No such thing as sociological excuses? Performativity,rationality and social scientific expertise in late liberalism

Introduction

What does sociological knowledge do? Even the staunchest among today’s objectivists would probably agree that social science does more than ‘just’ describe the world (Gane, 2019). While late nineteenth- and early twentieth-century debates were dominated by discussions of the relationship between interpretation and explanation (e.g. Glynos & Howarth, 2007; Apel, 1984 [1979]), towards the end of the twentieth century, the debate has decisively shifted towards the effects of scientific knowledge on, and in, the world.

This article explores the assumption that knowledge acts. This assumption, which I refer to as the performativity of knowledge claims (Bacevic, 2019), underpins a variety of approaches to sociology and its role in the world, from the idea that sociology can transform the world, or fight injustice and oppression, to the idea that sociological narratives provide justifications for bad, unacceptable, or criminal behaviour. Recent examples of the latter include ‘It is time people engaged in looting and violence stopped hearing economic and social justifications’, stated by Boris Johnson in response to the London ‘Riots’ at the time when he was Mayor of London, and French Prime Minister Manuel Valls’ repeated denouncement of ‘social and sociological excuses’ in the wake of terrorist attacks in France (Brandmayr, 2021a, 2021b). Traces of this assumption, however, can also be identified in British and French governments’ current opposition to Critical Race Theory and the related critique of ‘postmodern’ (in the United Kingdom) or ‘foreign/American’ (in France) theories.

The effects of social scientific knowledge are a long-standing object of sociological research and debate (e.g. Camic et al., 2011). They feature prominently in, for example, the critique of the role of economics in the 2008 economic crisis (e.g. Çalışkan & Callon, 2009; Esposito, 2013; MacKenzie & Bamford, 2018; Muniesa, 2014), research on the use of behavioural models to institute forms of governance (Guston, 2014) and the application of psychological and economic expertise in education policies (Williamson, 2019). Yet, this tells us very little about why specific disciplinary discourses attain performative power in specific contexts. Understanding how and in what sense(s) sociological statements can be said to act, and how and when they fail to do so, is relevant not only for the positioning of sociology in public debates but also for our own understanding of the discipline and its role in the world.

Rather than assume sociological statements are performative, the ambition of this article is to clarify the conditions under which they can – or, alternatively, fail to – do things. This line of inquiry shares a foundational concern with J. L. Austin’s (1956) work on excuses as well as the broader concept of speech acts (Austin, 1975, 1961; also Ahmed, 2017; Bauer, 2015; Butler, 2010). Austin became interested in the kind of words that do more than (just) describe the world. More importantly, he became interested in what he dubbed ‘conditions of felicity’: that is, (social) conditions under which (some) words can, or alternatively, fail to act. In other words, what would need to be true about social reality to enable sociological statements to do things?

The article is organized into four sections. In the first, I discuss the concept of performativity as it can be applied to sociological knowledge claims, highlighting the distinction between illocutionary and perlocutionary speech acts. The second section discusses the performative aspects of sociological statements, focusing, in particular, on their relationship to concepts such as explanation, justification, exculpation, and critique (see also Pleasants, 2021). To illustrate this distinction, I contrast Hannah Arendt’s Eichmann in Jerusalem with Adorno’s and Horkheimer’s Dialectic of the Enlightenment, to discuss what makes certain kinds of explanations ‘sociological’. I argue what this means for how we think about the performativity of knowledge claims, and, in particular, for why certain claims fail to act. The fourth part discusses what the attention to transformations of political power in late or (‘post’)-liberalism tells us about the sociopolitical conditions of performativity.

Doing things with sociology

The conditions under which some words can act is at the core of Austin’s typology of speech acts. In Doing Things With Words, Austin (1975) distinguished between statements that do something by the very fact of being stated, which he dubbed illocutionary; and statements that require additional conditions in order to do something, which he dubbed perlocutionary.

The example of an illocutionary statement would be ‘I apologize’: simply saying this is to apologize. The example of a perlocutionary statement would be ‘I hereby name this ship Queen Mary’: in order for that statement to act (or at least to act in the way the speaker had intended), other conditions need to be fulfilled. For instance, I must be given the position to utter these words in a particular ceremony, there might be a set of other ritual actions involved, like breaking a bottle on the ship’s bow, and so on. If these conditions are not fulfilled, the statement will be infelicitous: it will not have the power to create the intended effects.

The distinction between infelicitous (or ‘unhappy’) and felicitous (or ‘happy’) performatives is central to Sara Ahmed’s notion of non-performativity: that is, the conditions under which statements that should be performative act to prevent certain effects (Ahmed, 2006). Ahmed (2017) uses the example of ‘diversity-speak’ in institutions, which serves to prevent systematic reckoning with institutional racism and sexism. Similarly, Don MacKenzie (MacKenzie, 2006; MacKenzie & Bamford, 2018; MacKenzie & Siu, 2007) has introduced the notion of ‘counterperformativity’ to account for the capability of algorithms and other kinds of mathematical statements to create effects that run counter to those intended (see also Esposito, 2013).

Are sociological excuses, then, a case of ‘failed’ performatives? Are they explanations that have failed to explain, convince or justify, and are therefore consigned to a (mere) ‘excuse’? Furthermore, what is it about sociological statements that gives them the power to excuse? In order to answer these questions, we need to discuss what is it that sociological explanations do.

Social science normally entails an account of a phenomenon or course of events that makes reference to (1) the actions of (usually, though not exclusively) human actors; and (2) other supra-individual factors, from the most generic (‘society’ or ‘the social’) to specific categories such as class, gender, ethnicity, identity, institutions or their intersection. ¹ While the structure–agency dichotomy arguably no longer represents the key battling ground in social theory (e.g. Archer, 1995, 2003; Latour, 2005; Sawyer, 2002, 2003), most explanations in social sciences include factors that can be placed under one of these categories. Furthermore, sociological kinds of explanation tend to privilege elements of the ‘society’ or ‘the social’. Traditionally, this served to distinguish them from those thought of as ‘anthropological’, which tended to privilege culture, and ‘economics’, which tended to focus on relations of production. Of course, this distinction is, and always has been, tenuous: yet, pace Latour’s argument against the ‘sui generis’ nature of the social (e.g. Latour, 2005), most sociological research retains the concept of the social at least on the level of self-presentation.

On a minimal definition, then, all sociological explanations are illocutionary statements – to offer an explanation is, simply, to explain. The ‘success’ of sociological explanations, in this view, will depend on the proper utilization of sociological terms, like class, structure, normativity, power and so on, and the utterance of these explanations in a proper context – in a conference, scientific paper or a book. If one is committed to at least some version of realism, it will also depend on at least a degree of correspondence between those statements and social forces, or mechanisms, but this is not necessary for them to succeed in Austinian terms. The ‘success’ of sociological statements, in this form, is usually judged internally – that is, by other members of the sociological (or social-scientific) community: colleagues, mentors, reviewers and readers. They, obviously, can conclude that a sociological explanation has ‘failed’ in the sense in which it does not satisfy one or more of these criteria, but this does not mean that it has failed qua an explanation. In order for sociological explanations to (fail to) convince, justify or excuse, they need to do so on the level of the perlocutionary.

From explanation to justification

Justification seems to be the most frequent perlocutionary effect of sociological explanations. Statements like Boris Johnson’s ‘It is time…people who are engaged in looting and violence stopped hearing economic and sociological justifications’ suggest that politicians see ‘justification’ and ‘excuse’ if not as synonyms, then certainly as closely related. While some differences between excuses and justifications are obvious – after all, ‘excusing’ or exculpating connotes a response, or interpersonal agreement, whereas to justify is simply to offer a justification – it makes sense to explore this distinction. ²

Arguably the most comprehensive sociological treatment of justification is Luc Boltanski’s and Laurent Thévenot’s On Justification (2006 [1991]). Boltanski and Thévenot argue that existence in a common world presupposes an agreement on principles according to which things and acts in that world can be assessed, compared and evaluated. The discursive elaboration of these principles is what takes place in justification. All forms of sociality, in this sense, rest on practices of commensuration: it is by evaluating things, persons and actions according to a shared standard of acceptability (or unacceptability) that groups form, reproduce and survive. Importantly, this operation applies to consensus as much as to conflict or dispute: disagreement can only proceed through an underlying agreement on the principle through which competing claims will be evaluated. In the absence of this principle, we are consigned to violence. ³

This identification of violence as ‘that which lies beyond the realm of discourse’ converges with Jürgen Habermas’ concept of communicative rationality (e.g. Habermas, 1996 [1992]). Justification is thus a form of ‘speech-act’ that demarcates acts that are entirely unacceptable (those we consider antisocial or to which we attach adjectives like ‘inhumane’, ‘beastly’, ‘barbaric’ or ‘uncivilized’) and those we judge negatively (and punish) but that can still be evaluated from the standpoint of one of legitimate orders of worth. Boltanski and Thévenot (1999) identify six different orders of worth, or regimes of justification: civic, inspired, domestic, industrial, market and reputational; Boltanski and Chiapello (2007 [1999]) add a seventh, projective. What these regimes have in common with Habermas’ view is the belief in the centrality of justification as the key discursive operation that constitutes modern societies. In other words, it is not so much what certain acts can be evaluated in relation to as the fact that they can be evaluated in the first place.

This distinction is key for strongly contextualist accounts of morality, such as Gilbert Harman’s (1977). Harman argued that the statement ‘Hitler was morally wrong to kill Jews’ is meaningless – not because killing Jews was morally right or justifiable, but because trying to justify or condemn an act from the standpoint of a different moral order made no sense. While this argument can be used to support different kinds of moral relativism (Baghramian, 2004), justification does not preclude condemnation – it simply offers the possibility of evaluating a specific act from a standpoint presumably shared by all members of the community. Inclusion within the realm of discourse, therefore, serves to distinguish between those acts that can be evaluated, and those that cannot: whereof one cannot speak, thereof one must be silent.

From this point of view, to explain sociologically inevitably entails justification. After all, to explain sociologically is to give an account of something in social terms. ‘In social terms’, on one level, simply means within the realm of discourse: sociological explanations involve utilizing disciplinary language, often in specific social settings (such as conferences) or platforms (such as opinion pieces), conforming to culturally accepted standards of explanation. But sociological explanations are not only ‘social’ in the sense in which any discursive practice is necessarily social. They are social because they invoke elements of ‘the social’ – class, gender, identity, social structure and so on – to explain why people do things they do. ⁴ In other words, they use social or structural factors as reasons for action (cf. Parfit, 1984; Sawyer, 2011). ⁵

Is a recourse to social factors as reasons for action sufficient to qualify for a sociological excuse? Austin’s own work seems to suggest this is not the case (1956). To understand the difference, it is useful to contrast two classics of social and political theory: Hannah Arendt’s Eichmann in Jerusalem (1962) and Max Horkheimer and Theodor Adorno’s Dialectic of Enlightenment (1997 [1947]). Both were written by mid-century Jewish émigré intellectuals who had escaped pogroms. Both address the question of the Holocaust, not only as a moral or historical, but as a sociological problem: namely, how was an event as monstrous as that possible in societies that were, supposedly, paragons of rationality? Yet only one of them was accused of offering an excuse, and only one was sociological.

From justification to exculpation

Subtitled ‘A Report on the Banality of Evil’, Arendt’s Eichmann in Jerusalem provoked controversy shortly following its publication. First serialized in the New Yorker as reports from the trial of Adolph Eichmann, SS officer who was in charge of implementing the ‘Final Solution’, the book was published in 1962 with a slightly edited introduction. In this introduction, Arendt responded to some of the reactions the reports provoked – in particular, to the charge that explaining Eichmann’s actions by referring to social factors amounted to excusing those actions.

Arendt argued that what made Eichmann’s acts ‘banal’ was repeated reference to bureaucratic procedure to justify his actions. Rather than weighing in and out the moral implications of running a concentration camp, Eichmann and other officers were focused on procedure: in this sense, their behaviour was shaped by adherence to the socially and politically explicated rules that were not only legal but also widely accepted at the time. To Eichmann, organizing the transport of Jewish prisoners from occupied territories to camps or from camps to gas chambers presented a logistical, not a moral problem. In Arendt’s interpretation, his fault was in the unquestioning (in her formulation, ‘unthinking’) obeyance of procedures. In this sense, Eichmann and others like him were not, in any exceptional sense, ‘evil’: that is, they were not possessed by particular traits of character that impelled them towards acts of irreparable harm.

To many of her readers, this seemed to relativize the cruelty and inhumanity of the ‘Final Solution’. Arendt, of course, had no such intention, and vehemently argued against this interpretation. But the link had been made. When Eichmann in Jerusalem came out, the wounds of the Second World War were still fresh; Eichmann’s trial was possibly second to only the Nürenberg Trials in establishing not only legal, but also moral culpability of the Nazi regime. In the context of the Eichmann trial, Arendt’s ‘failure’ to ascribe the Holocaust to individual characteristics of the perpetrators stood out in part because of the nature of criminal justice: suggesting that complex events like the Holocaust cannot be explained by actions of a single individual could have been seen as ‘muddying’ the case for prosecution.

It is very likely that gendered perceptions also played a role: Arendt’s ‘cold-hearted, cynical analysis’ – sine ira et studio – stood at odds with those who expected that, as a woman and a Jewish woman at that, she should offer compassion and sympathy for the victims – in particular because her analysis involved the question of complicity of some of the Jewish leaders in mass deportations (Nelson, 2017). That a renowned political philosopher – who was, furthermore, Jewish – would have been more interested in analysing the causes of the Holocaust than in condemning it seemed unbelievable. Arendt’s ‘crime’, in this sense, was to offer reasons why someone acted the way they did, while her audience expected a moral denunciation.

This gives some insight into why Arendt’s analysis could have been received as an ‘excuse’. If to offer an explanation, in illocutionary terms, is to explain, this also means that it is also not to engage in another particular perlocutionary effect – namely, to condemn. For certain acts, such as large-scale atrocities or criminal acts, the absence of condemnation can by itself be constructed as an exculpation. To prioritize explanation over condemnation, as social scientists are likely to do, is to risk being perceived as providing an excuse. Yet, are all kinds of explanation via recourse to social factors ‘sociological’?

To understand the specificity of sociological ‘excuses', we turn to another classic: Adorno and Horkheimer’s the Dialectic of the Enlightenment (1947). Similarly to Arendt, Adorno and Horkheimer set out to explain the events of the Second World War as a world-historical political and philosophical event. The Dialectic of the Enlightenment was written during the war, in exile, and published in 1947. Rather than focus on the agency of specific individuals, however, their argument is much broader: namely, that the process of rationalization and demystification that enabled the rise of European modernity, eventually leads to concentration camps. Auschwitz, in their reading, represented the extreme but logical end of the process of rationalization.

One of the obvious differences between the Dialectic of the Enlightenment and that of Eichmann in Jerusalem is the genre. The Dialectic is much more infused with theoretical language; it started out as an attempt to explain the failure of Marx’s theory to bring about a global revolution of the proletariat, and, in particular, the fact Marxism did not predict the rise of National Socialism. To Adorno and Horkheimer, events of the Second World War were primarily a challenge to and an impetus to reform theory. While Arendt does weave theoretical arguments into Eichmann, they are supervenient on the event itself – ‘the event’ both in terms of the trial, and the broader set of historical circumstances.

This explains why Arendt’s analysis should not count as sociological. Sociology, for Arendt, was inclined to reduce the interaction of different factors to relatively deterministic social causes and that minimized not only the role of individuals but also their responsibility (Baehr, 2002; Walsh, 2008). Her claim, by contrast, was that the Holocaust could not be explained only through the analysis of existing tendencies – regardless of whether they are attributed to the Enlightenment, capitalism, modernity, industrialization or any combination thereof. We needed to remain attentive to ‘identifying the unprecedented’ (Baehr, 2002) – that is, tendencies that could not be deduced from theoretical frameworks that were dominant at the time, be they Marxist or functionalist. In other words, while Arendt was interested in explaining the Holocaust as unique, unprecedented event, Adorno and Horkheimer were interested in explaining it by reference to social factors – and, furthermore, those that could have been conceived within a Marxist (if adjusted) theoretical framework.

There is another important difference between Dialectic and Eichmann. Both drew attention to the irrationality inherent in supposedly rational institutions of modernity. However, Arendt did not mean to imply that modernity or rationality was somehow at fault. Her critique of Eichmann boiled down to him not being rational enough: in other words, not thinking for himself. By blindly following the rules, Eichmann gave up on what made him an individual: the possibility of disagreement with the line of action set by the dominant norms of the society. For Arendt, the social order provided individuals like Eichmann and others with the possibility to justify their own acts; but the responsibility was theirs alone. For Adorno and Horkheimer, in contrast, the source of this (un)thinking lay not in individual actions, but in the tendency of the Enlightenment to elevate a particular form of thinking to the level of doctrine. For Adorno and Horkheimer, the Enlightenment explained the Holocaust: Western capitalist modernity, with its emphasis on rationalization of production, provided the framework for the rationalization of death and killing. The Holocaust, for Arendt, was a reason for reclaiming rationality and individual responsibility (Benhabib, 1996); for Horkheimer and Adorno, it was a reason to be suspicious of the entirety of the Western liberal project.

Of course, reference to social factors does not erase individual responsibility, just like individual responsibility does not preclude identifying specific configurations of power that provide individuals with a justificatory framework for their actions. Explaining why people act in sociological terms inevitably involves drawing attention to structural conditions. This, in turn, opens the question of the responsibility of social actors or institutions that create or reproduce those conditions. This highlights another important characteristic of sociological explanations. Rather than analysing the origins of individual action – or lack thereof, as it were – they provide an epistemic position from which it is possible to construct a social critique.

From explanation to critique

The relationship between sociological knowledge and the possibility of critique is one of the key questions of Boltanski’s later work. On Critique (2011 [2009]) engages with the discursive formulation of critiques as a regularized element of both social transformation and social reproduction. In this framework, ‘critique’ is not the exclusive preserve of sociologists, academics or the intelligentsia. Yet, the use of social science to construct a critique opens up important questions of the relationship between diagnostic (descriptive) and evaluative (normative) registers of sociological writing:

(…) compared with sociological descriptions that seek to conform to the vulgate of neutrality, the specificity of critical theories is that they contain critical judgements on the social order which the analyst assumes responsibility for in her own name, thus abandoning any pretention to neutrality. (Boltanski, 2011 [2009]: 4)

Boltanski further identifies two kinds of, or approaches to, sociological critique. One is reformist: critique that uses social explanations to suggest and advise how social institutions could be designed to lead to different – more just, and more equal – outcomes (Boltanski and Chiapello, 2007 [1999]). This position Boltanski identifies with the role of the ‘expert’:

The project of taking society as an object and describing the components of social life or, if you like, its framework, appeals to a thought experiment that consists in positioning oneself outside this framework in order to consider it as a whole. In fact, a framework cannot be grasped from within. From an internal perspective, the framework coincides with reality in its imperious necessity. This engineering perspective is the one often adopted by sociologists when they are attuned to the officials in charge of large organizations (be it firms or organizations dependent on the state) and prove open and attentive to the problems facing these officials and the issues they pose. This position is one of expertise. (Boltanski, 2011 [2009]: 7)

Another position, however, takes aim at social institutions (and relations of power) as a whole. This is the kind of critique Boltanski refers to as ‘radical’: in this form of critique, sociological knowledge is used as a way of decentring and destabilizing the dominant social order.

The social sciences free themselves from expertise, and hence define themselves as such, by positing the possibility of a project of description…from a position of exteriority. In the case of ethnology or history, adoption of a position of exteriority is favoured by the distance – geographical in one instance, temporal in the other – that separates the observer from her object. In the case of sociology, which at this level of generality can be regarded as a history of the present, with the result that the observer is part of what she intends to describe, adopting a position of exteriority is far from self-evident. In sociology the possibility of this externalization rests on the existence of a laboratory – that is to say, the employment of protocols and instructions respect for which must constrain the sociologist to control her desires (conscious or unconscious). It is thus that descriptive social sciences can claim that they sustain a discourse of truth. It must be added that this truth claim, which is bound up with a description carried out by occupying a more or less extra- territorial post vis-a-vis the society being described, generally gives the social sciences, whatever they are, a critical edge (and this even, albeit in highly limited fashion, in the case of expertise). (Boltanski, 2011 [2009]: 7–8).

Many sociological explanations, of course, stop short of saying that the penal code is unjust and should be dismantled as such. Many take the route of reformist critique, furnishing suggestions on how it could be reformed to make it less unjust to those who are disproportionately affected – for instance, women, working-class people or ethnic minorities. Yet, evenin these forms, they point out the failures of the existing juridico-political order.

One of the effects of sociological explanations, therefore, is to point out to the unfairness or unjustness of specific social arrangements, and thus open the discussion about the responsibility of different kinds of social actors in creating, maintaining and transforming them. Of course, in cases like patriarchy, racism or economic inequality, it is often difficult to point to a single actor, institution or authority that can be held culpable. Yet, it is certainly possible to draw attention to specific ways in which specific institutions, regulations or practices – for instance, heteropatriarchal marriage or unemployment benefits – can and do reproduce forms of social relations that explicitly or implicitly privilege or oppress certain groups.

What sociological explanations do, in this sense, is furnish a mode of justification that is ‘extraterritorial’ to the dominant juridico-political order. Under contemporary legal systems of France and the United Kingdom, acts such as ‘terrorism’ or ‘looting’ are doubtlessly illegal. Depicting them as such also serves to position such acts as unequivocally condemnable and morally reprehensible; ‘beyond the pale’ in social terms. The effect of explanation is to bring something (back) under the order of discourse, making it legible in relation to a specific social setting, including its attendant value system – the order of justification. Rather than, for instance, seeing protesters as ‘rioters’ and ‘looters’, immigrants as ‘illegal aliens’ and women who steal tampons from shops as ‘thieves’, sociological explanations reach into the justificatory frameworks provided by actors themselves and realign them with the conditions and constraints generated by the social.

What sociological explanations can do, in addition, is reframe or redescribe behaviours that are condemned or punishable by law as the product of the same society that gave rise to those laws. In other words, they can ascribe responsibility not only or principally to the actors themselves but to the social and political order and, by extension, to those actors responsible for creating it. This reverses the question of ‘blame’ or culpability: rather than individual, it becomes institutional, or collective. In this sense, sociological explanations can, indeed, ‘exculpate’, if to exculpate is to turn the question of individual responsibility into one of institutional or structural determination.

This, of course, is by no means a necessary, or always intended effect of sociological explanations. In Austin’s terms, it is perlocutionary, not illocutionary. This requires us to pay specific attention to conditions under which sociological explanations are likely to be dubbed ‘excuses’ – that is, the social, historical and political conditions of felicity.

Independent irrational animals

The changing status of scientific knowledge has, in the past decade, acquired an influential diagnostic framing: the ‘death’ or ‘decline’ of expertise (Drezner, 2017; Eyal, 2019; Nichols, 2017). In this diagnostic, the weakening influence of scientific knowledge on decision-making is usually connected to trends such as the rise of alt-Right and populist movements in Europe, the declining public funding for and trust in traditional institutions of knowledge production and the rise of ‘new’ and social media with their potential to disrupt traditional ‘gatekeeping’ channels of information. Yet, even if we accepted that there is a shift towards a Weltanschauung that denies the role of science and, more broadly, criteria of truth, rationality and reason in governing societies, this tells us very little about its effect on sociological knowledge more specifically.

Understanding this requires us to look at the post-World War II transformations of the relationship between expertise and forms of governance. Two world wars – and, in particular, the Second World War – led to the disappointment of liberal intelligentsia not only in the capacity of governments to prevent mass slaughter but also in the ability of public reason to act as bulwark against such events. Adorno’s and Horkheimer’s critique of the Enlightenment, as well as Arendt’s analysis of the success of totalitarianism in eliminating ‘thinking for oneself’, herald this sentiment. The suspicion towards governments – socialist or capitalist – that aimed to influence the thinking of their subjects, combined with the recession following the post-war boom, eventually coalesced into a potent ideological project, usually known as neoliberalism.

Margaret Thatcher’s famous quip that ‘there is no such thing as society…there are individual men and women, and there are families’ is often taken to be the paragon of neoliberal political ontology. In this view, whatever else there may exist in the realm of the ‘social’ – friendships, workplace networks, unions, political parties – is ultimately irrelevant, and individuals are the main, if not only, subject (and object) of politics (Ledger, 2018; Rose, 1996). In lieu of large-scale planning attempts to deliver the ‘greatest good of the greatest number’ through, for example, education and healthcare, neoliberal governance prioritized markets as the way of ensuring the optimal distribution of goods and services. As the aggregate of choices made by rational, utility-maximizing individuals (Amadae, 2015), the idea of ‘market choice’ thus became synonymous with the freedom to choose.

Neoliberal emphasis on individual choice certainly did not chime well with sociological explanations, particularly the kind that that emphasized structural determination (cf. Gane, 2014). The fall of the Berlin Wall in 1989 and the decline of the Soviet Union further challenged Marxist critique – certainly in its radical version. The process of integration of Left critique into the neoliberal logic of capitalist enterprises in France is well documented by Boltanski and Chiapello in The New Spirit of Capitalism (2007 [1999]). In Britain, a corresponding process involved integration of sociological analysis in the reform agenda of New Labour (Hefferman, 2001). Exemplified in Anthony Giddens’ concept of structuration, this kind of analysis drew attention to the interplay between structural and agential forces (Giddens, 1998, 1984). Individuals were free to choose, but choices were not of their own making; therefore, the challenge to public policy became how to design institutions in ways that would enable people to make the right kind of choices (Hay, 1999; Rose, 1993; Shamir, 2008).

The end of the twentieth century posed a particular challenge for this approach. Consumers’ preferences – for instance, for salty, sugary, quick and cheap (‘fast’) food – were compounding a health crisis that came to be known as the ‘global obesity pandemic’ (e.g. Swinburn et al., 2011). This presented a problem for governmental regulation both in countries with public healthcare, like the United Kingdom, and in privatized systems like the United States: obesity meant sickness and sickness meant lower productivity. A popularization of behavioural economics dubbed the ‘Nudge Theory’ (Thaler & Sunstein, 2008) provided a workaround for this issue.

Departing from the notion of rational, utility-maximizing actor of neoclassical economics, behavioural economics argued that people’s decisions were guided – ‘bounded’ – forms of rationality, often leading them to fail to consider the broader implications of their actions. The way to change how they act, then, lay not in persuading them to think (or reason) differently, but in subtly altering the environment in which they make choices, what is known as ‘choice architectures’. This, however, did not involve transforming institutions or principles of distribution (Haydock, 2014). All kinds of choice remained reduced to market choice: for instance, supermarket aisles were ordered in ways that position fruit and vegetables closer to the entrance, but there was no intervention into pricing mechanisms or work/life patterns that would have made it easier for people to eat healthily.

This shift was significantly aided by the development of methods and techniques of data collection that did not require the kind of painstaking research or fine-grained analysis usually associated with (qualitative) sociology and anthropology (Savage, 2013; Savage and Burrows, 2007 – for instance, Big Data. Yet, a shift in modes of justification was more relevant. Drawing on Zygmunt Bauman’s dichotomy of ‘legislators’ and ‘interpreters’ as models of intellectual engagement, we could say that sociologists were no longer required as either. In lieu of ‘dependent rational animals’ (MacIntyre, 1999) of post-war welfare state, or utility-maximizing individuals that were epistemic exemplars of neoliberalism, the ‘model epistemic agents’ of late liberalism could best be described as ‘independent irrational animals’: isolated individuals, free to make choices, but not guided by reason. Consequently, scientific expertise involved in policymaking shifted from explaining why people act in certain ways, to shaping how they act.

A succinct illustration of the closure of performative space for sociological knowledge can be seen in the UK Government’s approach to the Coronavirus pandemic in 2020. The emphasis was on public ‘messaging’: that is, presenting information about the virus in ways that were meant to ensure compliance with specific forms of behaviour. When people diverged from these norms – for instance, when Londoners violated the ‘stay at home’ orders to hang out in parks – they were criticized (and, often, penalized) for non-compliance. Both punitive and ameliorative measures were targeted primarily at individuals. While sociologists drew attention to the reasons why some people may not be able to obey ‘stay at home’ orders – for instance, because they were in tiny and crammed flats (Fitzgerald, 2020) – addressing structural factors such as inequality, poverty or racism remained the domain of charities or volunteer organizations, while public policy remained content with ‘nudging’.

A similar focus on individual versus social factors is observable in the relationship to questions of race, racism and legacy of colonialism. United Kingdom’s Conservative government has maintained that racism is an individual issue, that is, a problem of personal prejudice, and can thus be addressed through individual measures, such as ‘unconscious bias’ training (Arday, 2019). The refusal to countenance analyses that speak of institutional or structural racism, for instance, reflects the post-liberal combination of economic neoliberalism and political authoritarianism, including a significant level of securitization and carceral justice (Burnett, 2017). This orientation to public policy requires downplaying structural and emphasizing individual factors contributing to behaviour. From this perspective, sociological explanations cannot but count as ‘excuses’.

Conclusion

Regardless of whether they diagnose its decline, or argue for its return, sociological narratives rarely question the assumption that knowledge can act. This idea, which I dubbed the of performativity of knowledge claims, underpins accounts of the role of social sciences ranging from self-affirmingly positive (sociology can transform the world) to critical and negative (sociology can provide excuses for criminal behaviour). This article analysed this assumption from the perspective of Austin’s distinction between illocutionary and perlocutionary speech acts, in order to draw attention to the sociopolitical conditions of performativity – in other words, specific conditions under which certain knowledge claims can or, alternatively, fail to act.

In and by themselves, sociological explanations do little more than explain. In order for them to justify, excuse or exculpate, other conditions need to apply. The article located the popularization of the idea of performativity of knowledge claims in the historical context of post-war welfare state, where sociological knowledge could be used to inform policies and public projects that aimed to transform social dynamics. The performative capacity of sociological expertise, however, lost its appeal with the weakening of the public sector and the growing influence of neoliberal policies. The shift in post-war political epistemology, away from ‘dependent rational animals’, towards rational, utility-maximizing agents and finally towards individuals whose rationality is limited and thus needs to be ‘nudged’, augured a different kind of expertise: one that shapes how people do things, rather than tries to explain why they do them.

This suggests an important corrective for the idea of performativity of social and human sciences. Disciplinary knowledge – sociological, anthropological, economic or psychological – only acts in the context in which there is an overlap between modes of justification common to a specific discipline and modes of justification common to a specific kind of political regime. By providing a justification for human action, sociological explanations make certain acts legible from the logic of the dominant order. What sociological explanations do, in this sense, is not provide ‘excuses’ for individual acts, but draw attention to structural factors contributing to them. Sometimes, this can lead sociologists to criticize social arrangements, and, by extension, those who have more power in shaping them: politicians and policymakers. Yet, even when sociologists do not take on the mantle of ‘radical’ critique, their explanations point to the possibility of transforming the existing order. In the sociopolitical environment shaped by long-term austerity and growing authoritarianism, sociological knowledge can only serve as an unwelcome reminder that there could indeed be such a thing as ‘society’.