The other side of freedom: On the sociality of ethics

Introduction

Some anthropologists have of late been responding to the call of freedom. They have been summoned by, among others, James Laidlaw, who set out the case for ‘An anthropology of ethics and freedom’ in his Malinowski Memorial Lecture of 2001, published the following year as a much-cited article with that title in the Journal of the Royal Anthropological Institute (Laidlaw, 2002). After commenting on the absence of an anthropology of ethics and the lack of ‘serious sustained reflection on ethics’, despite ‘some individually brilliant discussions of morality by anthropologists’ (Laidlaw, 2002: 312) and lamenting the lack of a serious dialogue with moral philosophy, Laidlaw went straight to the source of the problem. ‘And then,’ he told his audience, ‘of course, there is Durkheim’ (Laidlaw, 2002: 312).

Most interesting arguments need an antagonist and Laidlaw’s is no exception. Anthropologists, Laidlaw urged, must abandon what he sees as Durkheim’s ‘peculiarly narrow conception of ethical life’ (Laidlaw, 2002: 317): in particular, his ‘equation of the moral with society’ (2002: 314), where society is ‘understood  … to be based on moral obligations, and indeed defined  … as being a system of moral facts’ (Laidlaw, 2002: 312), and where morals derive ‘directly from social collectivities’ (Laidlaw, 2002: 312). The result is a conception of the social that ‘so completely identifies the collective with the good that an independent understanding of ethics appears neither necessary nor possible’ (Laidlaw, 2002: 312). In The Subject of Virtue: An Anthropology of Ethics and Freedom (Laidlaw, 2014a), Laidlaw renewed and considerably elaborated on this call for an anthropology that links ethics to freedom and escapes from Durkheim.

The shadow of Durkheim: Ethics and morality

For Laidlaw, what is missing from Durkheim is ‘what is distinctive of ethical life, the complexity and specificity of ethical reflection, reasoning, dilemma, doubt, conflict, judgement, and decision’ (Laidlaw, 2014a: 23) His central claim is that what he calls the ‘ethical dimension of social life’ must be ‘reclaimed for social analysis’ (Laidlaw, 2014a: 44). An anthropology of ethics, thus understood, would explore ‘regularities and variations in how moral reflection is conducted in different contexts’ (2014a: 13). As an important step in his argument, which is meant to lay the groundwork for such explorations of particular cases and to establish the centrality of freedom to our analyses of them, Laidlaw introduces a conceptual distinction between ethics and morality.

If ethics is to be distinguished from morality, there are multiple ways of doing so and thus considerable linguistic and conceptual confusion about the matter. This matters, since the confusion is not a mere cacophony: different ways of drawing, or not drawing, the distinction express different, often incompatible first- and second-order views. Accordingly, Laidlaw deploys particular definitions of ethics and of morality that are carefully tailored to make his argument with maximum perspicuity. As he explains in the Malinowski lecture, his definition of ethics in relation to morality draws upon Nietzsche, Foucault and Bernard Williams.

From Nietzsche he derives a deep puzzlement about asceticism: how can we explain the prevalence of distinctively ‘moral’ as opposed to ‘non-moral’ values—in particular the self-denying values of morality embodied in and inherited from the Judeo-Christian tradition? From Foucault he takes over the idea of ethics as ‘the conscious (réfléchie) practice of freedom’—‘the considered form that freedom takes when it is informed by reflection’ (Foucault, 1984 [2000]: 284, cited in Laidlaw, 2002: 324). Hence the later Foucault’s preoccupation with projects of self-formation—with the ways individuals ‘take themselves as the object of reflective action, adopting voluntary practices to shape and transform themselves in various ways’ (Laidlaw, 2014a: 111). Ethics is thus distinct from moral codes—‘rules and regulations enforced by institutions such as schools, temples, families, and so on, and which individuals might variously obey or resist’ and which, according to Foucault, ‘are more similar across societies and historical periods than are the forms taken by ethical practice’ (Laidlaw, 2014a: 111). And Laidlaw follows Williams (himself much influenced by Nietzsche) in seeing ethics as ‘any way of answering the Socratic question “How ought one to live?”’ so that ‘morality, […] is a certain kind of answer to that question, one which, though it has come to play a major part in the ethical life of the modern West, is none the less in broader historical and comparative terms a “peculiar institution”’ (Laidlaw, 2002: 316). For Williams, unlike Nietzsche, what distinguishes morality is its form, consisting in rules, rights, duties, commands and blame. Moral thinking, on this account is ‘a matter of weighing obligations and deciding where one’s duty lies, and moral judgement rests on whether one chooses, whatever one’s desires or inclinations, to act in accordance with this duty’ (Laidlaw, 2002: 317).

Kantian morality is the purest version of morality thus understood and ‘Durkheim, in taking Kant as his starting-point, was already beginning from a peculiarly narrow conception of ethical life’ (Laidlaw, 2002: 317). So Laidlaw, combining these views, reserves ‘the term “morality”  … for ethical systems where self-denying values inform law-like obligations in this way’ (Laidlaw, 2002: 317). Morality is thus framed as at once a subset of ethics and in some sense opposed to it, as a form of ethical life concerned with placing constraints upon individuals and their freedoms rather than embracing these. Freedom defines the larger theoretical problematic, while the explanation of how people come to embrace constraints on their freedoms appears as the narrower question; it is a reversal of the priority Durkheim accords these questions, at least as Laidlaw presents him.

At the heart of Laidlaw’s concerns is his preoccupation with the ‘[h]uman capacity to reflect and criticize, to imagine other and higher standards than those that are prevalent in the surrounding society’ (Laidlaw, 2014a: 13), so that ‘there is never a simple equivalence between social requirements and moral obligation’ (Read, 1955: 249, as cited in Laidlaw, 2014a: 13). This focus on critical reflection is central to his idea of freedom. Hence his aversion to Durkheim, as he interprets him, for whom, he writes, morality is ‘just the corpus of socially authorized rules and values’ and ‘a function of social structure’ (Laidlaw, 2014a: 19).

Freedom as reflection

By contrast, Laidlaw relies upon a conception of freedom that we will put in question, in two ways. On the one hand, we shall argue that it is individualistic in the specific sense that it views freedom exclusively from the standpoint of the individual human subject, neglecting the dynamic between the individual and the collective (see Hausner, 2013). ¹ On the other hand, we shall argue that Laidlaw offers an unduly reflective conception of freedom that excludes the enabling of non-reflective action. We will suggest that these two limitations of the proposed conception of freedom are interconnected. We begin here with the second, since this is the aspect to which Laidlaw himself gives particular importance.

It becomes clear in The Subject of Virtue (2014a) and in his later writing (Laidlaw, 2010, 2014b, 2017) that Durkheim is not Laidlaw’s only antagonist. His case claiming that his way of conceiving freedom is ‘intrinsic to the very idea of ethics’ (Laidlaw, 2014a: 149)—a ‘concept we think about but also think with’ (2014a: 10)—has two other targets: sociology as ‘the science of unfreedom’ (2014a: 3) and the view of morality held by ‘the dominant schools of anthropology in the mid-twentieth century’ (Laidlaw, 2017: 187). The sources of the dismal ‘science of unfreedom’ (among these he lists ‘some versions of evolutionary and/or cognitive psychology,’ ‘structural Marxism,’ ‘Lacanian psychoanalysis,’ and, indeed, ‘sociological theory descended from Comte, Marx, Durkheim and Bourdieu’ (Laidlaw, 2014a: 3, 9)) lead us to ‘re-describe the conduct of responsible agents as the effects of causal “forces” or the mechanical self-reproduction of “objective structures” ’(Laidlaw, 2014a: 10). As for the mid-20th-century anthropologists (he cites British structural functionalism and Boasian cultural anthropology), they reduced the study of morality to explaining ‘how custom binds’ (Fortes, as cited in Laidlaw, 2014a: 24) and studying ‘socially approved habits’ (Benedict, as cited in Laidlaw, 2014a: 24). This, Laidlaw argues, is to bypass ‘all that is most interesting and puzzling about their declared subject matter’ (Laidlaw, 2017: 187). As we will argue, one cannot conceive of freedom without reference to some specific set of constraints to be overcome; his foils shape and limit Laidlaw’s own conception of freedom in ways that make it less ethnographically useful than it might be (see Keane, 2016).

In this article we examine the case, advanced by Laidlaw, for an anthropology of ethics that defines ethics in terms of freedom. We argue that his discussion of the concept of freedom is flawed and propose an alternative conceptualization that is less restrictive than his exclusive focus on individuals’ ‘reflective freedom’ and focused instead on the dynamics that link the social to the individual. There are many ways of being free that include freely embracing norms that inhibit free-ranging reflection. We shall then seek to show, through an ethnographic case study, that his argument has significant limitations concerning freedom’s normative background—limitations which further attention to Durkheim’s thinking, despite its own limitations, might have avoided.

Laidlaw begins from what he calls the descriptive claim that ‘people…are evaluative:’ (Laidlaw, 2014a: 3) that is, they engage in self-interpretation—the practice of what Charles Taylor calls ‘strong evaluation’—where we approve, identify with, and commit ourselves to some wishes and desires while being ambivalent about, deprecating, regretting and struggling against others (as cited in Laidlaw, 2017: 188–189). Henry Frankfurt, similarly, sees such ‘second-order desires’ as basic to freedom of the will and required to makes sense of moral agency (as cited in Laidlaw, 2017: 189). Laidlaw also cites Webb Keane’s account of ethical life as ‘roughly my own’, which views ethics in terms of social interaction and as pervasive in everyday life while ‘built on basic facts about human action’, within which ‘morality systems’ invoke ‘self-conscious and institutionalized reflexive systematization’ (Laidlaw, 2017: 183, 186). With all this we have no quarrel. Indeed, we regard Keane’s account of ethical life as the most promising currently available. A key innovation of Keane’s is to link first-, second- and third- person perspectives in his account, with interaction as the mediating term between interior ethical experience and public, explicit moral norms: ‘what links the psychological and historical dimensions of ethical life is the dynamic of everyday social interaction’ (Keane, 2016: 33). Starting from a ‘scene of accountability’ in which individuals reckon with one another, this approach reminds us that freedom is a product of and always exercised within a ‘social and historical world’ (Keane, 2014: 452). Where we part company with Laidlaw, as we shall now argue, is over his particular Foucault-inspired conception of freedom as reflective, involving ‘ethical complexity, dilemma, reasoning, decision, and doubt’ (Laidlaw, 2002: 315) and ‘standing back or self-distancing’ (Laidlaw, 2017: 188). ² This conception of reflective freedom has the undesirable effect, we shall suggest, of limiting our appreciation of ways in which people can be free and thus, if we follow Laidlaw’s way of linking ethics to freedom, of limiting our view of the range of the ethical.

Freedom: Concept and conceptions

In Chapter Four of The Subject of Virtue, entitled, ‘The “question of freedom” in anthropology’, Laidlaw deploys a general concept of freedom as he understands it—of which the so-miscalled ‘“Western liberal” concept of freedom’ is one instantiation (Laidlaw, 2014a: 138). As we shall explain, we prefer to speak of a range of conceptions that fall under a single concept of freedom, of which they are interpretations. While criticizing his understanding of the concept, we agree with Laidlaw in seeing a range of variation that the constitution of agents and the forms of reflection in which they engage can take, thus confirming his important observation that this range will turn out to be greater than ‘mere thought experiments might suggest’ (Laidlaw, 2014a: 149). His project is to engage in a ‘comparative analysis of the varying social forms that freedom can take’ (2014a: 139), which will require an ‘ethnographically usable understanding of freedom’ (Laidlaw, 2014a: 93). How, then, does Laidlaw conceive of freedom?

Referring to what he calls ‘one of the defining essays of twentieth-century liberalism’, namely, Isaiah Berlin’s ‘Two Concepts of Liberty’ (1969), Laidlaw correctly observes that ‘Berlin’s positive-negative distinction is not itself fine grained enough to serve as a framework for anthropological analysis’ (Laidlaw, 2014a: 143). Although Laidlaw does not explicitly set out such a framework, this chapter can be read as suggesting the general outline of such a framework and, more importantly, ways of rendering it more fine grained. The chapter can be read in light of one of the earliest critiques of Berlin’s essay by R. B. McCallum (1967), a critique further developed and sharpened by the legal philosopher Joel Feinberg (1973). This would require us to accept (which Laidlaw does not) that there are not two concepts of liberty, or freedom, but one.

Freedom, let us say (building on Feinberg’s analysis), is a unitary concept with a triadic structure. Stating the point abstractly, when we attribute, identify and compare (even aim to measure) freedom we are concerned with agents who are subject to constraints that limit or prevent their experiencing or bringing about some outcome.

Feinberg elegantly portrays this structure with the following schema:

___is free from___to do (or omit, or be, or have)___(Feinberg, 1973: 11)

How the agents or subjects of freedom are conceived and conceive of themselves has long been a topic of widespread contemporary interest and it is, of course, largely influenced by and central to the thinking of the later Foucault, discussed in detail in Laidlaw’s Chapter Three, entitled ‘Foucault’s genealogy and the undefined work of freedom’. The topic is ‘techniques of self-formation’, the question posed by Foucault being how people have ‘thought that they had to make themselves as subjects’ by ‘crafting a self fit to exercise freedom’ (Laidlaw, 2014a: 100), how ‘subjects are formed in power relations, including how the self acts on and shapes itself’ (Laidlaw, 2014a: 101). Foucault’s discussion is historical, or rather genealogical, and (significantly) confined to the West, beginning in ancient Athens, where ethical work was aesthetic in nature in the service of moderation: it consisted in ‘establishing a proper relation to one’s desires, by establishing domination by the rational and reflective over lower parts of the self: keeping one’s desires under command’ (Laidlaw, 2014a: 120). Foucault’s story is one of a transition from ‘the active shaping and styling of the self into a skilled and reflective user of one’s pleasures, to the relentless inquisition of the self in order to seek out and overcome the unacknowledged desires that might lead one into sin’ (Laidlaw, 2014a: 120–121).For Laidlaw this suggests the anthropologist’s task: namely to ‘describe diverse forms of ethical reflection and practice so as to bring their distinctiveness—together with the contingency of those we take for granted—into focus’ (Laidlaw, 2014a: 124).

How the agent is conceived is inextricably bound up with the question of how the constraints upon freedom are to be understood. Here Feinberg’s acute analysis is most helpful. He makes two distinctions, which cut across one another, as illustrated in Table 1: that between positive and negative constraints and that between internal and external constraints.

OPEN IN VIEWER

Table 1.

The concept of freedom.

Positive constraints, most obviously, involve the activity or presence of obstacles, ³ but, as Feinberg remarks, there is ‘no doubt that some constraints are negative—lack of money, strength, skill or knowledge can quite effectively prevent a person from doing, or having, or being something he might want. Since these constraints are absences, they are “negative,” and since they can be preventive causes, they are constraints’ (Feinberg, 1973: 11). Of course, the absence of something is the presence of its opposite: lack of strength is the presence of weakness, lack of knowledge the presence of ignorance, and so on: whether a constraint counts as negative or positive depends on how it is described.

How we draw the distinction between ‘internal’ and ‘external’ constraints depends, as Feinberg noted, ‘on how we draw the boundaries of the self’. If we contract the self sufficiently so that it becomes a dimensionless non-empirical entity, then all cases are external. Other narrow conceptions of the self would attribute to its ‘inner core’ a set of ultimate principles or ‘internalized values’, or ultimate ends or desires, and relegate to the merely ‘empirical self’, or to a world altogether external to the self, all lower-ranked desires, whims and fancies (Feinberg, 1973: 12–13).

Moreover, how the agent or subject is conceived also determines which among internal constraints is going to count as restrictive of freedom, for, as Feinberg further remarked, whether the internal restraint is taken to be a restriction of the self’s freedom to act depends upon ‘how we model the self, that is, upon which of the element of the “total self” we identify most intimately with’. If we are prevented by some internal element—an impulse, a craving, a weakened condition, an intense but illicit desire, a neurotic compulsion—from doing that which we think is the best thing to do, then the internal inhibitor is treated as an alien force, a kind of ‘enemy within’. On the other hand,

 … when the inhibitor is some higher-ranked desire and that which is frustrated is a desire of lesser importance albeit greater momentary intensity, we identify with the desire that is higher in our personal hierarchy, and consider ourselves to be the subject rather than the object of constraint. When the desire to do that which is forbidden is constrained by conscience, by the ‘internalized authority’ of the prohibiting rules themselves, we identify with our consciences, and repel the threat to our personal integrity posed by the refractory lower desire which we ’disown’ no matter how ‘internal’ it may be (Feinberg, 1973: 14).

Finally, what freedom enables and unfreedom hinders or prevents—what we have called the ‘outcome’—can be extremely various. It can be material or ideal, this-worldly or other-worldly, private or public, individually or collectively oriented (as in Caroline Humphrey’s account of Russian-Soviet freedom as Mir—‘the euphoria of the universalized “we”’ (Humphrey, 2007: 5, as cited in Laidlaw, 2014a: 147) in ‘all-encompassing community’ (Laidlaw, 2014a, 146)—it can take the form of having, being or doing. It can be expressed in many different and incompatible ways, ranging from the language of choice, describing, say, the goods and options that the citizens of capitalist societies are, according to the Friedmans, ‘free to choose’ (Friedman and Friedman, 1980) to what Laidlaw, referring to the Jains, describes as ‘a project for the attainment of spiritual perfection and enlightenment through the rigorous ascetic elimination of all desire, passion, and attachment’ (Laidlaw, 2014a: 168). Here, too, how it is understood is inseparably linked to the other components of freedom.

Some sense of the range of conceptions of freedom and how they relate to the shared concept can be gained if we focus on the question of constraints as illustrated in the table above. Laidlaw is entirely right, as we have indicated, to write that ‘there is no single liberal conception of freedom’ (Laidlaw, 2014a: 142) and, in particular, we endorse his distinction between the Kantian conception and accounts by ‘liberal thinkers of a more pluralist persuasion’ (Laidlaw, 2014a: 167). Nonetheless, it seems helpful to think of liberals as predominantly concerned, in the spirit of J.S. Mill and Tocqueville, with protecting individuals from external interventions in their lives by state authorities and the pressures towards conformity of public opinion, thus focusing their attention on the upper left segment of the table. Within the liberal tradition, libertarian or ‘neo-liberal’ or ‘market fundamentalist’ views take the narrowest view of what is to count as constraining freedom. Hence, for example, Friedrich von Hayek’s definition of freedom as ‘the state in which a man [sic] is not subject to coercion by the arbitrary will of another or others’ (von Hayek, 1960: 11). This occupies, we might say, a small slice of the upper left-hand box in Table 1. Isaiah Berlin’s account of ‘negative liberty’ was, when first presented, located in that box but, interestingly, under pressure from critics, he extended it into the lower left-hand box to incorporate as restrictive of liberty lack of resources and the like that are attributable to political failures to remedy it. In general, social democrats and socialists are concerned with such restrictions as limits to liberty. As for Foucault, we can see that all the action lies in the right-hand section of the table: self-fashioning and technologies of the self all, in their various instantiations, involve prioritizing and re-ordering internal desires and motivations, prioritizing some, demoting and generating others, identifying the ‘total self’, in Feinberg’s phrase, with those that are higher-ranked.

We suggest that this unitary account of freedom is a helpful framework within which to interpret what is at issue in the debates raised by Saba Mahmood’s (2001, 2004) and Charles Hirschkind’s (2001, 2006) studies of Islamic reformist piety movements in Egypt. Laidlaw writes about these authors’ approach that they portray the pious individual’s ‘making of some of one’s desires (the desire to please God, etc.) into the motor of one’s actions’ (Laidlaw, 2014a: 154). According to Mahmood, women in the Egyptian piety movement aim thereby to ‘fully inhabit the norms to which they subscribe’, reinforcing authority and conforming to Islamic teachings (Laidlaw, 2014a: 140). Thus far, Laidlaw writes, we can agree with Mahmood that they pursue a kind of freedom unrecognized by ‘liberal feminism’, that involves the re-ordering of desires and the reassigning of constraints. It is the further step of aiming to cultivate incapacity—to render submission involuntary and visceral, ‘outside the purview of consciousness’ and ‘a nondeliberative aspect of one’s disposition’ (Mahmood, 2005: 137, as cited in Laidlaw, 2014a: 152) that would, were this achievable (which Laidlaw argues it is not in a world of conflicting plural values), render them, according to him, unfree. For they would then be unfree, not just in the liberal-feminist sense, but in the generic sense that they would no longer be acting selves with inner lives, with what Anthony Kenny (1975) calls ‘two-way powers’, able to identify with some desires and not others, ranking some higher and treating others as constraints.

Laidlaw, as we have seen, takes ethics to be conceptually inseparable from freedom as he understands it. Something like the conception of freedom advanced by Henry Frankfurt and others, in which ‘we “step back” from and evaluate our own thoughts and desires, and decide reflectively which desires we wish to have and to move us to action’ is, he claims, ‘intrinsic to the very idea of ethics’ (Laidlaw, 2014a: 148-9). Moreover, such reflection, distinctive of ethical life, involves ‘reasoning, dilemma, doubt, conflict, judgement and decision’ (Laidlaw, 2014a: 23). He thinks that this is so because ‘value pluralism and conflicts of values’ (Laidlaw, 2014a: 167) that are irresolvable (to which some but not all liberals draw attention) render ‘reflective and thoughtful self-direction  … always necessary because there is no way to resolve the conflicts definitively’ (Laidlaw, 2014a: 168). Thus he chides Mahmood and Hirschkind for their inattention to this inescapable feature of the human condition, seeing them as giving a distorted and misleading account by accepting as true the picture of their subjects’ lives as consisting in ‘submission to God that is internalized to the point of being instinctive’ (Laidlaw, 2014a: 147) and ‘in the realization of one’s true self in absolute and involuntary fearful obedience to God’ (Laidlaw, 2014a: 154). This critique relies on an argument: that ‘[a]ctually living a life requires doing so with reference to values that make conflicting demands, and managing the inherently irresolvable tensions between them’ (Laidlaw, 2014a: 169). Laidlaw presses the critique yet further by alluding to Berlin’s famous denunciation of the abuses of the so-called concept of ‘positive liberty’ becoming ‘an astonishingly reliable and flexible resource in the justification of tyranny’ whereby ‘political thinkers and statesmen’, in the name of freeing the ‘real’ or ‘rational’ self from oppression, ‘invoke the name of liberty, in this “positive” guise, to legitimate the imposition by force of quite other values and to pass off compulsion to submit to those values as liberation’ (Laidlaw, 2014a: 144). The suggestion is that the Mahmood-Hirschkind account involves ‘another such positive concept of freedom as the realization of a prescribed version of a true or perfected self’ (Laidlaw, 2014a: 147). To this Laidlaw has two objections: that it is empirically impossible and that it invokes a perverted (positive) notion of freedom. ⁵

It is true that Laidlaw raises the question of ‘how far along [the] trajectory it is ever possible to go’ towards ’the realization of one’s true self in absolute and involuntary fearful obedience to God’ (Laidlaw, 2014a: 154). But this in turn raises the question of whether Laidlaw’s conception of reflective freedom (involving ‘stepping back’ and evaluating alternatives) is the best way to capture the ‘ethical dimension of social life’ and the range of variation across it. For this idea of ethics as involving ‘reasoning, dilemma, doubt, conflict, judgement and decision’ is, we shall argue, partial. It offers a one-dimensional, deliberative view of its object, focusing on reasoning and reasons rather than on intuitions, described by Mercier and Sperber as ‘like mental icebergs’ where ‘we may only see the tip but we know that, below the surface, there is much more to them, which we don’t see’ (Mercier and Sperber, 2017: 7). To claim that ethical life must involve stepping back from and evaluating thoughts and desires is to hide from view all the ways in which people avoid and, to a lesser or greater extent, succeed in eliminating these experiences. It also, we shall argue, occludes the normative background of freedom: what shapes the identity of agents, the alternatives they face and the constraints that limit these.

Freely embracing unfreedom

Must we accept the claim that all projects and practices worthy of the name ethical involve or encourage such reflection? Mahmood’s and Hirschkind’s works portray a project that is not oriented toward the cultivation of participants’ critical capacities. But why exclude it for ethnographic consideration under the banner of ethics? Laidlaw may be right that it is almost certainly not possible for the subject to eliminate their own capacities for moral freedom, even where they wish to, but what should we make of their claims and aspirations to do so? Even when undertaking an ethical project aimed at making oneself less free (in Laidlaw’s reflective, deliberative sense), the subject is still electing to do so freely.

To see how this can be so it helps to think adverbially. For the noun ‘freedom’ denotes a condition that involves the coming together of internal capacities and external opportunities (bearing in mind Feinberg’s observation concerning the variability of where the boundary between what is internal and external lies). To attribute freedom to people lacking either one or the other can be a kind of mockery. A door may be open but are you free to exit if immobilized by fear or despair or uncertainty? (Self-censorship can be a good example). Or real material circumstances can render people’s aspirations unrealistic. (Are the poor, to adapt Anatole France, free to sleep under the bridges of Paris?) The adverb, by contrast, focuses attention on the presence or absence or, better, the extent of self-direction. To what extent are your thoughts and actions your own? To what extent are they the result of disabling coercion or, in some way, pre-determined or pre-programmed or, in G.A. Cohen’s (1989) phrase, ‘engulfed’ by the ideas and values of others? Cohen writes of ‘engulfment’ as occurring where one’s will is not one’s own. Unlike constraint, where an ‘agent has a desire or project which he and we can identify as his own and whose fulfilment is made impossible or difficult by something felt as outside him’, engulfment ‘envelops the will’. But, Cohen writes, ‘no engulfment is total. The self is never effaced by environing circumstances or other selves’ (Cohen, 1989: 189, 190).

That the work of self-cultivation is never complete is acknowledged by both Mahmood and her interlocutors in the piety movement. As much as they strive to cultivate the habits and emotional attunements that make, for example, prayer an inner desire, even a compulsion, they do not aspire to make action automatic. ‘Insofar as the point is not simply that one acts virtuously but also how one enacts a virtue (with what intent, emotion, commitment, etc.), constant vigilance and monitoring of one's practices is a crucial element in this model of ethical formation’ (Mahmood, 2001: 838). And yet the freedom exercised by the pious subject seems to be rooted less in critical deliberation and judgement than in spiritual discernment and emotional responsiveness, and the constraints she seeks to overcome are precisely those capacities—bodily needs, desires, and wilfulness—that constitute the core of the subject in many liberal conceptions of freedom. In Feinbergian terms, God’s help is recruited by the pious subject in overcoming constraints, conceived of as internal to the subject, that hinder the achievement of her ethically desirable ends.

It may be true that the pious are not critical or deliberative with respect to the form of life they are turning to embrace. Moreover, their very goal is to cease to exercise their capacity for reflective freedom in Laidlaw’s sense. Indeed, in his most recent discussion of the question, Laidlaw appears to recognize this, writing that we ‘do not have to be exercising this capacity continuously, or indeed at any specific time, for it to make a difference to the quality of our conduct’. For, ‘if we were not at times capable of conscious reflection, or if we never in our lives exercised this capacity, our everyday and habitual conduct would not be pervasively saturated with meaning, as the ordinary ethicists rightly insist that it is’ (Laidlaw, 2017: 188). But on this interpretation, reflective freedom fades into being merely a background capacity that may be only rarely exercised, rather than a recurrent manifestation of ethical life.

This leads us to question whether deliberation, judgement, decision and doubt should be thought of as the definitive hallmarks of freedom. The actual practices of piety (from discussion circles to sermon audition) are intensely intellectual and evidently involve reason, deliberation and debate. However, their goal seems to be, instead of a ‘stepping back’, as though to a more impartial or even ‘third-person’ perspective, rather a kind of ‘stepping into’ their pious lives. We speak of ‘leaps of faith’; if you step back, you can’t leap (metaphors are important clues to meaning). To the extent that their acts of commitment lead them to live lives that precisely view deliberation, dilemmas, reasoning, decision and doubt as constraints from which they seek to free themselves, are we to conclude that their lives are no longer the subject-matter of an anthropology of ethics?

Furthermore, the people that Mahmood and Hirschkind describe are converts to their new-found piety (like Joel Robbins’s Urapmin (2004). What about the vast majority of people who are simply continuing to live their lives mostly free from the constraint of reflective deliberation? Though plainly contrary to his purpose, Laidlaw’s proposal runs the risk of violating Keane’s injunction that ‘we shouldn’t decide in advance what ethics will look like’ (Keane, 2014: 444) and even of falling into what Robbins has called ‘the trap of promoting western common sense models of social action to the lofty position of universal theories’. (Robbins, 2007: 295). Those writing under the banner of Ordinary Ethics (Lambek, 2010), by contrast, have emphasized the unremarkable, tacit quality of ethics underlying everyday action, and in the back-and-forth between Laidlaw and Veena Das this distinction has emerged as a debate over which form of ethical life to privilege (Das, 2014; Laidlaw, 2014b, 2017). This sort of binary distinction between deliberative reflection and the relatively unreflective behavior, which has become a feature of several contemporary framings (see Zigon, 2007 for another version) is misleading, we suggest. There is a wide range of alternative ways in which people can seek to live their lives in relation to the prevailing norms of their society. The ethnographic task is to identify the forms that ethics takes within and across situations; it requires tracing the ethical lives of the pious across their entire lifespans, not only in the moment of their deciding to embark on a career of piety or in their most unreflective inhabiting of their ethical selfhood.

Cases like the piety movement reveal ways in which unfreedom in the form that minimizes reflection as critical deliberation can indeed be freely embraced, even if never fully attained, and as such they do not plausibly fall outside the bounds of an anthropology of ethics. As we have argued by way of Feinberg’s formulation, all actually-existing conceptions of freedom include both a particular understanding of the acting subject and some notion of what it is that the subject must be free to do; this helps us to see how any ethnographically-describable ethical project is conceived in relation to a particular set of available or imaginable alternatives, from which other possibilities are excluded. Individual freedoms and social constraints cannot be so neatly divided from one another in the situations in which ethical subjects actually find themselves. It is, in short, misleading to tie ethics to individuals’ freedom, understood in contrast with conformity to constraining social norms, if what matters is their mutual entanglement. For purposes of anthropological thinking about ethics, freedom should be understood in relation to the moral norms that both shape and constrain it. It is to the shaping of freedom in relation to norms, values and what people take to be their obligations that we turn now.

The voluntary and the obligatory

The interrelationship between individual freedom, on the one hand, and relatively explicit and binding moral norms on the other, is a central focus of the ethnographic analysis that one of us, Miller, has developed on the basis of fieldwork in the Hunza Valley in Northern Pakistan. There, the tension between freedom and moral obligation is thematized across many domains of social life. What follows is a brief sketch of a particular local conception, or set of conceptions, of freedom that illustrates our argument that a more open-ended concept of freedom is needed.

Hunza is an administrative district (tehsil) located along the western side of the Hunza Valley in Pakistan’s rural, mountainous, northern Gilgit-Baltistan region. In this rural and relatively remote region, nearly all households rely on a mixture of agrarian subsistence activities, cash-cropping, and wage work, which often involves migration to the urban centres of ‘downcountry’ Pakistan or abroad. In Central Hunza, where Miller conducted fieldwork in 2006–2008, most people are Burushaski speakers and about 85% are Nizari Ismaili Shi‘a Muslims, members of an esoteric minority branch of Islam who acknowledge the fourth Aga Khan as their Imam or spiritual leader. The institutions of Ismaili Islam include not only arms concerned with religious education and the dissemination of doctrine but a set of non-governmental development organizations, the Aga Khan Development Network (AKDN). In the course of their routine participation in these institutions, which ranges from attending communal prayers (namaaz) to engaging with the NGOs of the AKDN as beneficiaries, employees or volunteers, to quotidian acts such as reading and discussing articles in The Ismaili magazine, people in Hunza are encouraged to imagine themselves as part of a unique transnational community (jama’at) that is defined by faith and modern in outlook. Though subject to a diverse array of media, institutions and discourses, their faith gives Ismaili in Hunza access to a particularly complex and persuasive vision of how to live and what it is to live a good life in the present era. Central to this vision is the precept that one’s primary duty to God in this world is to care for and improve the lives of one’s fellow humans.

A central ethical practice by which people in Hunza attempt to achieve this ideal is a form of service to others that is referred to as khidmat (service) or, alternatively, by the English term ‘volunteerism’. These two terms were used almost interchangeably, the choice seemingly determined by linguistic context: referring to the practice itself, khidmat was more commonly used when the language of talk was Urdu or Burushaski, whereas volunteerism was used when speaking in English or with English-dominant interlocutors. Referring to those engaged in the practice, on the other hand, volunteers was used more frequently regardless of the language of conversation. In spite of the apparent semantic opposition between the two terms, emically, people in Hunza do not mark a distinction. In practice, the two dimensions of the practice are fully entwined; what counts as ‘volunteering’ is specified by the religious obligations of service.

Volunteering usually takes place within a relatively formal, organized framework; it can refer, for example, to the usher duties that typically fall to boy scouts and girl guides at formal village functions (for which they wear khaki uniforms similar to those worn by scouts in other places, except for the green and red handkerchiefs that mark them as Ismaili scouts); or to the task of preparing tea and snacks for Friday prayers at the jama’at khana, which is carried out by women and allocated according to a regular rotation; or to taking part in work crews assembled for a one-off task involving common property or collective benefit. Examples of the latter included repairing the gusulkhana (the room, located on the grounds of the jama’at khana, where corpses are washed and prepared for burial) or clearing away the rubble of a collapsed house so that a new structure could be built for a ticket-ghar and teashop connected to the newly-renovated Altit Fort (an AKDN project). Teaching ‘night-school’ religious education classes was largely conducted by volunteers, as was minding the shoes of the congregation while at prayer. More spontaneous acts of service were sometimes also referred to as khidmat, but often, it seemed, by analogy with these more formal, public practices. Importantly, the proximate beneficiary of khidmat is a collectivity (a village, the local congregation (jama’at), the wider Ismaili community (also jama’at) or even humanity as a whole) whose overall physical and moral state is improved by the act of service, rather than a single individual in need. The ultimate motivation for of all such forms of volunteerism was said to be an impulse stemming from the recognition of God’s boundless generosity to humankind (insaniyat); this generosity cannot be repaid directly, but upon recognizing the full force of this truth, the natural response is to follow his dictates on earth, in particular with regard to one’s fellow creatures.

The ethical status of volunteerism is thus somewhat ambiguous; it was often referred to as a ‘duty’—indeed, some interlocutors considered it the single most important religious duty for Ismaili, given that many of the other forms of worship enjoined on Muslims are understood by this community in esoteric terms. For example, Nizari Ismaili seldom observe a literal fast during Ramadan, interpreting the inner (batini) meaning of the Qur’anic injunction to do so as a reminder to refrain from acting, thinking or speaking wrongly throughout the year. At the same time, as implied by the glossing of the term as ‘volunteerism’ in English, the non-obligatory nature of the practice was equally stressed. Not everyone volunteers, or volunteers equally; indeed, a person’s readiness to volunteer was a sign of character both looked for and nurtured by others. Finally, such service was, as described above, considered a natural impulse arising within the soul of a person who possesses spiritual wisdom (haqiqat) in response to the fact of God’s creation; as such, it was less a matter of deliberation than of disposition, for which an individual subject is partly, but not wholly, responsible.

The interchangeability of the two, seemingly antithetical terms, khidmat and volunteerism, in ordinary speech is not incidental; it is not a simple matter to disentangle the freedom that makes a person’s service ethically significant from the moral obligations that make it desirable. We might say that the self, upon recognizing certain profound truths, is freed from narrow self-interest (and, one might add, from doubt, decision-making, etc.) to act in accordance with these evident truths through devoted service to others. Human obligations to God are an unalterable truth, but whether or not to embrace that truth and respond to it with particular acts of service is, necessarily, a choice. At least for those who embraced this particular ethical project, the obligation to do so was not experienced as constraint, but as a particular, ethically significant kind of freedom. What, then, is the form that this freedom takes?

It is true that the kind of deliberative reflection with which Laidlaw is principally concerned has a place in the context of khidmat or volunteerism. As a public, formalized practice and a prescribed religious duty, volunteerism was not infrequently a matter of discussion and debate, including a rather formulaic debate that is apparently common to Ismaili discourse transnationally about whether it is better to perform one’s volunteering ‘badged’ (in uniform or, by extension, in a publicly visible and acknowledged way) or ‘unbadged’ (privately, without recognition for one’s actions). The terms of the argument were laid out most clearly in a conversation in an online message forum: to engage in volunteering through formal channels means that one’s efforts can be allocated sensibly and combined with the efforts of others, therefore doing more good, but at the risk of muddying one’s own motivation by potentially attracting the admiration and praise of others. Such reflection was not only a matter for abstract argumentation but could also take far more personal and deeply-felt forms. One man, Danish, a well-known and successful shopkeeper, narrated the history of his own attempts to wed the ethos of volunteering to that of entrepreneurship by creating a handicraft business that would provide employment opportunities, skills-training and a broader market to the benefit of many people besides himself; in the course of this story, he criticized those (implicitly including his younger self) who engaged in ‘useless volunteering’, indulging in the virtuous feeling of doing good without actually producing any benefit for anyone.

Danish’s account exemplifies precisely the kind of conscious, deliberative reflection that Laidlaw sees as the definitive feature of ethical life. For an ethnographer, such moments at which ethical reasoning emerges in explicit and self-conscious form are relatively easy to recognize and to capture. People in Hunza take pride in their abilities to reflect and reason, especially on matters spiritual, as well as to debate and argue (not unlike the Chitrali villagers described by Magnus Marsden (2005). Undoubtedly, reflection is one modality in which they engage in ethical life in and around practices like volunteerism, but it is only one mode among others.

Volunteering, and even the disposition to volunteer, are not talked about in Hunza principally as the result of a deliberative process, but as a kind of desire. That volunteering exercises a form of attraction was evident even in the story of Danish, above, when he cautioned against being the kind of person who wastes all their time ‘hanging around the jama’at khana’ engaging in ‘useless volunteering’. The habitual tendency toward volunteering is an expression of character and is viewed as such by both self and others, rather than an external requirement. The desire to engage in volunteering can be cultivated, principally in others, and especially in children. ⁶ As an expression of character, this inclination differentiates some people from others; it is sometimes said to be visible in a person’s face or general appearance, from which it ‘shines’, affecting those around them. Such a person takes active enjoyment in helping others; voluntary activities, as with less formal kinds of mutual help (discussed below) are often taken as opportunities for fun, joking and sociability. The spirit of care and service to others thus joins the ethical character of the individual to that of their social milieu by way of their interactions with other people who recognize, respond to and nurture it. Deliberation has a role to play in shaping the subject’s freedom, but it is one capacity among others to do so.

What we have sought to demonstrate is that the forms by which freedom is recognized and understood are more variable than the conception advanced by Laidlaw allows. This is so because what counts as freedom is influenced by the way that selves, constraints and outcomes are described and delineated. Thus, freedom and constraint are entangled: what we might think of as constraints on individuals’ freedoms can also be seen as shaping that freedom into forms that are ethically significant within particular contexts. What is needed for an anthropology of ethics are better tools for thinking about this entanglement.

Ethical freedoms and moral rules

Even where people’s ethical commitments are readily expressed in terms of rules, norms and duties, such strictures may be made relevant to actors and their actions in a variety of ways. Rules do not dictate what people do with them, as much work from sociolinguistics has demonstrated: contextual factors as well as a host of metapragmatic rules inflect the ways in which they can be brought to bear in particular situations. In other words, the fact that people may express their ethical attitudes at times in the form of moral rules does not dictate that their behaviour takes the form of rule-following (or of breaking or rejecting rules).

Drawing on a variety of linguistic materials, Webb Keane has made a similar point: ‘although people may …invoke social rules or norms,’ he writes, ‘the ways they do so are products of interaction as much as they are guides for it’ (Keane, 2016: 133). Between the relatively fixed and systematized codes of morality and the situationally responsive ebb and flow of ethical life there is a two-way relationship. ‘Morality systems draw on features of interaction. The ethics of interaction in turn responds to the precepts made available by morality systems’ (Keane, 2016: 134). He further clarifies that in his view ethics is ‘not just a matter of decision making’ nor is it confined to the internal deliberations of an individual ethical thinker (Keane, 2016: 134). Rather, ethical awareness emerges prototypically in the form of interaction. The stances people take in speech position them in various and shifting ways toward moral principles and precepts, often embodied in the presumed perspectives of others. This attention to the complexity of the connections between systems and actions in Keane’s formulation helps us see the mutual entanglement of freedom and norms. Within ethical frameworks we might call moral, the possibilities for action and thought are not exhausted by the binary of obedience or resistance to moral rules (Laidlaw, 2014a: 111). ⁷

A tension between freedom and rules appears in many well-known ethnographic cases (see Robbins, 2004). Laidlaw’s own fieldwork among Jains in India describes how an almost unimaginably strict moral project of ascetic renunciation is lived out within a community divided between dedicated renouncers and ordinary householders (Laidlaw, 1996, 2000). In his finely-observed ethnography, he shows how adherence to the impossibly demanding rules of Jain asceticism is maintained as a liveable ideal through the logic of karma and the mediation of reciprocal exchange managed to appear as non-reciprocity (Laidlaw, 2000). The ethical telos—renunciation of the world and all affective and social ties to it—emerges as a collective achievement of the ethically heterogeneous social world.

In other cases, people may look to rules as a way of shoring up the limitations of the self. As described by Amira Mittermaier in an article that aims to ‘trouble the link between compassion and voluntarism,’ middle-class Egyptians engaged in charitable volunteer work turn to notions of duty and responsibility as a way to circumvent the limitations of compassion as a driver of ethical action (Mittermaier, 2014: 518). While compassion and care for others are considered virtues by many of these volunteers, they also struggle with compassion fatigue and recognize that, within their moral framework, the pleasures of giving potentially undermine the altruism of their efforts. For these volunteers, ‘humans merely mediate divine justice’, and ultimately, human motivations like compassion are a less trustworthy guide to their desired ends than divine rules (Mittermaier, 2014: 517).

Emphasizing obligations to God over fallible and unevenly-elicited human motivations like compassion seems a way of shoring up the self’s orientation toward ethical ends against weakness from within. In Feinberg’s terms, constraints on the fulfilment of the self’s higher-order desires arising from less-valued aspects of the self are counteracted by the embrace of external rules. Rules may, then, not merely be followed or ignored, but be embraced in service of ethically-desired ends; constraints in the form of obligation or duty can be recruited to shore up internal incapacities and bolster the self’s efforts toward desired ends. When people enlist external constraints in order to selectively engage or enhance particular aspects or capacities of themselves, it opens up questions about the constitution of ethical selfhood (the ‘subject of virtue’) in ways that require an ethnographic inquiry into the composition of that self if we are to provide an ethnographically adequate account of any particular conception of freedom.

Of course, people not only recruit external constraints in service of ethical projects—they also find themselves recruited into relations with other selves, collectivities and powers. In recent works by Mittermaier (2010, 2012) and Taneja (2017) for example, dreams convey the tangible trace of another will that exercises a compelling—though not compulsory—moral force on the dreamer. Other studies find their subjects already embedded within such relations, as in Johann Rasanayagam’s account of Muslims in Post-Soviet Uzbekistan, for whom everyday forms of conviviality are a key site for the authentication of Islamic ideas and ritual practices (Rasanayagam, 2011). For Rasanayagam, sociality itself is a site of transcendence, in which people encounter and ground their moral frameworks in orders of existence that exceed their own immediate experience (Rasanayagam, 2011: 27). As in Keane’s discussion of the social contingency of rules, in this account, systems of moral rules and ideas derive their force and legitimacy from their rootedness in the ordinary ethics of routine interaction, not the other way around. Nor is conscious, deliberative reflection the primary driver of ethical action in this account; moral emotions and intuitions are at the core of an ethical sociality.

The interrelationship between formal obligations and systematic conceptions, on the one hand, and freedom, on the other, arises in Hunza not only in the context of the codified, institutionalized moral system of Ismaili Islam, but also in more ‘ordinary’ forms of ethical life. ⁸ This tension is also manifested in the conceptualization of social relationships in and around the household. In daily life, two fundamental forms of relatedness are frequently juxtaposed in the idealized form of a person’s close patrilineal kindred (sukúyo) and neighbours (hamsaya). Sukúyo are prototypical figures of obligation; regardless of how a person may feel about the individuals and households who make up their sukúyo, there are certain core duties toward them that must be observed, most importantly those surrounding death and the nested periods of mourning that follows (marked by the tisra [third day], satwa [seventh day] and chaliswa [fortieth day] observations). Sukúyo thus stand in for the rule-governed aspects of social behaviour. On the other side of the relational coin are neighbours and friends—people between whom proximity, unforced affection and a history of acts of mutual assistance have forged bonds based on inclination rather than rule.

It could be imagined on this basis that in Hunza kinship is a moral domain and neighbourliness/friendship is a terrain of ethics. But the relationship between obligation and freedom is again more complicated. Many ‘neighbourly’ relationships were in fact layered atop recognized kinship ties, but in situations where participants wished to highlight the easy and affectionate nature of the bond, their voluntary character was emphasized in their choice of language. Mutual help (amantsé) around everyday household and agricultural tasks formed the substance of these relations; although developed in the context of ongoing exchange and reciprocity, however, many interlocutors downplayed any sense of indebtedness and emphasized that such acts were a matter of choice, based on free and unconstrained affection. Thus the neighbour was a prototypical figure of freedom, representing the fruits of action based on inclination and choice rather than pre-existing bonds. Similarly, kinship could be used as an idiom to invoke the ties that bind, though people also at times acknowledged that even the closest of kin relations were, in the end, products of choice and of careful maintenance over time.

Obligation (in the form of formal kinship) and freedom (in the form of neighbourliness) thus represented the two poles of relational potential. While the two ideals formed a contrasting set, they were linked within local discourses of identity. The ideal lay not in a deliberative process of balancing the principles of obligation and volunteerism but in achieving an expansive and open-ended sphere of kinship-like relations of trust, cooperation and mutually enmeshing ties. This idealized view of Hunza society was contrasted in people’s minds with other regional models of sociality characterized locally as ‘tribal,’ in which overly-strict adherence to the rules of kinship kept people locked within their own lineages, permanently suspicious and unable to cooperate with outsiders for a greater good.

Such relations were endowed with spiritual as well as social value; the similarity between the ethics of kinship and of khidmat or volunteerism, were implicitly acknowledged, as in the above-mentioned discourse of identity. A productive tension between freedom and obligation is a defining feature of both modes of action. Both practices take place across an ethical terrain in which questions of obligation (to kin, to God, or to fellow humans in general) and freedom (the neighbour’s or the volunteer’s unforced choice to give) are constantly present to the participants and in which the interplay between the two constitutes the distinctive ethical character of the practice. In Hunza the tension between obligation and freedom does not so much present itself to people as a problem of competing and incompatible demands on decision and action as it appears jointly constitutive of the ethical domain itself.

Volunteerism, whether in the guise of freely-chosen bonds based on personal inclination and mutual assistance, or in the form of religiously-motivated acts of service, represents one local model—in the terms of Feinberg as we have been interpreting him, a particular conception—of ethical freedom. In this conception, we might say that the subject, endowed with wisdom/discernment (haqiqat) and love (mohabat), is free from the constraints of selfishness and closed-mindedness, to enter into valued forms of relationship with fellow humans (and thereby, with God). Both the constitution of the subject and the outcome are the same in the two contexts, while the constraints may be understood as primarily internal in relation to volunteerism (as selfishness) and as largely external in the domain of kinship (the closed and divided character of the social world constrains the formation of enriching relationships).

Such an understanding of freedom is not easy to account for in Laidlaw’s terms. For one thing, it is rooted in the social, emotional, and spiritual capacities of its subjects rather than, or in addition to, those of reason and reflection. Furthermore, it presumes a close relationship between the subject’s freedom to choose and the ties of obligation, duty, affection, and aspiration that make claims upon that freedom. On the one hand, this conception shapes and defines the options available to the ethical actor: she can perform her service or not, enthusiastically or not, but she cannot ignore it entirely. On the other, it presupposes a particular notion of the good—the telos of ethical action—configuring an array of powerful inducements, both external and internal to the subject, in favour of that telos. As a local model of freedom, volunteerism gets its specificity and its power from the way it configures capacities, opportunities and constraints; it has no ethical content without them. Finally, volunteerism as a conception of freedom is conceived in relation to a corresponding model of moral constraint—what we have been calling obligation—both of which are equally central to people’s conceptions of what it means to live a good life.

The dualism of human nature and its social conditions

The ethnographic cases we have been discussing suggest the limitations of a theory of ethics rooted in a conception of freedom in the guise of reflection, deliberation, decision and judgement, as opposed to one incorporating inspiration, intuition, imagination and other non-deliberative capacities. These capacities call into question the boundaries of the ethical subject and suggests a constitutive openness. Furthermore, these modes of ethical experience do not line up in a clear way on one side or the other of the binary of ethical freedom and moral constraint, in part because they take individual subjects out of themselves and place them in relation with other selves, other temporalities and other ethical projects. Encounters with extra-ordinary modes of experience, of course, lie at the heart of Durkheim’s theory of moral life, and as the essay from which we have drawn the title of this section makes plain, the tension between individual freedom and the compelling force of society is for him irresolvable: ‘how can we belong altogether to ourselves and altogether to others, or vice versa?’ (Durkheim, 2005: 37). Durkheim’s late reflection on the dualism of human nature invites us to envisage the interaction between the ‘first nature’ of innately given dispositions and capacities (which Durkheim called ‘organico-psychic’) and the ‘second nature’ of social and cultural life. To do so, in the light of current knowledge, requires us to focus on the multiple ways in which the constitution of selves and normative constraints impinge on one another.

Ethical life (whether in the form of conscious deliberation or in its various other modes) is lived by subjects in the presence of and in response to others, and an anthropology of ethics must be able to account for the persuasive, affective, and normative currents that pull at them with as much complexity as do their freedoms. In this respect we suggest, along with Joel Robbins (2016: 13), that one further way in which Durkheim’s work may have value for a contemporary anthropology of ethics lies in his having asked the question, not of how it is that society imposes its rules on us, but how it is that it comes to seem ‘desirable and desired,’–to the extent it does—that we should conform to them (Durkheim, 1974: 45).

Conclusion

We have sought to make two arguments. First, that, in proposing an ‘anthropology of ethics and freedom’ (Laidlaw, 2002) and arguing that reflective freedom is ‘intrinsic to the very idea of ethics’ (Laidlaw, 2014a: 149), Laidlaw offers us unduly restrictive notions of freedom, of ethics and, accordingly, of the scope of anthropology. We can agree that freedom thus conceived is a central aspect of ethical life but not that it is itself sufficient to define the ethical from an anthropological point of view. Secondly, we have argued that the push and pull of normativity, whether in the form of duties and rules or of reciprocal obligations and relational commitments, make up the other side of freedom and ethical life. The exercise of freedom cannot be empirically separated from its normative background. Liberty, as Durkheim remarked, is the fruit of regulation.