The “Normative Turn” in Sociological Theory: Sociology’s Garden of the Forking Paths

The sociological tradition that Émile Durkheim inaugurated is deeply entangled with questions of normativity; morality [la morale] was the “centre and end of his work” (as reported by Georges Davy, cited in Lukes 2020). Against philosophers and psychologists, he attempted to build a science of morality intertwined with a science of society. Morality was construed as a property and function of societies that allows for solidarity, integration, and coordination. As societies change their organisational structures, different normative problems condition social action. Through naturalistic and scientific methods of observation and explanation, the sociologist can give a causal account of the normative determinants of social life. The argument, as is well known, is that social “facts” can be objectivised through the methods of the positive sciences.

The family of reinterpretations of Durkheim’s work that we classify under the rubric of Neo-Durkheimianism has, in the past couple of decades, aimed to push back against externalist, sanction-based ideas of the normative in favour of an internalist view of it. This shift, which may be connected to a “late” rediscovery of Durkheim’s (1995) Elementary Forms of Religious Life (P. Smith 2020, 131–32, 136–40, 180–220; see also P. Smith and Alexander 1996; Alexander 2005), leads also to acknowledgment of the dual character of morality as involving duty and ideals.¹ The former constrains us from the outside and is hence reducible to causal analysis. But the latter are created from within sociality through rituals, symbolic representations, and social effervescence and thus are irreducible to causal patterning (Weiss 2012). Normativity appears not so much as an external, factual determination bounding social action from without but as an intrinsic element in the texture of social action—normative ideals appear as symbolic or even sacred objects within the experience of participants themselves (Lukes 2020). At least part of the intrinsic complexity of the structure of social action consists in how all these emotional and meaningful elements combine and become moral ideals.

Agents gain cognitive and expressive representations of social reality through the ritualistic and symbolic aspects of social action. These take the form of beliefs, which then constitute the realm of social normativity as ideals. Representations that result from processes of symbolisation are not the reproduction of external reality but are expressive of normative commitments that bind people together and orient their engagement with each other and the world. These representations are particularly important in times of collective effervescence, when social groups gain an enhanced consciousness of their normative commitments. Mythical narratives, together with totemic and sacred representations, are paradigmatic of this inner complexity from which social experiences gain their normative texture.

This normative turn in Neo-Durkheimian sociological theory is not homogenous, however. In what follows, we discuss two trajectories through which it has crystallised in contemporary sociology. These variants share an antiscientificist reading of Durkheim and a focus on normativity as an internal, constitutive aspect of social action. They also share a view of normative ideals associated with spaces of ritualism in which groups gain consciousness of those past normative commitments that bind them together. Yet they differ on their conceptualisation of the normative aspects of reality as well as on the ways of making explicit the evaluative dimensions of sociological descriptions.

A first group of scholars makes the turn towards normativity on strictly sociological grounds and has become known as the strong programme in cultural sociology. In the work of its most significant proponent, Jeffrey C. Alexander (2003), this version emerges from a dissatisfaction with the externalist, sanction-based views of normativity that they see in Parsons’s work (Alexander 1988; Alexander and Seidman 1990). The (functionalist) autonomy of the cultural system is predicated upon an identification of morality with expedience, which then paradoxically ends up adopting the mechanistic language that it had attempted to reject, as it focused on the sanctions and rewards that set bounds to social action. This antihermeneutic approach separates the social sciences from interpretative techniques and loses sight of those “symbolic” aspects of social action (ritual, sacralisation, pollution, myth, metaphors, narrative) that suffuse social life with normative meaning from within. An explanatory account of values and norms displaces the symbolic modes through which reality is normatively experienced from within, thus conflating “presuppositional and empirical normativity” (Alexander 2021a, 160).

To overcome the Parsonian deficits, Alexander’s early theoretical model focuses on how, through interpretation (which entails both typification and invention) and strategization, agents rearrange their material and cultural environments in the light of situational elements. Interpretation and strategization are coterminous and symbolic processes, involve interactional and meaning-oriented aspects, and are not exclusively oriented by rational standards of sanctions and rewards. These symbolic aspects construct, typify, and evaluate reality from within and thus endow social action with an irreducible normative texture. Narratives and rituals allow agents to shift their “conscious attention” to the “components of meaning itself, to the classifications and valorisations upon which the existence of culture depends” (Alexander 1988, 322–23). Agents can now gain consciousness of those past normative commitments that bind them together and that are continuously reenacted for the continuation of collective life.

In his later work, Alexander (2011) shifts his attention to larger processes that he refers to as social performances that operate between rituals and strategy. For this shift, he draws upon aesthetic theorising of meaning-production and no longer upon so-called micro-sociologies. The normative texture of social experience now lies in its theatrical aspects, sacred and profane elements, icons, and purification rituals, rather than in the interpretative structure of social action itself. For normativity to be successful, and for the sociological observer to describe it adequately, it is crucial to recognise the emotional energy and performative power that is attached to symbols—how symbolic activity engenders effervescence, solidarity, and a sense of belonging. These aspects not only convince audiences but also performatively constitute them as collectives that share normative orientations. These representations contain cognitive, normative, and expressive dimensions all at the same time. Yet this “consciousness” does not display the self-transparency or self-sufficiency that is typically attributed to reflexive reason; instead, it relies on opaque, condensed, and emotionally charged objects that emerge out of rituals. This is how normativity works in every society—“modern” or “secular,” “traditional” or “religious.” The sacred-like properties make the normative texture of the world inevitably coarse and bristly.

Pushing against views that equate modernity with processes of rationalisation or secularisation, Alexander (2021b, 2021c) argues that rituals and performances determine the fundamental normative texture of modern society—the modern mind itself is ridden with taboos and totems. Modernity has not given rise to more rational contexts of normative action, nor can normativity adopt symbolic forms outside or beyond performances. There are rituals and theatrical elements underlying even the most seemingly rational practices, and these are the ones that determine the shapes and forms of social normativity. At most, modern societies have seen their context of meaning-production altered because the elements of social performance have been progressively “defused” and hence rituals that directly convince audiences have become rarer (Alexander 2011, chap. 2; 2017, introduction). Hinging on the question of their authenticity, performance in modernity acquires a theatrical rather than a rational orientation that aims at “re-fusing” script, actor, codes, and audiences. Experiences of transcendence, the sacred, the profane, the polluted, et cetera, are more difficult to perform authentically under modern conditions.² Still, they underlie every process of civil repair, any “societalization” of social problems through which groups engage in moral change (Alexander 2019). This overaestheticised, antirationalistic take on normativity is nowhere more evident than in his civil sphere theory, where he aims at deciphering the aesthetic scripts and codes upon which hinge the performance of civility (Alexander 2006).

All in all, Alexander seeks to retrieve the “primitive” elements underpinning even the most “secular” practices, thus arguing that our modern rationalist prejudices are to blame for not allowing us to see that the normative texture of civility is, in fact, religion-like. In framing the retrieval of culture’s autonomy within the retrieval of the primitive, all normativity is cast in the same ritualistic mould. He thus loses sight of—or even distorts—the specific normative challenges that emerge from a (post-) secular society (see Chernilo 2022). Solutions to social problems cannot be rationally self-directed by the participants or rationally assessed by the sociological observer. Instead, they depend on performances of sacred meanings that can be blocked by power dynamics and are never transparent vis-à-vis rational argumentation (Alexander 2019).

A first shortcoming of this approach results from this overly aesthetic view of meaning-making; namely, it precludes any notion of self-correcting practices and institutions operating in social life. The “sacred” may indeed be a crucial element of normativity, but conflating modern morality and religion leads to a weakening of the former. By generalising the religious-ritualistic-totemic model of normativity, this approach loses sight of the different empirical or historical enactment of the normative in social reality: of the substantive difference between a religious and a secular moral background (Blumenberg 1983; Taylor 2007). If normativity is held to coincide with the sacred, empirical instantiations of different modes of normative experience are dismissed from the outset.

Another problem concerns the lack of substantive or even formal standards that may orient the sociological observer in her descriptions of the social world. Alexander’s Neo-Durkheimianism recognises the intrinsically normative texture of the social world but is unable to develop normative standards to guide sociological observation. His work on the civil sphere has thus been criticised for displaying a consistent bias against conservative Americans and, even worse, for being normatively irreflexive (see Binder 2017; Baehr 2020a, 2020b). The sociologist is deprived of the resources to evaluate social reality or change, let alone to diagnose social pathologies. At best, she can take a celebratory stance if what she deems as sacred has been, to her eyes, successfully performed. Alexander’s conflation of the sacred and normativity weakens a crucial aspect that Durkheim (1973; see also Lukes 1969) himself had already envisioned in relation to the unstable role of intellectuals in a secular society (Callegaro 2012, 2016; Weiss 2021).³

In demarcating the space for a new science of morality, Durkheim had already argued against moral philosophers who grant themselves a privileged position to rationally assess or justify moral claims. In contrast, a science of morality is at most able to understand already existing moral beliefs in order to offer their best possible interpretation (Durkheim 2010, 35–62; 1973, 116–17; 1975, 330). If, however, the moral specificity of secular societies is cancelled, a major aspect of the Durkheimian project relevant to the normative turn is undermined.

A different approach towards normativity, one that keeps Durkheim’s earlier insight alive, has developed in France. Instead of grounding their reinterpretation on the theoretical and methodological problems of sociology, they offer a more philosophical and historical reading of Durkheim that sets him as part of a longer conversation about the normative and symbolic foundation of society in general and of modern society in particular. We may start with the question of normativity and symbolism in social practices and institutions. Norms and values not only are parts of action; they also appear as objects that are felt, and elevated to thought, by agents themselves. If externalism, conformism, and objectivism can no longer be attributed to a Durkheimian view of morality (Karsenti 2013, 2021), then symbolism becomes the intrinsic element of the structure of human agency. It is central to the immanent normative texture of social life through which agents gain a sense of moral personhood and their different ways of apprehending normative stakes (Callegaro 2012).

Marcel Mauss’s work is central to the development of such an anthropology of the symbolic and is paradigmatically introduced in his Essay on the Gift (Tarot 1999). But what is still missing in the early Durkheimian tradition is a full-blown conceptualisation of the multiple empirical instantiations of the symbolic as tied to a philosophy of symbolic forms. It is through the symbolic (which cannot be reduced to the totemic model of religions and the sacred) that agents gain different modes of consciousness of past commitments as they become sedimented in their social worlds.

Apart from ritualistic or totemic representations, social practices also traffic with linguistic communication. The symbolic pertains to all human practices but unfolds differently in different spheres and epochs. Symbols give rise to different modes of representations of our normative commitments, some more ritualistic—where social effervescence is vital—and others more rational, where intersubjective recognition and coordination are at their centre (Callegaro 2021a; Karsenti 2021). Even if it may be treated as primordial, the sacred remains only one medium in which the normative is felt and thought. Other symbolic forms emerge historically and give rise to variable modes of normativity. Put metaphorically, conceptual verbalization can indeed render the normative texture of the social world smoother or more refined. Alexander’s earlier approach might well be able to incorporate some of these insights, if interpretation is granted not only a typifying and creative function but also a clarifying one. But the displacement of interpretation for a notion of performance precludes such clarifications of meanings or the ironing out of “irrational” elements of sacred-like ideals. The sociological study of normativity demands a substantive analysis of the symbolic forms in which moral realities and normative practices find their concrete shape and some potential for rationality.

On the one hand, a philosophy of symbolic forms needs to give space to different symbolic expressions; that is, not only ritual representations and performances, but also figurations, enactments, or even meta-pragmatics. Minimally, this puts into question the thesis of the supposedly arbitrary or unmotivated referential character of all signs, upon which many of Alexander’s theorems rest—most notably, the notion of “stable signifiers, shifting signifieds.” On the other hand, it needs to account for the relationships among different symbolic forms, ranging from myth and religion to science, art, and democratic debate. It needs to account for the different normative understandings, cognitive gains, challenges, and interactions brought about by the alphabetisation of faith and the linguistification of the sacred, by the shift from totem to verbum, from nonaxial to axial religions, from enaction to narration to conceptualisation, from gestures to rituals, myths and icons to theories. Alexander’s approach seems to preclude the possibility of these different epochal modes of normativity.

In contrast, following Marcel Gauchet’s ([1985] 1999) philosophical history of the political and religion, Karsenti (2013) and Callegaro (2015) argue that Durkheim’s scientific discovery of the “social” is framed within the modern project of autonomy. This normative project finds historical concretisation under modern conditions of secularity—that is, when the normative force behind our practices no longer refers to the sacred as its ultimate ground. Modern secularisation entails not the normative emptiness of social life but the progressive rationalisation of its normative commitments and sources. Modern individuals no longer feel an attachment to their groups only in an emotional manner; rituals no longer fulfil their binding role, at least not without being challenged, because modern contexts of action demand at least some sort of rational awareness of the possible reasons for conduct and normative commitments.

Karsenti (2013) and Callegaro (2021a) return to Durkheim’s encounter with Pragmatism to provide the theoretical basis for the analysis of reflective thought as it emerges out of practical experiences. Categories of logical thought, critique, and rational problem-solving all have their roots in the inner symbolic complexity of social experience. These more “reflexive” modes of normative experience do not entail a self-sufficient, isolated consciousness, furnishing beliefs and later confronting a world of objects. Instead, they bring to light and make explicit the normative commitments underlying social practices that envelop actors themselves.

Cyril Lemieux (2012; see also Callegaro 2021b) has provided a particularly illuminating version of this normative turn in Neo-Durkheimian sociology. As sociology must describe the different normative grammars underlying social practices, its observations should be oriented towards increasing the possibilities of rational criticism, autonomy, and reflexivity within them. The sociologist draws her normative resources from a substantive analysis of the collectivity she observes. If, in her practice of observation, the sociologist increases the levels of reflexivity of social practices, this is because its object (the modern world) is normatively constituted by the project of autonomy that, in turn, entails that individuals must—or at least should—make normative commitments in an unconstrained manner. In this sense, Neo-Durkheimian theory finds normative foundations for its observations in the contexts it aims to describe. It is, in Weiss’s (2021, 20) words, “an internalist evaluation of socially constructed morality.”

Neo-Durkheimians “discovered” the internal aspects of the structure of human action by making explicit the role of normativity in the constitution of the social world. In so doing, causal and mechanistic explanations based on sanction-models are not altogether discarded but must play a secondary role. Moral ideals are at play in ritual-like situations, whereby individuals feel attachment to their groups by bringing into symbolic representation those past normative commitments that bind them together and guide their common actions. The study of normativity demands that we understand those spaces of ritualism that cannot be naturalistically explained but require paying attention to how symbolic forms operate.⁴ In the next section, we turn to Neo-pragmatism.

The history of twentieth-century American sociology is intertwined with the history of Pragmatism. While it would not be appropriate to signal something like a normative turn in Neo-pragmatist philosophy, it is nonetheless possible to pinpoint at least two waves of pragmatist influence on sociological theory. The first wave opened up a field for pragmatist-oriented empirical inquiry that was marked by a return to contextualised interactions; the second rendered explicit normative issues in the social world as well as in the activity of theorising that world. Both moments are inspired by fundamental pragmatist insights into the nature and structure of human agency.

Above everything else, Pragmatism is a philosophy of action (Joas 1993). Pragmatism develops from the premise that human agents find themselves constantly coping with ever-changing surroundings. Their experiences of the world are marked by flows of stability, uncertainty, and resolution. Communicative, cognitive, and reflective capacities must be seen in relation to situated contexts and coping with problematic situations—hence their fundamental insight that world-directedness is fundamentally marked by action rather than representation. Every action takes place in a particular temporal and spatial context that enables various forms of practical judgements from which specific modes of coping with environmental uncertainty and contingency emerge. Situatedness, the flow of experience, and practical judgements constitute Pragmatism’s triptych.

Rather than a theoretical development per se, the first pragmatist wave within sociology aimed at correcting concept formation through a recourse to case studies (Joas 1993, 14–54; Gross 2007, 193–201). Akin to phenomenology or ordinary language philosophy, Pragmatism was for sociology an early reminder that its concepts and abstractions ought to return to the flux of everyday human existence. Social relations are not fixed within systems and institutions that guarantee normative integration and shared meaning. Instead, these find stabilisation through situated interactions and are thus tied to ongoing transactions.

More than effectively theorising in a pragmatist vein, the early Chicago School conducted empirical research following these main tenets. In the words of Andrew Abbott (1999, 196–97, 205–17), the Chicago School’s commitment to case studies demonstrates that “one cannot understand social life without understanding the arrangements of particular actors in particular social times and places […] Every social fact is situated, surrounded by other contextual facts and brought into being by a process relating it to past contexts.” Not general laws, but local patterning of events; not systems and institutions, but interactions and situations in their own times and places. The categorial picture that emerges from these early works is one of ecological orderings attaining forms of dynamic equilibrium. The principles of ecological orderings could be discerned in naturalistic terms as sets of events and reciprocal adaptations causally (instead of reasonably) patterned into quasi-organic regularities within environments organized around dynamic forms of competition over scarce resources (autonomy, audiences, etc.) (Thomas 1909, 17–26; Thomas and Znaniecki 1926; Park 1936; for empirical application, see Bogue 1974; Hughes 1971).

Yet this early pragmatist-inspired sociology did not operate in a normative void. The classical American Pragmatists were all social reformers in different fields who had a strong commitment to science and democracy as rational modes of collective problem-solving. Traces of this normative attitude remain implicit in the works of the Chicago School (Gross 2007, 191–93). Building on what we have said in the previous section, we contend that a more recent normative turn also occurred here, when normativity is explicitly addressed as inherent to social processes. This is best expressed in Andrew Abbott’s work.

There, it is possible to find, first implicitly and then explicitly, a challenge to naturalistic construals of social orderings. Moments of equilibrium, however dynamic, cannot be defined from the outside, since the rules and goals orienting social processes prove to be rather undetermined and nonspecific. Every problem, either functional or moral, has a normative dimension that remains underdetermined and operates as the principle guiding reciprocal adaptations. Goals, strategies, rules, and norms all prove to be a matter of controversy among those actors who belong together in particular processes. All local adaptations or stabilisations involve different types of agreements, which in turn shape the definition of the process itself. Put differently, every local adaptation involves a meta-layer that allows agents to find normative grounds to decide how to adapt, creatively, to contingent events. Reflection, dialogue, disputes—that is, all the elements that structure from within the normative terms of social processes—can no longer be explained in causal or naturalistic terms alone. They involve emergent and contingent spaces of reasons that provide the grounds in the present upon which agents recode past events and shape possible futures (see Jouvenet 2016 for a succinct overview of Abbott’s work).

A first step to overcome the tendency to overnaturalisation of social processes consists in enlarging the metaphoric imagery of social dynamics (Abbott 1988). The goal here is to move beyond denormativised versions of social life and ecological orderings which, resembling Bourdieu’s theory of fields, draw upon economic-utilitarian metaphors that cast these contextualised sites as permanent struggles over resources among unequally positioned actors (cf. Liu and Emirbayer 2016). Against notions of supply and demand, Abbott resorts to legal and political imagery. Processes of ecological ordering involve changing “jurisdictions” over a range of events, which may be more temporary (“settlements”) or unstable (“bundles”) and do not involve complete alignment amongst actors with specific normative orientations. However, the goal is to grasp spaces of reasons whereby events are socially construed as problems that find different normative redefinitions and resolutions at different moments.

Abbott’s shift to a jurisdiction-like vocabulary is crucial to understand that values and norms are not masks that actors may wear in order to get their own way. In fact, Abbott has since argued that there is a “profound reason” why “processualism requires a commitment to normative reflection”; namely, that social processes consist “of the congealing of values into social things, the gradual weaving of choices and actions into the stable lineages of events that we call social entities. Value is central to this process, and so there is no escape from value in sociology” (Abbott 2016a, 253). Local adaptation processes are no longer seen as dynamic and contingent causal patterns of social forces from which regularities emerge, but as normative patterns of giving and taking reasons that shape what counts as a problem—and indeed the appropriate answers to them (Abbott 2016a, 202–27). For Abbott, the causal and the moral are not two separate ordering patterns; they are temporally distinctive lineages as they order any present, contingent event. The past encodes itself causally into the present, where contingent problems open the possibility of finding the grounds and reasons to normatively recode practices and institutions that, in turn, open new futures (Abbott 2016a, 270–76).

Abbott not only provides us with a complex view of the autonomous operation of the normative in the social world in terms of “linked ecologies” (Abbott 2016a, 33–74) whilst dropping the notion of dynamic equilibrium (Abbott 2016a, 202–27). More radically, he tackles the question of what the normative stakes are when describing such a world (Abbott 2016a, 277–92; 2016b). He rightly recognises that, compared to sociology’s descriptive tools, its normative lenses are rudimentary—they may even require the foundation of a subdiscipline within sociology itself (Abbott 2018). Sociology needs to find its way out of an unproductive struggle between proponents of value-freedom (normative “dualism”) and of unreflective value-drivenness (normative “monism”). Abbott’s (2018) solution is for sociology to turn to itself to clarify its normative commitments and descriptive practices in a “legalist” or “canonical” way. Rather than turning to reality to find definitive normative foundations, Abbott suggests that sociology must first clarify its own commitments through an analysis of its own practices (Abbott 2016a, 2018).

Two main challenges remain, however. The first has to do with how to describe this emergent normative reality without distorting its texture—to which his answer is that sociological descriptions must take a narrative, “lyrical” form (Abbott 2016a, 77–122). Central to this is a thin commitment to humanism: sociology can sympathetically describe moral dilemmas from within because it recognises its subjects in a humane way. Such an approach foregrounds the moral investment of participants, their agency, and meaning-making, as well as the emotions, actions, and meanings of the sociological observer (Morgan 2016). For Abbott, this approach can coexist with more positivist ones. The point is to make a “humane translation” of the social process that avoids “the Scylla of self-referential disengagement […] and the Charybdis of dogmatic politicisation” (Abbott 2016a, 287–88).

The second challenge is that sociological descriptions are intrinsically moralising because they give shape to the very processes they aim to describe. In a critical engagement with Burawoy’s proposal of a “public sociology,” Abbott argues that it is a mistake to anchor sociological descriptions on a sometimes unreflective and unmotivated desire to change reality. This desire is based mainly on prior normative judgements that disregard how people experience their own lives. The goal is to avoid falling into the patronising trap of diagnosing participants with “false consciousness” whenever the normative judgements of observers and participants are not aligned (Abbott 2018, 170). Moreover, sociological descriptions with an unreflective allegiance to a group or an ideal can be weapons for further polarisation in societies where mutual understanding is already weak. Sociology’s “humane sympathy” may ideally contribute to easing such normative challenges.

The trajectory of Neo-pragmatism in US sociology shows us that there has been an explicit turn to address the place of normativity in the social world—and indeed in the descriptive claims of sociology about that world. We want to briefly review another normative turn spurred by Pragmatism in a different intellectual context. What characterizes both trajectories is the emphasis on problematic situations and different forms of practical judgements. In this sense, Pragmatism provides a postmetaphysical impulse to the study of moral reasoning that is context sensitive.

In France, this line of inquiry was initiated by Luc Boltanski and Laurent Thévenot (2006). The pragmatist theory of truth serves them to underscore, and ideally correct, the asymmetry of moral and cognitive capacities that haunts the critical sociology of Pierre Bourdieu. Against Bourdieu, they argue that social actors do have the capacity to criticize in a way that is not altogether different from the “professional” critique of the sociologist. In fact, a crucial aspect of the social world is that it is ridden with acts of critique and justification that give shape to most practices and institutions. Actors give and take reasons according to the different senses of justice that are constitutive of their practices. In moments of uncertainty or controversy, the structuring power of these normative justifications becomes more evident. Boltanski and Thévenot seek to provide a sociological framework to think about situated processes within particular normative orders. Their emphasis is on how the present constitutes the grounds upon which normative justifications can legitimately take place.

Boltanski and Thévenot focus on reconstructing the “grammars” and “orders of worth” underlying patterns of justification and critique—that is, the moral contexts within which giving and taking reasons is required. They focus on the plurality of regimes of justification and valuation that are at stake in localised processes of reparation and stabilisation—that is, on the conditions of felicity of claims to justice that involve a delicate balance between generalisation and distance (Boltanski 2012). Thus, for Boltanski there is a fixed set of grammars of justification and critique, according to which different situations find contingent stabilisation, whereas for Abbott every problematic situation is already a composite of values and norms that entangle and disentangle itself at every turn. Hence, Boltanski reconstructs different “orders of worth” according to the model of a polity—that is, a shared sense of justice organised around some view of the common good.

It is not difficult to see how his model of testing structured around common goods resonates with the American Pragmatist’s hail of (American) democracy and science. Yet this may also be a source of certain theoretical aporias. To assume that polities of reason-giving are principally rational may be said to be as reductive as the opposite view that treats normativity to be exclusively ritualistic and performative. In this case, the implicit model is that of rational discourse itself. As the former is based on the sedimented normative force of the past, the latter attaches itself to the creative normative freedom of the present.

Unlike Abbott or Baert (2011), who developed their pragmatist approaches in dialogue with Buroway’s “public sociology,” we have said that Boltanski set his position vis-à-vis the critical sociology of his old mentor Pierre Bourdieu. The question is not so much about engaging nonacademic audiences or connecting sociology to a democratic civil society. Instead, for Boltanski the issue settles around questions of emancipation which, in a pragmatic sense, means orienting our descriptions to a “reinforcement of the role of critique” (Boltanski 2011, 150)—a position, incidentally, that is not altogether different from pragmatist Durkheimianism. What we want to underscore here is the fecundity for Pragmatism to make normative issues explicit.

There are of course other versions of the Neo-pragmatist tradition, but their crucial contribution lies in that they offer tools to analyse normativity in postmetaphysical terms.⁵ They emphasise the relations between knowledge, symbolisation, and action-problems as well as on how concrete, situated norms and narratives of contingent adaptations are constantly remaking themselves and shaping human activity. Its commitment to nonrepresentationalism and antifoundationalism enable a postpositivist conceptualisation of the normative that inheres in empirical sociology—without, however, dropping standards of objectivity altogether.

Two challenges to this tradition remain, however. One concerns a deeper problematisation of the disclosive dimensions of language; the other, a hasty generalisation of scientific reasoning to other social spheres. In both cases, these defects may be said to mirror the shortcoming reviewed in the previous section on the Neo-Durkheimian tradition: normative ideas are neither completely irrational and deployed through rituals nor wholly rational and subject to the rules of practical logic alone. It is the constant process of negotiation between either side to which sociology must always pay attention. Furthermore, as we also have indicated already, the locus of normative ideals in this tradition lies in the present. Normative ideas set up spaces of reasons that allow agents to reorganise and reimagine the social worlds they currently live in. A crucial aspect is that spaces of reasons cannot be studied naturalistically. The anthropological feature of creative and situated adaptation centres the operation of normative ideals on a contingent present that poses problematic grounds upon which entangled and plural moral claims can be made.

In the final section, we turn to Critical Realism, which stresses another anthropological feature—namely, reflexivity and a temporal orientation towards the future.

The “British” tradition of Critical Realism took form in the 1970s as part of conventional debates in the philosophy of the social sciences. Originally associated with the work of Roy Bhaskar (1975, 1979), at least part of its driving force was to offer a critique of the epistemological presuppositions of the positivist mainstream that was prevalent at the time. Not unlike the critique of naturalism and mechanistic causation that we saw in the previous sections, Critical Realism’s critique of positivism may be captured in the following three arguments:

(1) For positivism, the ontological gap between the “is” and the “ought” means that the realm of science and the realm of morality are fundamentally incommensurate. The laws of nature that apply in the former have no role in the latter and, conversely, the evaluative propositions of morality have no place in science. Against this dogmatic separation, Critical Realism contends that a science of society must attempt to bridge that gap without either making one dimension more relevant than the other or collapsing their specificity.

(2) For positivism, science is the study of “facts” alone, and these are always and necessarily observable. Things that cannot be measured, and preferably quantified, are hardly the subject of scientific knowledge. Against this, Critical Realism developed the notion that the task of theoretical explanation in the social sciences, as opposed to the mere description of empirical regularities, lies in uncovering those hidden mechanisms that become primarily apparent through their causal effects. There is, therefore, a much-heightened role for theoretical elaboration within Critical Realism.

(3) For positivism, normative considerations lie outside the scope of science because they are neither observable nor measurable. More radically, they are seen as intractable vis-à-vis the possibility of rational discussion. Whether they are conceptualised as subjective preferences or the realm of values itself is understood as ontologically irrational, a similar consequence ensues: normative considerations are to remain in exile vis-à-vis the development of scientific knowledge. Against this, critical realism doubles down on the view that normative considerations are subject to rational knowledge—though not to causal explanations. The question, then, is how to advance positive arguments as to how exactly this is possible in practice.

By the mid-1990s, the central tenets of Critical Realism had established themselves through a sustained elaboration of these propositions in both their critical and their positive registers. In fact, with regards to the rejection of positivism and the need for theoretical elaboration that transcends observable trends, Critical Realism does not stand too far apart from at least some of the key motifs of critical theory—not least in its own dispute with the positivistic tradition (Adorno 1976). Yet unlike critical theory, a unique feature of its project is the extent to which Critical Realism understood early on that a more complex, and eventually explicit, concept of normativity is required. Not unlike Boltanski’s critique of Bourdieu, for Critical Realism there is not one single immanent rationality underlying all social processes so that a single emancipatory project may be derived from the direct observation of social and political struggles. The realm of normativity emerges out of the interaction between the autonomous properties of both social structures and agents. To that extent, their critique of ontological and methodological reductionisms (i.e., individualistic and collectivistic positions) as well as their critique of “conflationary” social theorising (i.e., the idea that “practices” and “discourses” are both structural and agential at the same time) applies also to their defence of the autonomy of the normative (Archer 1995).

One way of explaining this understanding of normativity is to look at it from the standpoint of what a “normative fact” is. Here, a dual argument is required. On the one hand, Critical Realism posits that normative facts exist in the world independent of our will and desires. Indeed, their autonomy is not unlike the autonomy of all social facts: institutions are construed in order to protect or promote certain normative ideas; various discourses are available in society so that people engage in normative debates within cultural frameworks that are thus defined; individual and collective actions are triggered when certain values are violated, need to be defended, or are to be promoted. If “normative” facts are a particular kind of “social” facts, then understanding their normative specificity must be subject to the same “sociological rules” as those that apply to the conventional fields of the economy, politics, or religion. On the other hand, however, normative facts are unique, unlike other social facts, because of the specific role of normativity in society. A sociological understanding of what makes the normative unique requires that the evaluative commitments they trigger become internal to their very empirical description. When social practices are seen as “unjust,” institutions are deemed “corrupt,” and people behave in ways that are “untrustworthy,” the very adequacy of the sociological descriptions of those phenomena requires the sociologist to make value-laden statements that relate directly to what is normatively at stake in each case. Withdrawing moral judgement from these sociological descriptions would betray rather than enhance our comprehension of what is specific, indeed central, to how people experience the significance of normative facts.

As was also the case in previous sections, Critical Realism is anything but a unified field—the more so as its influence has now reached such fields as economics, international relations, and educational studies. The short space available in this article makes it impossible to account for the nuances and differences of its many versions; below, we will focus on how these arguments have been articulated within the sociology strand of Critical Realism. The main ideas have been most explicitly developed in works by Douglas Porpora (2015), Christian Smith (2003), Margaret Archer (1995, 2000, 2007, 2012), Andrew Sayer (2011), Dave Elder-Vass (2010), and Daniel Chernilo (2013, 2014, 2017).

If positivism played a central role in delineating Critical Realism’s target for the critique of mainstream social science in the 1970s, postmodernism may be said to have played a similar role in its more recent debates about normativity. A main shortcoming that Critical Realism underscores is that postmodernism’s dissolution of the subject leaves us with no reference point against which normative claims can be made, criticised, or redeemed. Normative facts, in the sense we defined them above, presuppose a subject for whom things matter in the world and who is able to make this felt through discourse as well as action. In Archer’s formulation:

One of the main contributions of this tradition for our understanding of normativity may be termed anthropological, as it is concerned with delineating explicitly those shared generic properties through which human beings are able to make, and then criticise, normative claims. For Critical Realism, these properties are necessarily generic, indeed “universal,” in the sense that while they are always and necessarily actualised in particular sociohistorical settings, at their source they are independent vis-à-vis social forces. All human beings, regardless of time and place, are able to pass moral judgements on the things that are relevant to them, are able to communicate the reasons for which these judgements are made, and are able to understand (i.e., accept or reject) the reasons that others give for their own assessments. Crucially, at stake here is the “reflexive” ability humans possess so that they can transcend a self-centred outlook and consider, albeit imperfectly, any situation from somebody else’s point of view. The cognitive property of “reflexivity” (Archer) is thus intimately connected with human empathy that also requires the development of an emotional bond (Chernilo 2017, 64–85, 181–205). The argument on the autonomy of the normative is reinforced because, at the same time as there can be no normative considerations that are devoid of substantive content—they always emerge as we busy ourselves in our relations with the world and with others—the very ability to pass normative judgements requires that we temporarily suspend our necessarily particular “being in the world.”

More importantly, Archer (2000) wants to resist the reduction of normativity to adopting the perspective of others or to society’s conversation. Rather than normativity being rooted exclusively upon answerability to others—in which case, the normative would resemble the pragmatist model of giving and taking reasons—Archer argues that agents are first and foremost answerable to themselves through the internal conversation. In direct engagement with American Pragmatism, Archer pushes back against Mead for oversocialising the “I” and the “Me” and argues instead that human beings develop crucial capacities in their presocial engagements with the world. More relevant, they gain the capacity to form self-monitoring projects that allow them to see the present from a future perspective, and to gain an awareness of their bodily and practical concerns that exist before a direct engagement with social normativity.

Archer’s internal conversation is not a mechanism by which society talks to itself through its members. As they talk to themselves, agents are able to put society in conversation with their own humanity, in its vulnerabilities and needs. Archer’s internal conversation is a mechanism by which agents are able to mediate society’s conversations and structural arrangements in the light of ultimate concerns that they themselves develop in their inner forum that take into account presocial aspects of their humanity. A social arrangement that doesn’t allow bodily and practical flourishing, that doesn’t keep up with our natural concerns and provide a sense of achievement, can thus be submitted to critical scrutiny. The internal conversation fuels a dialectic between social normativity, structural realities, and agential humanity. It constitutes the missing point that allows connecting particular lifeworlds and systems with universal aspects of humanity throughout time (for further discussion of Archer’s approach, see Caetano 2015).

However important normative considerations may be in social life, the argument is not that they constitute the “centre” of society—or, to remain within an anthropological language, whether “moral” concerns are central to who we are as a species. There seems to be two different versions of these arguments. In a “thick” formulation, to be found most clearly in Sayer’s (2011, 107–38) work, specifically moral concerns are central to who we are as evaluative beings. Moral questions are the most important to people; moral outrage or admiration is unique because it fundamentally appeals to the core of who we are as human beings—they are the things we care about the most.⁶

In its “thin” formulation, on the other hand, human beings remain evaluative beings, and their evaluations do have moral implications, but the events in the world that trigger moral feelings need not be only or specifically moral. Given the complex and pluralistic nature of contemporary society, it seems sociologically more accurate to accept that there is a wide range of interests, or objects of concern, that people develop through their lives, and they may develop normative considerations with regards to many of them. It is theoretically unnecessary, and makes little empirical sense, to suggest that there is a necessary hierarchy between concerns for world peace, our family, our professional development, or even the football teams we support (Elder-Vass 2010). Of course, some are of greater significance than others, but, insofar as people’s judgements are concerned, the key is that what matters to them, and how they matter to them, changes over time and is not set in stone. At stake is more the kind of relations people establish with different things that matter to them than with the substantive properties of the issues or “things” themselves.⁷ Still, both thin and thick versions concur in the claim that some evaluative relations to the world are of a uniquely normative kind, and their main difference is that the former predecides that only certain domains of social life are worthy of this attachment, whereas the latter leaves this open to what people themselves decide or are able to do with their lives.

Critical Realism has also made an explicit contribution to the methodological question of how the sociologist ought to study normative facts. Critical Realists have sought to develop arguments about how social scientists can grasp these concerns without, in the process, misconstruing them by theoretical or even ideological fiat. Here, a particularly relevant role is played by the notion of normative or evaluative descriptions. This is a way of making social scientific descriptions include rather than exclude the specifically moral dimensions of the questions under study. A classical formulation of this insight can be found in Hannah Arendt’s Origins of Totalitarianism. There, she describes Nazi concentration camps as “hell on earth,” and her argument is that her depiction of that particular institution becomes more objective than those that remain value free and only add negative moral judgements as something external to the “bureaucratic essence” of the camps.⁸ To be sure, most social research does not study such infernal institutions as concentration camps, and this offers a necessary note of caution for the use of normatively loaded terms. But to the extent that empirical studies are relevant to the things that matter to people, and that social scientists themselves are often concerned with the “big questions” of power, identity, or inequality, at least something of what is at stake in most social research lies precisely in the moral feelings that these issues trigger, the moral commitments they elicit, the moral implications they carry. As social scientists, then, it is simply not the case that remaining detached from the moral texture of certain—indeed most—social facts automatically helps us in our understanding of the phenomena under question. It is rather the opposite: being able to grasp these normative qualities is precisely what we need if we want to fully understand them, not least because the very issue of what exactly those normative considerations are remains open to interpretation (Sayer 2011, 7–11; Chernilo 2017, 230).

This touches on yet another epistemic implication of Critical Realism’s outlook—namely, the ways in which it understands the relations between lay and expert knowledge. Similar to what we have discussed before, Critical Realists also refuse to construe them as if the latter are always and necessarily superior to the former. Instead, as soon as the idea of normativity itself is at stake, the moral texture of normative concepts tells us that there is a mutual learning process between lay and expert conceptions of normativity. Social scientific concepts enter wider public debates and contribute to a clearer, or more reflexive, comprehension of some of our practices and institutions—as well as their underlying justifications. At the same time, lay concerns enter empirical and theoretical debates in the social sciences and help steer research towards or away from specific topics or issues. The goal here is to avoid the twin sins of critical theory’s patronising attitude towards people’s everyday concerns and postcolonial theories’ reified acceptance of people’s self-descriptions. To put it differently, the fact that all social actors are lay social scientists mirrors the fact that all social scientists are citizens and denizens of the social world they inhabit. As Margaret Archer (1995, 1–6) argues, even the theoretical centrality of the debate between structure and agency in sociology mirrors the ways in which people experience their personal concerns as partly autonomous and partly context dependent, and their social contexts as partly enabling and partly constraining.

Critical Realism connects its reflection about social normativity and the normative dimensions of sociological descriptions through what we may call a robust philosophical anthropology. The substantive normative foundations for sociological descriptions are drawn from an explicit account of humanity itself, in contrast to both Neo-pragmatism, which draws formal normative foundations from the structure of human agency, and Neo-Durkheimianism, which discerns its substantive foundations from the contexts it studies. This allows Critical Realists to engage with questions of the good and eudaimonia. Moreover, the focus on reflexivity as the anthropological capacity sets the locus of normative ideals in the future. Reflexivity opens up spaces of aspirations where actors shape their present and assess their past “for the sake of something.” The autonomy of human agency lies in the fact that things can always be otherwise—however strong structural concerns may appear, the possibility of constructing another, indeed better, future is central to people’s lay normativity.

We may perhaps bring our article to a close by offering an overview of the main dimensions for each of the three traditions. It is not our goal to integrate all three traditions into one but to stress how, while they are equally concerned with the problem of conceptualising the normative sociologically, they do so in their own particular ways. More importantly, while we set ourselves the task of mapping out various normative turns, we also submit to the Borgesian realization that the art of cartography is futile if the map contains all the vast details of what is being mapped.

Our road map has three routes to lay out different layers of sociological traditions. The first allows us to map out a methodological layer (see table 1). Sociological descriptions must come to terms with the fact that purely explanatory models run the risk of distorting how normativity unfolds in the social world. A running concern of the traditions we discussed above is the extent to which mechanistic or naturalistic explanations distort the normative texture of the social. The normative relates, in the last instance, to those aspects through which agents themselves muster meaning into their doings, through which these are connected to interpretative (dis)agreements, symbols, or imagination. In order to achieve this, sociology must include inter alia interpretative (Alexander), narrative (Abbott), and biographical (Archer) methods that grasp normative spaces of reasons, of ritualism, and of future aspirations. At the same time, sociological descriptions constantly attempt to find some normative foundations. As shown, these may or may not be context specific, may or may not be formal rather than substantive. Yet our inquiry reveals that there are options for these questions outside the Frankfurt School tradition—whether noncontextual and formal (Habermas 1984) or substantive (Rosa 2021).

What is crucial for all three traditions, therefore, is that any attempt to find normative foundations refers ultimately to the structure of human action. This leads us to our second route that allows us to map out a metatheoretical layer (see table 2). Or, in other words, it refers to the specific anthropological features—rituals, justifications, reflexivity—that are taken to be central to human sociality. Our point is that any attempt to seriously study social normativity cannot avoid a moment of reckoning with its own—explicit or implicit—philosophical anthropology. In a similar way, Frédéric Vandenberghe (2018; Caillé and Vandenberghe 2016) has argued for a moral sociology via an anthropology of the gift, phronesis (or practical judgement), and reflexivity. His work attempts to bring together different strands of the traditions we have examined here and points to a similar direction of, first and foremost, understanding our moral and ethical vulnerability qua human agents. In the end, what is at stake is a sociological engagement with the question of what makes us human. The Neo-Durkheimian tradition stresses symbolic-ritualistic aspects of human agency, their emotional-expressive orientation to rationality, and the integrative function of norms and values. The Neo-pragmatist focuses on adaptative-creative aspects of agency and the critical potential aspects of normativity in a deliberative-justificatory orientation to rationality. The Critical Realist foregrounds reflexive-caring aspects of human agency that relate to the imaginative dimensions of rationality. These also bring to light the internal contradictions of approaches that are exclusively concerned with decentring the human and subject.

Finally, this also proves that the domain of norms and values is complex and autonomous, and that it is possible to theorise their loci and temporalities in different ways. This is our third route that allows us to map out crucial theoretical layers (see table 3). The temporal dimension is a pressing question that must be tackled independently in theorising social normativity. We have shown that each approach emphasises one locus where normative ideas take a distinctive temporal shape associated with its function and agential property: through symbolic representations, spaces of ritualism retrieve past commitments that are binding for members; through justification and criticism, spaces of reasons are constitutive of present sites of contention for solving a problem or dilemma; through reflexive imagination, spaces of aspirations can devise novel paths of action in the light of some future state. As with the previous point, these may not be all the temporal connections that are needed to explain normativity, but they serve as a reminder of the temporal complexity inherent to the domain of norms and values.

As in Borges’s story of the garden of the forking paths, the more complex the internal structure of the normative maze, and the more apparently divergent the paths to its theorisation, the more clearly we realise how they eventually return to the foundational sociological questions about agency, norms, and values. Our threefold road map can indeed be useful to map out the normative layers of other sociological traditions.

Daniel Chernilo is Professor of Sociology at Universidad Adolfo Ibáñez in Santiago, Chile, and Visiting Professor of Social and Political Thought at Loughborough. He has written widely on the history of social thought, humanism, nationalism, and cosmopolitanism and is currently working on anti-Semitism and secularization. Some of his books are Debating Humanity: Towards a Philosophical Sociology (CUP, 2017), The Natural Law Foundations of Modern Social Theory (CUP, 2013), and A Social Theory of the Nation-State (Routledge, 2007). His personal webpage is danielchernilo.com.

Sebastian Raza is a PhD candidate in the Department of Sociology at the University of Cambridge. Some of his work has been published in the Journal of Classical Sociology and Distinktion: Journal of Social Theory. Together with Jorge Daniel Vasquez, he is coeditor of Theorein: Revista de Ciencias Sociales.

An earlier draft of this paper was presented in the Verstehen Colloquium in the Department of Philosophy at the University of Cambridge. Parts of this paper were commented on by Raquel Weiss and Filipe Carreira da Silva. We would like to thank them and the participants in the colloquium for their suggestions and remarks.