religion’ of US politics, wherein presidents and presidential candidates have the potential to grow a second, iconic, presidential body.

In his studies of Barack Obama's rise to power, Jeffrey C. Alexander has posited precisely this mythical relation to presidential candidates, who emerge onto the cultural scene via heroic narratives (often managed by campaign staff and public relations experts). Alexander reconstructs Obama's performance in political campaigns, and the media events that propelled him to victory, and then notes that

Alexander is able to trace the mythological projects of the campaigns of Obama and McCain, as competing attempts to embody ‘the people’ and thus sway the (relatively small number) of undecided votes in the 2008 Presidential election in the United States.

In the 1980s, American political scientist Michael Rogin argued that the phenomenon of the executive's second body in the USA precedes the televised age, and is not limited to campaigns but inheres in (some of the) public actions of installed presidents as well – especially those understood as ‘exceptional’, as somehow both inside and outside the law. Specifically, Rogin finds the executive's two bodies functioning in the presidential terms of Abraham Lincoln, Woodrow Wilson, and Richard Nixon. Each of them navigated crisis and war via a political semiosis whereby ‘confusions between person, power, office and state become accessible’. These executives, Rogin argued, ‘found problematic their bodies mortal and the human families and dwelling places that housed them … they sought transcendent authority and immortal identity in the White house, absorbing the body politic into themselves’. Thus, Lincoln absorbed and reformed the body politic via the Civil War, the Emancipation proclamation as a war measure, his public life and speeches, and his eventual martyrdom at the hands of a Southerner. Wilson articulated the American empire as a moral project that, like a ghostly second body, was built into and animated the American industrial machine. Nixon insisted on the mystical fusion of ‘his own person with the presidential office’ (Rogin, 1988, pp. 82, 99, 109).

Of course, Nixon was impeached, the American congress passed an amendment banning slavery, and American empire would take a form, in the mid-twentieth century, foreign to Wilson's imagination. These facts point to the complex difference between the King's Two Bodies as a way of arranging rector–actor relations and ‘the people's two bodies’ as a way of doing so – and thus to a series of difficult research questions about power, agency, and modernity. Kantorowicz's study started with jurisprudence; a study of the political semiosis of executive power would have to attend just as much to mediated popularity.

We would be remiss if we did not acknowledge that this research project has emerged, in part, as members of our group in Berlin, Charlottesville, and Hong Kong attempted to come to terms with recent events in global politics, particularly since the financial crisis of 2008. Social theory, after all, is ever struggling to link the long arc with the difficulty of the present moment. One aspect of the present that concerns us is the crisis of liberal democracy in the West (though this is certainly not our only concern; the next paper-in-progress to emerge from this project concerns the history of the Cultural Revolution in China).

Via the peculiar political magic of liberal democratic modernity, a close presidential election becomes ‘the mandate of the people’, and suddenly a whole set of actual people within the nation must – to varying degrees, for a variety of reasons and coercions – follow a leader they detest. But we can see from the theory of power set out here that Schmitt and those on the left who have used his work to examine this recurrent difficulty have insufficiently examined the complexities introduced by the non-identitarian relationship between social rector–actor–other networks and their symbolisation.¹² Democracies function not only via the mechanism of election and thus by the struggle to install an executive, but also via the representation of this struggle. Political representation in a republic itself must be culturally represented and (re)interpreted. From a semiotic perspective, the mythology of ‘the people’, and the struggle over the meaning of the people that takes place in the media, is precisely what fills the gap of Schmitt's inescapable concept of sovereignty.

For, in a representative republic, a democratic vote involves multiple citizen-rectors delegating various projects to a single actor, or small set of actors. This means, de facto, that those whose preferred candidate loses an election do not have their preferred agent-in-the-world. For those rectors, their actor (their candidate) is not empowered to run the state. The winner of the election, on the other hand, is legitimated as sovereign, in part, via the mythological interpretation of the election as the ‘will of the ruled’ (Szelenyi, 2016; Weber, 2005). Part of this legitimation, especially in presidential politics in the USA, is then constituted by the use of the President's first body – his mortal body and its iconic representation – as a signifier for his second, sacred body as a figuration of the people. And it is from this perspective that we can offer an interpretation of the current political crisis in the USA.

The United States – and the Federal government in the USA, in particular – is in the throes of a crisis of republican rectorship. This has certain dimensions that are clearly matters of political semiosis and performance. It has been shown that Barack Obama's election involved him becoming a charismatic leader to some (as most presidential candidates do), and thus that he was able in the process of seeking election, to grow a second body, and thus make a claim to represent to the body politic. Simultaneously, his elevation to the executive led to a radical increase in white supremacist activity in the USA, and surveys showed that a portion of the white electorate refused to recognise him as a legitimate executive at all, as well as imagining him as a Muslim (and sometimes Communist) subversive who had grasped the levers of the state – thus drawing upon the mythological others of American culture. In a sense, Obama's first black body rendered the growing of a second body extraordinarily difficult and contentious – even as he took on the accoutrements of office. Indeed, these ritualised aspects of his power position have been shown to be less culturally helpful to Obama's political power than they had been for previous presidents (Dyson, 2016).

And so, over the course of Obama's time in office, a racist and anti-republican regime of signification, according to which Obama could not be American and could not grow a second body, grew in prominence, especially on social media. Drawing from accounts already used by Obama's official conservative opposition during his candidacy and his presidency, this language then effloresced in the presidential campaign of Donald J. Trump, who rose to prominence contesting the legitimacy of Obama's place of birth (a very ‘Kingly’ concern, we would note).

The racial dimension of Trump's campaign has been widely commented upon, as has the misogyny of his attacks on Hillary Clinton. But the crisis in the USA extends beyond this, because after Trump's election, it is not at all clear that the very game according to which the people delegate their energy to the second body of the president will hold. Both during the campaign for president and while in office, Trump has acted well outside the basic game of representation and mediation in American politics. He has – to a certain degree – abandoned the game of ‘the people’ – according to which the president is an actor – for something more King-like, more buffoonish, and more pernicious. In this alternative semiosis, Trump is a rector, not an actor-as-agent-for-citizen-rectors, and as such is adored by some actors, hated by others, but ultimately possessed of sovereignty of, from, and by himself.

This alternative semiosis of power is not inevitably ascendant; Trump still uses the language of the people. He does so, however, in accompaniment to actions and interpretations that are well outside of the bounds of the positive side of the binaries of civil power as analysed by Jefferey Alexander (2006). It might be suggested, then, that he is no longer playing the game of becoming the figuration of the people. In this sense, the foreboding aspects of the Trump presidency disclose an underlying tension in liberal modernity itself.

In the modern, democratic-republican version of the people's two bodies, the approximately one-quarter of the American population that voted for Donald J. Trump were engaged in an electoral game of delegation of a powerful office to a person who they believed would help them bring their projects to fruition, via various chains of power, and over and against the conflicting projects of others, who were also in, but lost or abstained from, the electoral game. This is the basis for the legitimacy of Trump's rectorship, or so one is inclined to think, when thinking inside the people's two bodies. However, another possibility looms. The electoral game may not, in fact, be the organisational or cultural basis for Trump's ongoing power. Indeed, it is possible that the electoral game will soon be seen as a threat to that power. Such a possibility is indicative of the manner in which a Presidential politics, in particular, emanates from ongoing legacies of a royal social imaginary, and strives for the emotional appeal and satisfactions of the strong leader taken as naturally sovereign. This spectral seduction has always plagued even the most successful modern democracies in which a plurality of actors constitute themselves as rectors with the right and the duty to adjudicate the legitimacy of rectorship.

We have found that, when one adopts the distinction between action and agency that we have proposed here, and its accompanying language of rector, actor, and other, certain problems come into clearer focus. Each can be imagined as an area of future research, beyond those (marriage, revolution, presidential politics) already proposed here.

When the meaningful meta-framework for understanding who is rector, who is actor, and who is other is fundamentally contested, the persuasive performance of rectorship is enhanced as a route to power. To walk and talk the myth becomes a way to access command. Thus, the study of power – and even, of government or governmentality – will be, in part, the study of performance and social drama and its contested interpretation as routes to hierarchical command. Research projects on power and violence in the Cultural Revolution; the precarious politics of state-building in the early American republic; and how to do things with images in a digital age will mobilise the dramaturgic metaphor, and its accompanying philosophical interest, rhetoric, to study political theatre and its relationship to political theology, past, present and future.

Nowhere has the relationship between agency relations and othering – between the (mis)recognition that obtains between rector and actor, and the violence of the exclusion of other – been taken up more clearly in philosophy than in the encounter between postcolonial thought and classical political philosophy, especially Hegel (Bernasconi, 2000; Monahan, 2017; Taylor, 2007). Responding to this encounter, we propose a rereading of dialectics that moves out from Hegel in two directions at once – back to Aristotle (through Arendt's problematic yet promising appeal to ‘the Greeks’), and forward through Fanon and Butler to a critical-and-empirical interrogation of universality, modernity. This could be combined with a comparative-historical sociology of the Atlantic revolutions, and especially the revolution that featured the ‘avengers of the New World’ (Dubois, 2004).

The dynamic tensions created by interlocking and countervailing networks of rectors, actors, and others manifest in the competitive politics of contemporary liberal democracies via contests over what Benhabib (2004) has called ‘The Rights of Others’. At present, for instance, the ‘letter of the law’ would seem to imply that full equal rights must be granted to millions of persons who have found themselves forced to leave regions torn apart by violence; yet, democratic majorities within the nations where these ‘others’ (are they refugees? asylum seekers? undocumented migrants? illegal aliens?) find themselves will not, it seems, tolerate this. Thus, we must ask what is the normative basis for the legitimate expectation of hospitality under these conditions (Weinman, 2017)?

We conclude with a note derived from a reading we have discussed multiple times in the attempt to come to terms with agency, power, and modernity as concepts. In her reading of Hannah Arendt's account of Adolf Eichmann, Judith Butler notes an instability. Arendt famously described Eichmann as acting thoughtlessly – a description that fuses with her characterisation of bureaucracy as ‘rule by nobody’. Yet even here, at the very heart of the most inhumane of bureaucratic beasts, to make this interpretation Arendt needs it to be the case that Eichmann is both responsible individually for what he did and also that he did what he did on the basis of a lack of personality, rather than some evil intent, genius or madness. Eichmann, as individual actor with agency, is a paradox; he makes no sense. But, as Butler (2014, pp. 172–174) shows, by Arendt's own understanding of human plurality and the phenomenological fact (rather than a moral imperative) of cohabitation, thinking, in fact, only happens in plural. (Likewise, we would add, agency only happens in plural.) Furthermore, it is only possible for us, with Butler and Arendt, to judge Eichmann because Eichmann himself did make judgments of the National Socialist project to which he aligned himself, and of his superiors in the Nazi party (first and foremost Himmler) to whom he voluntarily (in Aristotle's sense) transferred his right to withhold his choice to act as directed, and on whose behalf he constituted himself as a subject who acts, but lacks agency, in the constant re-narration of his life as a member of the German Volk.¹³ Eichmann acted in a very particular, and very specifically constructed and represented, chain of rectors, actors and others.

Our project is to press – through research, theorisation, and conceptual development – on the ambiguity that Arendt could not ‘solve’ philosophically in her encounter with Eichmann, and thus launch a series of sociohistorical and philosophical research projects for the critical analysis of power, authority, and violence. For, even in this remarkably stark and horrible instance of inhumanity, we see the all-too-human ambiguities of power as a question of who is rector, who is actor, and who is other. The rational choice theorists and economists love to talk about ‘agency problems’ – by which they usually mean such banalities of exploitation as securing worker compliance with higher assembly line speeds, or ideologically justifying the exhorbitant pay of CEOs in shareholder capitalism. We wish, instead, to examine precisely those ‘agency problems’ that cannot be solved – they are, rather, the ambiguities and uncertainties of social life itself under conditions of plurality. A critical theory of power, authority and violence will not provide solutions that enhance the control of some humans over others, but rather show how such agency problems open onto that very human activity called politics.

The authors thank Jennifer Bair, Irit Dekel, Jeffrey C. Alexander, Risto Heiskala, Daniel Hirschman, Emily Erikson and Matthieu Desan for comments and suggestions.

No potential conflict of interest was reported by the authors.