The Responsibilised ‘Agent’ and Other Statuses

Introduction

Neoliberal discourses that present individuals as empowered and autonomous choice-makers frame status-based inequalities and institutionally supported oppression as personal ‘in a way that turns the private as political on its head’ (Gill, 2007: 155). The term ‘neoliberal subjectivity’ is gaining currency both within and outside sociology to describe the internalisation of this phenomenon – the attribution of responsibility to one’s self that excuses institutional factors (see, for example, Sweet, 2018). Evidence of the degree to which young Europeans display neoliberal subjectivity is revealed in a study by Karin Schwiter (2013) in which young adults in Switzerland were asked to explain their life plan. Schwiter (2013: 154) found that, regardless of gender or class, participants demonstrated a neoliberal subjectivity that ‘absolves the state, municipalities and employers of responsibility, handing it down to the individual’. Unequal burdens associated with interdependent statuses, such as the unequal division of labour within the home are accounted for as the voluntarily assumed life-choices of unique individuals (Baker, 2008; Schwiter, 2013).

Brannen and Nilsen (2005) suggest that a kind of neoliberal subjectivity has also become internalised within sociological research and shows itself in the manner in which data are analysed. Coffey and Farrugia (2014: 472) warn too of the use of conceptual frameworks that align agency with resistance to construct people who do not ‘resist’ as passive and ‘erases the efforts’ of the people concerned. At the level of the individual, some people become celebrated as ‘agentic’ while others are dismissed as ‘lacking active subjectivity’ for their failure to share the political orientation of the researcher (Coffey and Farrugia, 2014: 472). Choices that reproduce social patterns are held to reflect both stronger constraints and a ‘lack’ of agency while deviation from understood ‘standard’ biographies are interpreted as indicating a more authentic interiority or a greater personal will. At a group level, institutional effects are analysed as group-specific ‘trends’ that emerge from the free choice of autonomous individuals, as in descriptions of the ‘delayed transition’ to adulthood (Arnett et al., 2011). This leaves unexamined the social and economic forces contributing to the conditions with which a cohort is required to cope (Côté and Bynner, 2008; Young et al., 2011).

While debates on the relative importance of giving attention to agency and structure play out in disciplinary divisions (Brannen and Nilsen, 2005; Coffey and Farrugia, 2014) and in the development of middle ground positions (Evans, 2007; Evans and Biasin, 2017), some sociologists have called for a conceptualisation of agency that will allow the researcher to ‘get beyond’ neoliberal conceptual frameworks (Coffey and Farrugia, 2014; Harris and Dobson, 2015) and the ‘dualistic myths’ (Barnes, 2000: 99) that support them. This article responds to those calls. It examines the underlying connection between dominant libertarian conceptualisations of agency and a neoliberal construction of individuals as empowered autonomous choice-makers, solely responsible for their lives. It claims that Barry Barnes’ (2000) model of agency offers the means to move ‘beyond’ such individualistic accounts by providing a framework that is fundamentally social and monistic. The implications of Barnes’ model for understanding the individualisation process as a shift in the local expectations of the status of ‘individual’ are considered, and I suggest that Barnes’ model allows for investigation into the unequal burdens imposed by a requirement to construct the self as autonomous. Barnes’ conceptualisation opens up space to explore how the interplay of status expectations operates to enable or disable social membership. I indicate how Barnes’ model brings a new sensibility to the operations of voluntaristic discourses by indicating the methodological implications of Barnes’ model in a range of areas.

Neoliberal Discourses and the Libertarian Models That Support Them

Neoliberal discourses are grounded in libertarian conceptions of the individual as having a power or capacity to act in a manner that is not causally determined. The attractions of the libertarian position are that it accords with our intuitive notion of choice as autonomous and undetermined, and that it appears to offer the possibility of social equality. Where agency is conceptualised in libertarian terms as the potential power to intervene in the world through the undetermined action of an individual, individuals are conceived as a priori empowered and therefore potentially equal. Equality is conceived as limited only by external barriers to freedom of choice, so that the removal of constraints to choice promises a level of social equality.

Dominant libertarian models result in neoliberal narratives because they commonly fail to adequately distinguish the a priori attribution of agency (conceived as the ability to act otherwise based on alternative actions being possible) from the contingent attribution of agency (voluntary action) that underlies attributions of responsibility. I argue in this section that overlooking this distinction leads to neoliberal understandings of individuals as empowered autonomous actors responsible for their lives.

Libertarian social theories locate agency in the possibility of the undetermined individual action. Action may be conceived in some libertarian models as direct action, but often is understood as the undetermined ‘choice’ that then leads to action. This is frequently formulated in terms of the possibility of an individual being able to act ‘otherwise’, as in Anthony Giddens’ (1993: 75) famous formulation: ‘it is analytical to the concept of agency . . . that a person “could have acted otherwise”’. As Nigel Pleasants (1999: 117) points out, this is necessarily correct if analytic. However, the statement then operates as a definition. This definition conforms with the everyday understanding of voluntary action on which we base the attribution of social responsibility – and what follows from the definition is that a person is an ‘agent’ only if they could have acted otherwise. If they could not have acted otherwise, there is no possibility of agency, but being able to act otherwise does not guarantee agency. This leaves room for the empirical question of whether, in a particular case, the person ‘could have acted otherwise’: that is, whether they ‘had a choice’.

We generally judge a person to be responsible if they meet three conditions: the causal condition (their action or lack of action bears some causal connection to the event); the capacity condition (they have the rational capacity to understand the effects of their actions); and the choice condition (the action is voluntary). Confusion arises when the definition of agency understood as the contingent possibility of acting otherwise is vitiated and ‘agent’ becomes a synonym for ‘actor’. Consider, for example, Hays (1994: 63): ‘the central point that is implied in all definitions of agency: alternative courses of action are available, and the agent therefore could have acted otherwise’. In this statement, agency is not tied to notions of voluntary action but is linked only to the availability of alternative possible actions. We can picture this libertarian ‘agent’ as an autonomous vehicle at an intersection. The vehicle can proceed down the one path or the other. Or it can do a U-turn or reverse up the street or drive into a tree. Yet the fact that the vehicle could go one way or another does not mean that which way it goes is not causally determined. Either the direction of travel is causally determined (by, for example, pre-programming) or it is a frighteningly random action. In either case, we would not consider the autonomous vehicle to have acted voluntarily. The argument that a person’s action is not causally determined since alternative actions are possible is only convincing because the ability of the human ‘agent’ to act ‘freely’ (and yet not randomly) based on internal capacities of some kind is already assumed.

When agency is conceptualised as an internal capacity grounded on the possibility of alternative actions, individuals whose actions ‘could have been otherwise’ become understood as individuals whose actions are voluntary. Since all action could have been otherwise, the ‘voluntary’ condition under which we attribute responsibility falls away and leaves only the causal and the capacity conditions. All actions performed can, on this model, be deemed voluntary since the individual could always have acted otherwise, and the individual is consequently accounted responsible in all situations in which they can potentially act based on rational capacity alone.

Rule-Individualism

Within the framework of libertarian models of agency, where equality is conceived as limited by constraints to alternative actions, removing exterior constraints is understood to provide the conditions under which individuals can be equal. Nonetheless, individuals faced with no external barriers frequently abide by patterns of social behaviour that are not in their individual interest. A number of social theorists have attempted to reconcile the resilience of social order with libertarian understandings of agency through ‘development’ theories: theories that explain social order in terms of the development of the capacity for reflection along with the internalisation of social rules during childhood (Westlund, 2009). Although free to take alternative paths, individuals are, by virtue of their knowledge of or orientation towards social rules, subject to some form of ‘pre-programming’ that influences but does not determine their actions. Actions can still be accounted free since alternative actions are available. This is compatibilism of a kind, but it is one in which dualism is re-created and internalised: some actions can be considered truly agentic and others as routine or conforming.

Wittgenstein’s private-language argument has been utilised in a number of critiques to demonstrate that development theories rely on a rule-individualism that reintroduces dualism and is ultimately unsustainable (see, for example, King (2009) on Bourdieu, Habermas, Giddens and Bhaskar; Pleasants (1999) writing on the theories of Habermas, Giddens and Bhaskar; Schatzki (1997) on Bourdieu and Giddens). Wittgenstein’s private-language argument has a number of strands and interpretations, but central to them is Wittgenstein’s (1968: §258) insistence that if a rule is to be followed in private, ‘whatever is going to seem right to me is right. And that only means that here we can’t talk about “right”.’

Barnes’ (2000) critique of the social theories of Habermas, Giddens and Bhaskar also draws on Wittgenstein’s private-language argument, and on Saul Kripke’s (1980, 1982) exposition of Wittgenstein. Barnes argues that social practices are not reducible to the practices of individuals, all habituated to or attempting to individually follow the same rules, since individuals privately engaging in rule-following could not produce a coherent practice in a collective. Just the opposite, he argues, is true: individuals must alter their own routine practices continually in order to maintain a collective practice. Offering the examples of marching soldiers or members of an orchestra, he describes social practices as constructed by interacting social individuals who must constantly modify their behaviour – not to conform to internal knowledge of a rule but to adjust to the other participants (Barnes, 2000: 56). The mutual coordination of social agents does not rest on a presumed tacit understanding or private knowledge of social rules, but instead the standardisation of practices and actions must ‘arise out of interactions’ and rests on mutual intelligibility (Barnes, 2000: 67). Intelligibility does not require following a rule, nor is it violated by resistance. Both resistance and compliance, and indeed sanctions for non-compliance, operate in relation to a rule – and acting in relation to a rule is a social practice: it must be judged socially not understood and applied privately. It is only in a case where a person’s actions are unintelligible that the person’s capacity might be called into question. In such a case, the group may judge the individual not accountable by virtue of a lack of ‘competence’.

Barnes (2000: 55) recognises that the fallacy of rule-individualism applies to any theory which describes ‘macro-order as constituted by individual persons who separately orient themselves to the same rules’. The shift in conceptualisation that Barnes proposes is from one where macro-order is understood to be sustained through the orientation of the individual towards order, to one where macro-order arises when individuals orient their actions towards one another.

Barnes: Agency as a Status

Wittgenstein’s influence can also be seen in Barnes’ approach to the dualism that accompanies rule-individualism. Wittgenstein considered most seemingly intractable metaphysical debates to stem from errors that arise when similarities in grammatical structure are misinterpreted as similarities in language function – when words lose touch with their social function in discourse and ‘language goes on holiday’ (Wittgenstein, 1968: §38). Wittgenstein insists that, in the face of metaphysical dilemmas, we should re-examine how language is ordinarily used. Taking this approach, Barnes (2000: x) begins his discussion by giving attention to the everyday language of ‘agency’ and ‘choice’ (the ‘recognised means by which individualism and dualism are expressed in modern social theory’). Barnes (2000) defines ‘the everyday language of agency’ as the language through which voluntaristic notions are expressed and distinguishes this use of ‘agency’ from attributions of causal agency. This is not to make a metaphysical claim but to differentiate the use of the term ‘agent’ to denote the causal actor from the attribution of free will in discourses of choice and responsibility that ‘paradigmatically . . . applies to people, and only to people’ (Barnes, 2000: 3). In making this distinction, Barnes avoids the conflation of the social agent with the causal agent that so bedevils theories of agency, and – almost, it appears, incidentally – establishes what Martin Kusch (2008: 131) has termed ‘a new field of sociological inquiry: the sociology of free will’. The distinction between attributions of causal agency and attributions of social agency is determined not by their grammatical or lexical structure but in their language function.

To illustrate this, we can consider these three sentences:

 The cow is on the road.

 The accident was caused by the cow walking across the road.

 The accident was caused by the man walking across the road.

Examined functionally – that is in terms of ‘use’ – we might consider the first sentence to describe some state of affairs in the world. The second attributes causal agency. The third, however, is somewhat ambiguous. It might attribute causal agency in the same manner as the second. Equally, we can hear it as a voluntaristic statement that assigns blame and assumes choice or free will. To the extent that we are tempted to hear the second statement (‘The accident was caused by the cow walking across the road’) as a voluntaristic statement, it is not related to the assignment of responsibility to the cow, but rather the responsibility is passed up the causal chain until it meets someone deemed agentic – perhaps the person who left the gate open. Similarly, we would not generally say ‘the cow chose to walk on the road’, and this despite the fact that the cow might have taken a different path.

One problem for libertarian models is to explain why the causal chain of action and effect is held to disappear at the human actor by virtue of the actor’s ‘agency’. Why is the person who left the gate open responsible and not the cow who walked on the road? Both the person and the cow could have behaved differently. Using Barnes’ model, the answer would be that it is because the person is an (adult) person and the cow is a cow. Only people and their institutions are accounted as socially agentic in voluntaristic discourses, and the temptation is to ask what sets them apart. Barnes insists that what sets them apart is not some intrinsic quality of people (a ‘state’) but the ‘status’ we assign them. It is because individuals within a society have the social status of a responsibilised (agentic) individual that voluntaristic notions apply to them. Statuses are understood here to mean classifications of all kinds. Statuses differ from states in that their assignment depends not on the intrinsic qualities of the status-bearer but on ‘the orientation to her of everyone else’ (Barnes, 2000: 150). A status is the focus of the ‘rings of action and expectations that define powers and prohibitions, rights and responsibilities, entitlements and obligations’ (Barnes, 2000: 148) as these are socially constituted. Voluntaristic discourses are used in the creation of social statuses, and one of the statuses created is that of accountable individual ‘agent’ within a society.

The status of ‘autonomous agent’ is turned into a state in the way other statuses are turned into states: that is, by using ‘processes of naturalisation, reification and hypostatisation’ (Barnes, 2000: 149). The autonomy is then projected onto the individual in the endowment of individual agency – so that the endowment of agency to the social ‘individual’ (as a status) appears to be the acknowledgement of a power or freedom that resides within the individual (as a state) and is, or can be, oppositional to the constraints imposed by a society. Barnes suggests that it is this understanding, with its implication of the power of the rational individual, that connotes agency as a ‘good’. What gives rise to this formulation of agency is that ‘individuals with rights to choose are easily rendered as individuals natively imbued with powers of choice. What is treated as autonomous becomes what is of its nature autonomous’ (Barnes, 2000: 149). This transition from individual ‘agency’ as an endowed status that distributes responsibility towards agency as a power residing in the individual is the sleight-of-hand that renders the choice autonomous and the agent both free and responsible.

Our intuitive insistence that agency is located in the individual as a capacity or as a power, Barnes argues, can be seen as consequent on the function of voluntaristic discourses. They function to create individuals as accountable social actors with the rights and social accountability of an agentic ‘individual’ as that status is socially (re)created. Voluntaristic discourses are not descriptive of internal states but are the medium through which responsibilities are distributed and individuals (causally) influence one another’s actions and so make social coordination possible. Barnes (2000) argues that problems arise when socially agentic discourses – voluntaristic language that is used performatively to assign social expectations and accountability – are understood in descriptive or causal terms to say something about the inherent quality of the individual. To put Barnes’ argument in Wittgensteinian terms, it is an error of ‘grammar’ that occurs when the performative attribution of agency to an individual as a status (using voluntaristic discourses) is understood to say something about the individual (to describe the state).

Accountability, Susceptibility and Intelligibility

Barnes’ theory of agency accounts for both routine and oppositional or ‘resistant’ action within the same paradigm: following a rule is not ontologically distinguishable from acting in resistance to it. Agency, for Barnes (2000: 65, emphasis added), is a status assigned by a group to a member of the group and being assigned the status is related to competence, where the ‘basic competence members presume in each other is that of accountability’. To be accountable, one must be able to intelligibly account for an action. This is not the same as following a rule. For Barnes, as for Wittgenstein, ‘grasping’ a rule is not an interpretation; it is not something that happens behind the scenes and is then demonstrated by conforming or resistant action. Rather grasping the rule is social and is shown by the action of either obeying the rule or going against it. It is not necessary to obey the rule in order to act in relation to it – but it is necessary to act publicly in a manner that is intelligible to be considered competent and held accountable. Disobeying a rule might incur sanctions, but these are determined by the rule. If an action is not intelligible with respect to a rule, the individual’s competence (their ability to be held accountable) could come into question.

Barnes conceives of people as essentially social and responsive to the judgements of others, and of voluntarism as reliant on that responsiveness. Barnes could perhaps have employed Wittgenstein’s notion of ‘form of life’ to argue that language can only be intelligible when our commonalities taken together create (in a very broad sense) ‘agreement . . . in judgements’ (Wittgenstein, 1968: §242). Barnes, however, makes the sociologically stronger argument (King, 2009), finding agreement in our inherent sociality and our need for the esteem of other people. Barnes (2000: 67) argues that ‘Agents who are disposed to co-ordinate their understanding . . . have to be agents who are affected by others.’

Acknowledging the influence of Goffman, Barnes employs the notion of ‘susceptibility’: as social creatures, people are inherently susceptible to the influence of others. Susceptibility is understood naturalistically and causally as a predisposition to be influenced by others’ judgements. This leaves room for judgement, in a particular case, of whether a person could have ‘acted otherwise’. In the presence of coercion or an overriding fear, a person may not be susceptible to the influence of other people. Paradoxically then, for an action not to be determined (for an individual to have a choice or to act ‘freely’) is for the individual to be susceptible to the influence of others.

Individualisation: Control through Choice Discourses

Barnes’ understanding of agency casts new light on the process of individualisation that I explore here by placing Barnes’ conceptualisation in dialogue with Beck’s Individualisation thesis. Readings that assume a libertarian framing of agency (see, for example, Brannen and Nilsen, 2005; Evans, 2007; Roberts, 2010) have suggested that Beck’s Individualisation thesis has been influential in fostering neoliberal subjectivity within sociological research. Dan Woodman (2009) contests this framing, arguing that Beck showed little interest in questions of agency and structure, and that a libertarian reading of Beck’s thesis is not justified. Certainly, Beck and Beck-Gernsheim (2002: 7) specifically reject an understanding of the individualisation process in terms of autonomy and empowerment: their stated concern is not to discount persistent inequalities but to explain why, in the face of the falling away of traditional operations that sustain class differentiations, ‘social inequality is on the rise’ (2002: xxii). They describe the individualisation process of late-modernity not as a liberating absence of constraint, but rather as ‘new demands, controls and constraints . . . being imposed on individuals’ (2002: 2). And one of these new demands is the demand to account for biography in terms of individual choice: to describe ‘All the messes into which one can get’ as self-made (Bauman, 2001: 10). A social demand to assume responsibility in narratives of choice does not reflect an understanding of agency in terms of autonomy from or resistance to social demand. Though Beck’s use of the more neutral German term ‘freisetzung’ (to describe ‘release’ from the traditional forms of control) is translated as ‘liberation’ in the English translation of Risk Society (1992), translator Mark Ritter (1992: 129) notes that he uses ‘liberation’ not in the sense of the English word ‘emancipation’ but as defined by Beck (1992: 128): as a release from the security of guiding social norms that is accompanied by a ‘control’ dimension that re-embeds individuals in the social. I argue here that, when viewed through Barnes’ conceptualisation of agency, the focus of Beck’s Individualisation thesis shifts from the ‘liberating’ dimension to emphasise that of ‘control’.

If, following Barnes, ‘agency’ is understood as a social (discursive) construct, locally, historically and situationally dependent, Beck’s notion of ‘freisetzung’ can be apprehended as a shift in the ‘rings of expectation’ surrounding the status of responsible individual or ‘agent’ as this is currently constructed in the West, such that the responsibilised individual is expected to ‘articulate individual lives in a way that excludes or suppresses (prevents from articulation)’ the means by which individual biography is linked to social structure (Bauman, 2001: 10, interpolation in original). Conceiving agency as status, individuals within late-modernity can be understood to live their lives, as they always have done, as responsible social agents, susceptible to the opinions of others and concerned with sustaining status within societies that are ‘ordered on the basis of status’ (Barnes, 2000: 84). Barnes’ model allows for the individualisation process of late-modernity to be coherently conceptualised as a socially produced requirement for individuals to ‘tak[e] responsibility for personal misfortunes and unanticipated events’ in ‘a culturally binding mode of attribution’ (Beck and Beck-Gernsheim, 2002: 24).

The individual within an individualised society is not then less governed or constrained. Absence of constraint is created as ‘choice’ – but it is a socially constructed absence of traditional sanctions and constraints, and does not endow power or offer liberty. Instead, it changes the social landscape in terms of sanctions and the institutions that express and sustain these changes. As Barnes (2000: 148) writes:

In the discourse of choice as absence [of institutional constraint], ‘the individual’ features as a social status, and what is permitted to the status is conventionally formulated in terms of rights to choose. The discourse is employed in the refashioning of social structure – in shaping and reshaping the rings of action and expectation surrounding statuses or social positions.

Beck’s Individualisation thesis can be understood, when so framed, not to deny but to explain the continued effect of status differentiations within an individualised society. While Barnes’ attention was confined to the status of an ‘individual agent’, people carry many statuses – adult, mother, man, worker – and these too can be understood as divisible into smaller statuses each defined by their ‘rings’ of actions and responsibilities that enable social coordination. Each individual attempts to balance many of these statuses, all carrying expectations that overlap and at times conflict. The shift in the status expectations of the late-modern ‘individual’ that requires the construction of autonomous-choice biographies can be seen to include, I suggest, a requirement for competent members of society to individually reconcile the status of autonomous choice-maker with differentiated status expectations.

Barnes’ monistic formulation of agency as status clears the ground for the investigation of the manner in which statuses are competing and being reordered as well as the means by which they are currently maintained. Barnes (2000: 116) describes the tension between status and the competent performance of status to be ‘an important part of the drama of everyday life’. In any situation where status-related expectations are not met, an individual, Barnes argues, has two options: to accept the consequences in the form of blame or sanctions that accrue, or to plead an excuse. While excuses can be successful in modifying expectations, the penalty is always a ‘degradation’ of status such as that described by Garfinkel (1956) and Goffman (2005). Though we may acknowledge differences in individual circumstances, even the ‘studied avoidance of blame’ retains ‘something of the quality of a miniature degradation ceremony’ (Barnes, 2000: 116).

Within an ‘individualised’ society, where a central social demand of the responsibilised individual is to demonstrate and claim autonomy, people become doubly burdened by responsibilities and expectations that curtail autonomy. At the same time, individual critique of the means by which they are burdened is silenced since to voice such critique is to make ‘excuses’ with a resultant loss of status. It is within this context that agency can only be claimed at an individual level if, simultaneously, ‘explicit analyses of a gendered structure of power [are] relinquished’ (Harris and Dobson, 2015: 150). For an individual to talk of structure as limiting or constraining is to use the language of the victim; it is to give up the ‘status’ of ‘empowered choice-making agents’ (Harris and Dobson, 2015: 153). This inability to describe personal experiences in terms of structural oppression is the other side of the requirement to explain biographies in terms of choice – the other shoe, as it were, of the individualisation process. In an individualised society, libertarian framings of agency in terms of the autonomy of rational individuals enlist ‘suffering actors’ (Harris and Dobson, 2015) in shifting the gaze of critical enquiry away from structural factors.

Understanding ‘the individual’ as a social status – one that is created and sustained through discourses – opens up space to rethink how status expectations intersect to enable or disable social membership. When the status of ‘agentic’ individual is tied to the construction of the self as empowered choice-maker, the status that is threatened by a failure to demonstrate autonomy and to self-attribute responsibility is that of responsibilised ‘individual’. Rather than being equally and a priori empowered with agency, individuals can be seen to be unequally impeded (through the responsibilities and expectations of intersecting statuses) in fulfilling the status expectations of a responsible ‘individual’.

Methodological Implications

Sociological concerns over neoliberal subjectivity have focused on two dimensions: attributions of self-blame and the creation of what Schwiter (2013) calls sociological ‘blind spots’ – locations where neoliberal discourses of choice-based responsibility serve to conceal rather than reveal institutional constraints that differentially delimit the choices available and require the choices to be made. People come to hold themselves responsible ‘for not only driving the car but also for paving the road’ (Frye, 2019: 728) while at the same time maintaining the notion that we all share the same roads. Schwiter (2013) illustrates these two dimensions in her study of young people’s approach to work–life balance. Schwiter (2013) argues that the non-gendered discourses of choice and individual responsibility in which the participants frame their responses conceal persistent gendered social norms and structural factors that may influence and restrict ‘choice’, such as the availability of child care or the risk of gender-differentiated criticisms of choices made. Rather than turning away from voluntaristic discourses towards analyses of ‘the structural facts of people’s lives’ (Brannen and Nilsen, 2005: 421), I urge that greater attention be given to the ordinary voluntaristic discourses in which people attribute free will and accountability. Analysing attributions of social agency in terms of their function exposes the manner in which voluntaristic discourses operate to differentially distribute responsibilities to individuals and hold them accountable. Importantly, it exposes the way that individuals and institutions participate in the construction and maintenance of inequalities in their use of voluntaristic discourses. To illustrate this, I indicate in this section some potential ways of ‘giving attention’ to the operations of everyday voluntaristic discourses.

Where local constructions of the responsible ‘individual’ include an expectation to construct the self as unconstrained choice-maker, self-attributions of accountability in choice discourses are inevitable – but they will not be the same choice discourses. Within everyday interactions, choice discourses are not generally used where there is freedom from constraint but where there are tensions between incompatible expectations. A person who feels no conflict between the status-related expectations of work and parenthood is unlikely to say, for example, ‘I chose to work’. Attention to disparities in the use of expressions of choice and of blame can serve to indicate locations where there are tensions between status expectations. Schwiter (2013), for example, speaks of gender-differentiated social criticisms of work–life balance choices. These criticisms will be expressed within voluntaristic discourses and will expose the way that individuals and institutions contribute to the construction and maintenance of gender inequality.

Investigation of changes to discourses of choice and blame over time can also reveal how conflict can be reduced or resolved through shifts in the status expectations associated with, for example, the ‘father’ or the ‘mother’: shifts that may serve to reduce or increase conflict for individuals in their struggle to fulfil the expectations of multiple statuses (McNulty Norton, 2021). I suggest that these shifts occur in response to status tensions and operate by reassigning responsibilities to interdependent statuses (and so perhaps relocating status conflicts). This has significance for how we understand choices ‘offered’ to children.

Giving attention to how voluntaristic discourses function to attribute accountability may also reveal where a range of options do not necessarily all operate as ‘choices’. To illustrate this, I draw on an investigation by Coxon et al. (2014) into the lack of uptake of non-hospital birth options in England. Despite a decades-long public policy that promotes ‘choice’ and supports non-hospital birth settings for cases of healthy pregnancy, hospital birth remains the preferred option for 93% of English mothers (Coxon et al., 2014). Pointing to prevailing beliefs of birth as risky and of the hospital as being the ‘normal’ place to give birth, Coxon et al. predict that the low uptake of non-hospital options is unlikely to change. They suggest that the positioning of the hospital as ‘normal’ may figure in women’s assessment of risk: that where hospital birth is the norm, a factor in assessing risk could be that ‘women who elect to birth without doctors present may be held to account if death or mishap occurs’ (Coxon et al., 2014: 65). Though this is framed in terms of women making rational choices in the face of risks of various kinds, quotations from participants’ narratives given in the article suggest that it might be useful to consider to what extent all options are constructed as ‘choices’. Two mothers narrate being asked if they ‘want a home birth’ by a midwife or birth partner while another explains that in her family, home birth is a ‘no-no’ as a result of the death of a baby when a family member ‘had a baby at home’ (Coxon et al., 2014: 58–59). These quotations, though limited, suggest that attention to the voluntaristic discourses that surround birth options might reveal that the attribution of individualised responsibility to the mother does not result a posteriori from her ‘choice’ of a non-standard birth facility – rather that the non-standard facility is already constructed as an individualised ‘choice’ within the discourses that surround child-birth in a way that hospital birth is not.

Barnes’ formulation also opens the way for investigation of how individuals employ discourses of choice to reassert status in the face of structural oppression. The young woman who cannot describe experiences of oppression without losing status as a responsibilised individual (see Harris and Dobson, 2015) is also the young woman who can gain status by publicly constructing herself as someone who ‘chooses’ to speak out – or even someone who chooses not to speak out, as when Anita Hill (1998: 132) famously says ‘I assessed the situation and chose not to file a complaint. I had every right to make that choice.’ Conceiving of the individual in terms of status offers new ways to understand the seeming dichotomy between the private silencing of the individual regarding what Harris and Dobson (2015) term ‘suffering-marked-by-structure’ and the public voicing of this suffering found in, for example, the #MeToo movement.

Conclusion

Libertarian conceptualisations of agency that incorporate notions of independent individuals all separately and privately following or resisting rules invite us to interpret findings dualistically, casting actors as either resistant or passive, either empowered agents or victims. At the same time, individuals required to self-assign responsibility become viewed as a priori endowed with an autonomy that creates their choice narratives as the expressions of ‘free’ individuals and silences social critique. I have argued that Barry Barnes’ conceptualisation of agency as status provides a model that can ‘move beyond’ the individualism and dualism of libertarian accounts.

Caldwell (2006) says of Barnes’ conceptualisation that it reintroduces naturalism, and Barnes accounts himself a naturalist of a sort (Roth, 1998). However, to construe this characterisation as criticism is to insist on viewing Barnes’ project as offering an ontological account of agency. Barnes provides a functional model of agency, one that allows the local and historical ‘meaning’ of social agency to be explored ‘within the grammar of the concepts within which it resides’ (Mahmood, 2005: 34) – that is, within voluntaristic discourses. For Barnes (2000: 149), the function of voluntaristic discourses is to make social coordination possible by attributing social accountability and distributing expectations, responsibilities and privileges. Choice discourses can be understood on this model, not as descriptions of unique individuals, but as one of the range of everyday voluntaristic expressions through which individuals influence one another and become created as accountable.

Barnes describes his model as ‘compatibilist’, but it is a compatibilism that is grounded firmly in collectivism rather than individualism. Instead of asking how order can be created and sustained given that individuals are autonomous, he begins by assuming that human beings are essentially social and enquiring into the social function of voluntarism. Barnes’ model offers an account of voluntarism as the medium though which people causally affect one another. Barnes provides, as Kusch (2008) notes, a sociological conceptualisation of ‘free will’. In principle at least, this account is complementary to naturalistic theories (such as the Actor Network Theory) that employ the term ‘agent’ to designate the actor as ‘the source of an action, regardless of its status as a human or non-human’ (Doolin and Lowe, 2002: 72). The two uses of ‘agency’ – that of the causal actor and the responsibilised individual – do not fundamentally conflict.

Barnes’ model does not promise equality: equality cannot be found in the differential distribution of expectations based on status. I suggest, however, that Barnes’ framework offers opportunities for exploring shifts in the distribution of privileges and responsibilities to statuses of all kinds through attention to the voluntaristic discourses that attribute and create them. The utility of Barnes’ model is that it offers a framework for understanding the function of choice discourses as the attribution of expectations to a member of society in the creation of the individual as responsibilised ‘agent’. I have argued that the individualisation process, when framed by Barnes’ model, can be understood primarily in terms of local changes in the mechanism of social control. While individualisation frees the individual from traditional means of control, the result is a socially created absence of constraint and a social demand that individuals create themselves as responsible, a demand that I link here with a local change in the construction of the ‘status’ of responsibilised individual ‘agent’.

This shift differentially imposes burdens on members of status groups of all kinds who are required to balance their varying status-based responsibilities with the status expectations of the late-modern ‘individual’. From a picture of the empowered individual pushing back against constraints, or freely choosing in the absence of constraints, Barnes’ model paints a picture of individuals attempting to maintain social esteem while balancing sometimes conflicting status-based responsibilities and expectations. Choice narratives can be understood as the ongoing efforts of individuals to maintain the status of competent ‘individual’ that requires them to conceal the furious paddling that is going on below the surface. It becomes then, not a question of whether sociological attention ought to be focused on agency or status, but one of paying attention to agency as status. In particular, it is a question of paying attention to how current local expectations surrounding the status of responsible individual ‘agent’ place different burdens and barriers on members of status groups in meeting those expectations.