Governing Opacity: Regimes of Intention Management and Tools of Legibility

ABSTRACT

The intentions of others are ultimately opaque: we can never know exactly the mind of someone else. Yet humans continually attempt to ‘read’ the mental states of others and throughout history have created institutions that attempt to do so by managing intentions and thus addressing the opacity of other minds. The contributors of this special issue argue that the form in which we meet the fundamental challenge of the opacity of mind is decisive for the kinds of government we are able to imagine. Our introduction provides the framework for the exploration of the correlations between the management of opacity and the forms of government humans create. We draw attention to different ways of creating legibility, and corresponding practices of accountability, thus linking particular forms of intention management with particular ways of doing and imagining politics.

KEYWORDS:

• Accountability
• intentions
• opacity‌
• legibility
• politics

The intentions of others are fundamentally opaque: we can never exactly know the mind of someone else. At the same time, some basic capacity to ‘mind read’ – the inference of the intentions, beliefs, and mental states of others in relation to their behaviour – is a human universal. Taking advantage of this capacity, tools of communication and social institutions create environments in which intentions can be managed, that is, interpreted, revealed, and concealed. In this special issue we propose that the particular ways in which humans meet the challenge of the opacity of mind are of decisive importance for the kinds of government they are able to imagine and establish.

Anthropologists have drawn attention to the different ways in which intentions are addressed in human sociality. In an influential edition of the Anthropological Quarterly, contributors argue that in many Melanesian societies the intentions of others are not addressed explicitly, or at least, not spoken about publicly and directly: Robbins and Rumsey (2008) point out that in these societies, the ‘opacity of mind’ has become a doctrine and public mind reading is disavowed. We follow the general recommendation of this earlier volume, to look for the correlates of the prohibition (and respectively, the encouragement) of public mind reading, and in particular the suggestion of Rupert Stasch (2008), who establishes a connection between the disavowal of public mind reading and questions of political coexistence. This is, in fact, the question that unifies the contributions to this special issue: What are the correspondences between intention management and political regimes?

Such correspondences might be conceived as ‘cultural ideologies of intention-guessing’ (Danziger & Rumsey 2013), or local variations of theory of mind, in the sense for instance that minds can be understood as more or less ‘porous’ (Luhrmann et al. 2011, Luhrmann 2020). Rather than cultural differences, in the following we try to identify the social conditions under which public discussion of intentions is discouraged and opacity of mind is cultivated, as well as the social conditions that favour public discussion of intentions, and thus relative transparency of intentions.

To understand the extent to which opacity and transparency vary in relation to political regimes, we suggest legibility as a useful concept for bridging knowledge of mind and statecraft. We argue that in political regimes where opacity reigns, legibility may not only look very different from those in which transparency prevails, but also some forms of state legibility might make it more likely to transform opacity in the direction of transparency.

By legibility we mean the general condition of making something legible, that is, decipherable, distinctive, and clearer to the senses, and thus susceptible to interpretation. Fundamentally, legibility always involves a relation between observer/observed or person/surrounding.¹ Legibility is the degree of distinctiveness that enables the viewer to categorise the surroundings. Legibility, per se, does not concern intentions, but rather provides a filter between people and contexts. Yet the kind of categorisations available for particular objects and populations – that is, their relative legibility – may make the need to deal with intentions directly more or less relevant. Hence our proposal that public mind reading, or the institutions of intention management, correlate in interesting ways with the tools of legibility.

The most common use of legibility in anthropology derives from James Scott’s (1998) use of ‘state legibility’ to refer to mechanisms of visibilisation and simplification that maximise the reach of the state to populations. He explains how he came across the idea while researching efforts by nation-states to settle or sedentarise nomadic peoples, including hunter-gatherers, shifting cultivators, and itinerant traders:

The more I examined these efforts at sedentarization, the more I came to see them as the state’s attempt to make society legible, to arrange the population in ways that simplified the classic state functions of taxation, conscription, and prevention of rebellion. (1998: 2)

For Scott, legibility is therefore a central problem in statecraft. In Seeing like a State (1998), Scott dealt with a centuries-long process by which states reorganised the societies they governed to make them legible to the apparatus of governance. Yet the creation of state legibility, at least in principle, is not aimed at understanding and changing the complex behaviour of the entities to be governed – these only needed to be superficially ‘legible’ from a central vantage point. Whether governance is driven by predatory aims or by the desire to improve the lot of the people, its goal is to simplify some features of the population in relation to a centralised and narrow viewpoint for the purposes of standardisation, measurement, and control. Scott’s first example for state legibility is scientific forestry in nineteenth century Prussia (1998, ch.1). The modern Prussian state needed to maximise tax revenues from forestry and for this purpose the acreage, yield, and market value of a forest had to be measured. So only these obviously relevant variables were included in the statist ‘mental’ model of the forest. Since wild and unruly forests were illegible to the surveyor’s eyes, scientific forestry transformed this diversity of wild species growing randomly into uniform rows of the highest-yielding varieties. Thus, the word legibility is not only a metaphor; it produces real effects on the world. Just like the promotion of literacy can have manifold social consequences that start from the transformation of spoken language into visible and readable words, the spread of state legibility potentially shapes the world into the mental image of efficiency and bureaucratic control held by the government’s centre.

In this special issue, however, we take state legibility, as outlined by Scott, to be only one – and perhaps the most extreme – form of making the world legible. The concept of ‘state legibility’ does not have to imply an essentialisation of the ‘state’. Anthropologists have long noted that states should be studied as state processes and state practices, or, in brief, by the ‘effects’ they achieve, including the isolation of subjects and the spatialisation of boundaries (Trouillot 2001). Such ‘state effects’ result in simplified fictions of social reality, and the work of simplification and fictionalisation is not carried out by a unitary, all-powerful entity acting under a coherent logic of rule; rather, it is executed by a range of organisations that take a variety of forms. This multiplicity of governance can nevertheless produce real state-like effects, so that the state resurfaces ‘as a powerful object of encounter even when it cannot be located’ (Aretxaga 2003: 399). Yet, on the recipient end of legibility, subjects may experience the state in terms of illegibility and uncertainty (Das 2004), or suspect powerful conspiracies (Sanders & West 2003). Similar insights can be garnered from the rise of transparency ideologies in contemporary politics, as a precondition to establish the sort of accountability that regulates functioning democracies in liberal states. Institutions built on the promise of transparency do not eliminate opacity so much as shift the boundaries that define, on the one hand, knowledge that must be revealed in public, and on the other, knowledge that should be concealed and kept private. Thus, together with the extension of transparency we often see a rise in the active promotion of opacity via practices of secrecy and notions of privacy; in contemporary digital publics, for instance, a new ‘right to opacity’ has been opposed to the promotion of transparency by governments and corporations (Birchall 2021).

To these important insights, we add that the will to legibility need not pursue statecraft at all. Not every political regime aspires to ‘see like a state’, that is, to simplify the contours of populations so as to make them visible from one central vantage point. Even in environments where state legibility is conspicuously absent, such as in so-called ‘egalitarian societies’,² we find numerous ways of creating social legibility. We thus expand the use of legibility to political forms that have other means and goals than those of states, and that may cultivate opacity with the same intensity with which contemporary liberal institutions cultivate transparency. Or, conversely, that might converge on the pursuit of transparency via different ideological avenues and historically specific projects of legibility. We further aim to understand how these forms of legibility – that is, deciphering and categorising reality – relate to the institutions of intention management that different societies have in place.

Expanding the scope of politics beyond the state, we also note that government, within or without ‘states’, cannot be limited to the interaction between humans, but includes the relative opacity, as well as the capacities to make others legible, of all kinds of beings – including metahumans, spirits, species-owners, and others. Marshall Sahlins (2017) has recently argued that even societies he terms ‘egalitarian’³ possess a ‘state’ and a ‘rule of law’ in the government of metahumans. More powerful than ordinary humans, such beings govern life and death, rain and sun, fortune and disaster. Sahlins points out that ‘something like the political state is the condition of humanity in the state of nature; there are kingly beings in heaven even where there are no chiefs on earth’ (Sahlins 2017: 91, emphasis added). Yet the nature of this likeness – ‘something like’ – is debatable. What kind of state is the rule of metahumans, the state of what Sahlins calls ‘the original political society’?⁴ Sahlins invites us to overcome the Cartesian separations structuring our thought, and specifically the assumption that only human subjects can create ‘state’ and ‘law’ to dominate the world. In the same breath, Sahlins also throws much received wisdom overboard – including some of the conventional anthropological understandings of species-owners, masters, and spirits in so-called ‘non-state societies’ – by arguing that their power did not amount to coercion, and that they were never radically separated from the human realm. Altogether, the argument is therefore a reductio ad absurdum of the concept of ‘the state’: if the state is everywhere, then it is nowhere.

Inspired by the decentring of humanity in Sahlins’ cosmopolitics, we note that the inevitability of the weather, the stars, birth, and death, which he describes as the state and government of metahumans, might be re-termed simply as the opacity of life. Yet we also note that in his outline of the ‘original political society’, Sahlins never considers the problems of the opacity of intentions and of state legibility. If we examine his argument from the perspective of opacity and legibility, it appears that in many of the non-state societies he includes in his argument, so-called ‘opacity of mind doctrines’ are common, and little to nothing resembling ‘state legibility’ can be found. While it is possible to find versions of opacity doctrines in hierarchical and more centralised societies, as we elaborate in the next section, it appears that opacity is frequently correlated with a preoccupation to reduce public accountability, whether in regard to one’s own intentions or those of others.⁵ As Duranti puts it:

from a sociocultural point of view, the phenomenon of opacity doctrine might be seen as a defence strategy against the accountability that comes with making claims about what others think or want … It could turn out that the opacity doctrine hides or at least implies a pan-human preoccupation with reducing one’s accountability. The study of the conditions that might increase or decrease such a concern is a worthwhile research project. (2015: 186)

We follow Duranti’s invitation and suggest that a critical variable increasing this concern of reducing one’s accountability is which tools of legibility are at hand. Tools of legibility do not by necessity increase individual accountability, but they facilitate the attribution of responsibility, that is, the causal and moral association of an agent or a chain of agents with an act. On this basis, the same tools afford the ascription of accountability, that is, the obligation of an agent to respond to the attribution of responsibility.⁶ Tools of legibility, therefore, enable the attribution of responsibility and the demand for accountability, and they make both more likely to occur. They do so in particular when legibility takes the form of ‘state legibility’, that is, the kind of legibility in which a large population can be seen from one central vantage point. Similar to other forms of legibility, state legibility in and of itself does not lead to accountability, but projects of state legibility generally expedite the public attribution of responsibility and accountability. As such, they tend to foreground a concern with the transparency of intentions, and possibly with public mind reading.

Decentralised societies typically do not possess ‘state legibility’, that is, the capacity to simplify populations to make them ‘legible’ and controllable, which is a core requirement of statecraft. In such societies no one – not even metahumans – can make everyone else legible from one privileged vantage point. In these situations, there is no ‘public benchmark’ of accountability, no public apportioning of responsibility and univocal enforcing of collective judgements, and therefore we expect individuals to decrease their accountability in everyday interaction. This might happen by concealment, secrecy and suspicion, the avoidance of explicit mind reading, and possibly, in the most extreme form, by general opacity doctrines.

Intrigued by Sahlins’ suggestion that ‘egalitarian societies’ always had a state in the government of metahumans, we will consider below whether the ancestors, spirits, and species-owners of decentralised societies possess the capacity to make populations legible in similar ways. Here we merely note the absence of state legibility and the avoidance of public mind reading in the social arrangements that provide the case studies for Sahlins’ ‘original political society’. Acknowledging this coincidence of circumstances, we can begin to examine the correlations between intention management, tools of legibility, and forms of government in different geographical and historical contexts.

Between the cultivation of opacity in decentralised societies and public mind reading in centralised government, there are countless forms and gradations of managing the intentions of others. One core variable is whether projects of ‘state legibility’ are present, that is, particular means that make it possible to simplify, to read, and thus to manipulate populations. While state legibility per se does not imply the increase of public mind reading and accountability, we propose that the expansion of state legibility has manifold consequences for the ways in which other people’s minds and intentions are addressed, identified, and verbalised.

On the other hand, even when there is little state legibility and public mind reading, there are tools that create social legibility and institutions of intention management. These tools and institutions influence the extent to which people conceal or reveal intentions in public, the ways in which they ascribe responsibility and demand accountability, and the possibilities for extending control or claiming autonomy. Contributors to this issue draw connections between different forms of government, tools of legibility, and regimes of intentions – that is, the institutions through which people establish accountability, manage the expression of intentions, and create legibility in different settings. Re-locating the conditions of politics in the opacity of life and mind, this special issue formulates an anthropological theory of politics, which reconsiders timeworn concepts such as power, state, and autonomy.

To prepare the ground for the contributions of this special issue, in this introduction we outline a framework that connects the challenge of the opacity of mind with the questions of legibility and accountability, so as to be able to understand the correlations between intention management and forms of government. We do so in three steps: First, we trace how the opacity of mind has been dealt with in recent anthropological literature, paying attention specifically to the question of when and how intentions have to be made explicit. We then expound our proposal that while the capacity for mind reading is universal, it is put to use in different ways, core variables being the tools of legibility and institutions of intention management that exist in any given situation. Finally, we explain why projects of state legibility are particularly important in this regard, even though their effect on the politics of intentions is neither unidirectional nor irreversible: it is possible to identify correlations between specific forms of state legibility and the tendency to address intentions in public, but the centralisation of perspectives that comes with state legibility also creates new possibilities for opacity. We thus identify a series of correspondences between specific forms of managing the opacity of intentions and particular ways of practicing and imagining politics, both human and cosmic. Governing opacity, therefore, refers not only to the politics of intention and legibility, but is a sine qua non condition for politics altogether.

Opacity of Mind

A number of developmental psychologists have argued that ‘theory of mind’ is universal to human beings, and fundamental to social interaction and language (Wellman 1992; Wellman et al. 2001; Astington 2006). From a young age, children begin to grasp that the mental states of others are causally independent of their own and of the state of the world. That is, they understand that mental states are crucial to understand people’s behaviours as well as the meanings and consequences of actions: for example, whether acts are intentional or accidental. This ability to attribute mental states to oneself and to others is called ‘theory of mind’; it is a theory because it refers to phenomena that are not directly observable and because it is used to make predictions, for example to explain and predict the behaviours of others.

Theory of mind thus refers directly to the intentions of others. In ordinary language, intention refers to the conscious or unconscious goals of human action. Philosophers of language have typically used the notion of intention to refer to how people go about making sense of what others say or do.⁷ In discussions of theory of mind and the attribution of intentions, it is important to distinguish between two levels: on the one hand, the level of explicit, conscious reflection about mental states, and on the other, the automatic, implicit, and largely non-conscious level of mind reading (Astuti 2012, 2015). Implicit mind reading does not require verbalisation or explicit reflection, it happens anytime we monitor our interlocutors’ attention during a conversation or predict where they will go next by following their gaze. Such implicit mind reading is fundamental for human action and happens almost all the time – but the resulting implicit understandings and interactions are not our main preoccupation in this special issue. Our reflections focus instead on how speculations about other minds are made explicit, and how differences in explicit mind reading impact on politics. When we refer to 'opacity of mind', we mean the indifference or otherwise avoidance and reluctance to engage in explicit speculation about the intentions of others.

Western philosophical reflection on the nature of meaning and action has tended to place great emphasis on intentionality, while assuming this to be a property of the individual actor, to be found within the self. This emphasis on intentionality leads to a philosophical impasse – ultimately everything that can be said about others’ minds is based on inference, and therefore a basic ‘opacity of other minds’ can’t be overcome.⁸ Since people are neither omniscient nor telepathic, they rely on all forms of mediation to infer mental states. Linguistic anthropologists typically refer to this in terms of semiotic mediation, in which language plays a central but not exclusive role; of importance is also pragmatics, that is, how language is used, since this too influences the intentions people impute to each other. What’s more, people may draw inferences about each other’s intentions from purposeful as well as unplanned signs: they may be drawing on conventional and more or less explicit codes but also from non-verbal, unconventional, and implicit cues that interactants only barely control such as bodily posture, smells, the intensity of light at a particular time of the day, etc. Noting the hyper-mediation of signs in humans’ intersubjectivity, Keane (2016: 83–85) points out that humans are not so much mind readers as sign-readers since anything from words, acts, events may be culturally construed as signs of intentions: it is impossible not to communicate.⁹

So far, intentions appear unproblematic: both the human capacities (a propensity to detect and attribute intentions) and potential limitations (the inferential and mediated character of intention-reading behaviour) of theory of mind can be considered universal. What is not universal, however, are the usages and assumptions people around the world make about intentionality. More specifically, the extent to which intentions are considered relevant, open, or disclosable; transparent or susceptible to revelation and accountability; and imputable in public, differs across societies and even across contexts of interactions. Anthropologists have tended to be sceptical of the idea that individual intentionality plays the same role everywhere (Keane 2016: 118). Robbins and Rumsey (2008) show for example that many Pacific peoples appear disinclined to mind read and routinely disavow knowledge of the mental states of others as ways to interpret behaviour and attribute responsibility for actions. While most recent literature has focused on Melanesia and the South Pacific (e.g. Rosaldo 1982; Duranti 1992; Robbins & Rumsey 2008; Throop 2008; Mayer & Riese 2013), there is some evidence that a similar phenomenon is discerniblec elsewhere, including North Africa (Carey 2017), across the Americas (Danziger & Rumsey 2013), and in particularly heightened form in the Maya region (Danziger 2006, 2010; Groark 2008), and in Lowland South America (Conklin & Morgan 1996: 682; Boster et al. 2003: 476–477, Johnson 2003: 125; Londoño Sulkin 2012: 63, Course 2013: 308; Walker 2013: 122; Brown 2014: 187; Ewart 2015: 209).

Linguistically-oriented anthropologists have tended to associate opacity with a local linguistic ideology whereby an utterer’s intentions are considered irrelevant to the utterance’s interpretation. This contrasts with the modern Euro-American ideology of language in which intentionality is considered central to linguistic production and interpretation (Duranti 2015). Yet other anthropologists have also shown that, in some places, opacity of mind makes people more rather than less attentive to others’ intentions (e.g. Groark 2008; Stasch 2008; Throop 2008; Robbins 2011).

A fundamental question in this regard is whether opacity doctrines override the universal human capacity to read other minds. Even when following opacity doctrines, do people still contemplate mental states despite their proclamations to the contrary? The diversity in which languages encode notions of intentionality, responsibility, and control, suggests that it would be misguided to delineate a strong contrast between indifference and interest in mind reading (Goldman 1993; Groark 2013; see also reappraisal by Duranti 2015: 2–15). Even where people seem to explicitly reject engaging in mind reading, there are cases of spontaneous interactions, e.g. gossip, accusations, anticipations, or ordinary interpretations of behaviour (cf. Besnier 1993; Duranti 2008: 486–489) in which most social actors, and perhaps especially children (Schieffelin 2008) ordinarily impute intentions and desires to others. For these and other reasons, some authors have argued that opacity claims are less about intentions than about ‘talk about intentions’ and should be taken as ‘metalinguistic and metapragmatic claims about the relations between public evidence and private states’ (Keane 2008: 474)¹⁰ – that is, claims about how acts of revealing and concealing are or are not to be taken as evidence for private states. The question is not what can be known about others intentions but rather what can be said about others’ intentions – or, in other words, ‘Knowing Minds Is a Matter of Authority’. This is the title of the seminal essay by Rupert Stasch (2008), already mentioned above, who observes that for the Korowai of West Papua ‘reflexive models of the possibilities and problems of knowing other minds are also models of the political terms of people’s coexistence’ (2008: 443).

Thus, an important question is the degree to which a cultural model of the opacity of mind affects local attitudes toward authority and accountability, a relevant consideration for the thesis we advance in this issue. Stasch (2008) argues that for the Korowai, political autonomy and opacity of mind are associated: people avoid talking about the intentions of others so as not to interfere with their autonomy and self-determination.¹¹ Political autonomy is often associated with the freedom to move and the absence of permanent settlements. The Korowai, for example, historically lived in dispersed tree houses; individuals were highly mobile and living apart was part of a larger pattern of rejecting relations of authority. But very different kinds of societies can endorse mind opacity: for instance, versions of the same ‘doctrine’ can be found in highly stratified societies such as Samoa (see Duranti 1992, 2008), where people identify with larger social units and see their lives as regulated by social obligations and ancestral rules. In such cases, therefore, the same disavowal of intentions may relate to a preference for institutional and social roles over individual psychologies and personal needs (Duranti 2015: 241). If, as we stressed above, opacity may have an elective affinity with the obfuscation or downplaying of individual accountability, the same cannot be argued for collective accountability, which may be intimately connected with the cultivation of opacity (see, e.g., Danziger 2013). Samoans, for instance, are said to be ‘more eager to act upon conventions, consequences, and cultural expectations about what a certain type of person (e.g. with a particular status or role) would or should do in a given context than they are to rely upon explicit reconstructions of what an individual’s intentions or unexpressed goals might have been’ (Duranti 2015: 67). While for the Korowai mind opacity relates to a cultural respect for the autonomy of others, among the Samoans it expresses a preference for standardisation in terms of social role (see also Bovensiepen, this volume). Another way of putting this is to say that in some situations, opacity doctrines buttress the performance of independence, whereas in others, they support the performance of interdependence and/or group homogeneity (Duranti 2015: 241). In the first case, people avoid assertions regarding others’ intentions for fear of speaking ‘on their behalf’ (Stasch 2008: 449), which would imply taking responsibility for others’ thoughts. In the latter, people resort to social conventions, including stereotype and prejudice, and thereby avoid the unpredictable consequences of individual action and thought.¹² In both cases, however, the avoidance of intentional language appears to reflect a certain anxiety over the shape of interiority, and becoming accountable for it.

In line with this observation, a number of authors have related the degree to which interiority matters to local attitudes toward authority and accountability, concluding that the use of intentional language often reflects a form of social control over interiority. Keane (2008: 480) for instance suggests that ‘opacity claims are not necessarily epistemological claims, but are claims about the sources of action, autonomy, possibly even freedom, and the expression of a great deal of anxiety about the role of others in my life’.¹³ Rather than about the nature of intentions themselves, opacity claims therefore speak of the risks and dangers that other people’s thoughts and emotions might pose (see also Rosen 1995; Groark 2008; and Throop 2008). As Duranti (2015: 242) puts it:

the very ability to imagine or predict what others perceive, wish, or fear creates the conditions for others’ social and emotional vulnerability, just as the transparency of our own intentions, motives, and wishes creates the conditions for our own vulnerability.

Luhrmann (2020) further proposes that one important dimension along which explicit reflection about theory of mind might vary cross-culturally is the extent to which minds are regarded as ‘bounded’ or ‘porous’.¹⁴ In the collection of short essays that follows her proposal (Luhrmann et al. 2011), we find a couple of examples of porousness coming from Melanesia (Stasch 2011) and Amazonia (Vilaça 2011). It appears that while in Melanesia, and in some parts of Amazonia (see above, and Buitron in this volume), the intentions of others are considered ultimately unknowable or undisclosable, selves are thought of as porous and susceptible to external attack or capture.¹⁵ These premises sit in interesting tension with one another, as they give shape to a strong form of individual autonomy that is nevertheless hyper-relational (that is, selves are conceived of as highly permeable to others). This likely inflects the patterns of responsibility found in each region: while autonomy and freedom in action are highly valued, attributions of responsibility typically draw attention to a chain of intentional agents that extend, blur and distribute the locus of agency and authority beyond the individual – as we would expect in situations where individual accountability is reduced. Perhaps for the same reason, the act of making public one’s speculations about others’ mental activity makes many speakers in Melanesia and Amazonia particularly anxious about external control, and therefore they refrain from presuming to know other minds or making public accusations in specific speech contexts.

To sum up, the disavowal of intentional discourse appears closely linked with an interest in reducing or distributing accountability (Duranti 2008: 493, 2015: 186; see also Danziger 2013: 260). If the prevalence of opacity doctrines is based on a certain anxiety with or even resistance towards accountability, what factors might increase or decrease this concern? In our view, projects of state legibility very likely play a crucial role. But before we deal with state legibility specifically, we will consider a range of tools of social legibility and institutions of intention management. In the discussion that follows, we pay special attention to the roles that metahuman agents play in governing opacity.

Tools of Legibility and Institutions of Intention Management

The classic Durkheimian error of twentieth-century anthropology was to reduce the agency of non-humans and metahumans to representations of the social interactions between humans. The counter-reaction of some branches of ontological anthropology is to see agency everywhere, and to imagine radically different planes of being that remain socially and politically interconnected. Sahlins’ recent essay on the ‘original political society’ (2017) is part of this tendency, but rather than a levelled ‘flat ontology’ in which humans are simply the same kind of actors as spirits or stones, the world that Sahlins writes into existence is a mirror image of Cartesian naturalism. In this world, the conditions of agency are set by metahumans, and thus human society is de-centred. Yet, as we have already noted, Sahlins never addresses the questions of opacity and legibility. Seen from this perspective, there seem to be indeed some remarkable differences between various forms of human government, as well as between different types of metahuman government. The Jade Emperor of China as well as the Catholic God of rural Spain, for instance, clearly possess the means to make populations legible in ways that are very similar to this-worldly bureaucracies and empires (see for instance Christian 1989; Feuchtwang 2001). But what about the meta-human governors of societies in which human-to-human relations are generally non-hierarchical? In the absence of agriculture, sedentism, and writing, it might be difficult even for metahumans to know where particular humans, plants, and animals are at a given point, and how many of them are together in a particular place. Even though perfect surveillance is impossible even for modern states, they at least have wholly different and powerful tools at hand to read (in simplified forms, as noted above) entire populations, and thus to establish optimal control. Hence, what distinguishes metahuman government from what we commonly understand as the state, is that the former does not aim to create a unified centre from which the different points of view of its subjects appear comparable and manipulable, let alone imputable.

This unification in the form of the state, by necessity, is based on (a) a privileged vantage point of view from where to see and read others; (b) technologies that make it possible, on the one hand, to simplify social reality, such as mapping or the creation of a census, while on the other, to discriminate individual differences to apportion judgement and responsibility; and (c) social institutions that allow for such vantage points and technologies. In the absence of state legibility, however, there are numerous other tools of social legibility, as well as institutions of intention management, that are crucial when dealing with the opacity of other people’s minds, as well as the opacity of the forces of life or metahuman government.

Now, even though metahumans do not have the tools of state legibility at hand, they still play crucial roles in the government of opacity. As several of our contributors demonstrate, in situations when the opacity of mind pervades interactions between ordinary humans, the same opacity is frequently extended to interactions between humans and metahumans, even if a fundamental asymmetry between humans and metahumans is recognised. To some extent this is to be expected, given that humans recruit the same cognitive mechanisms to interact with metahumans as they do with humans.¹⁶ Conversely, the fundamental opacity of the forces of life (which govern the weather, the seasons, birth, and death) is precisely how metahuman agency manifests itself in human affairs. The ways in which humans construe relations with metahuman agency set the conditions for opacity among humans themselves, by enabling the constitution of particular kinds of persons and group relations.

Among the Shuar of the Ecuadorian Amazon, for instance, the local framework to deal with the opacity of intentions is inseparable from the limit case of arutam vision-encounters that allow eminent men to see clearly. In her contribution to this special issue, Buitron outlines a ‘rule of self’ built on the relations with arutam, the spirits of deceased eminent elders. Shuar seek to incorporate arutam spirits in visionary quests that imbue them with strength, a sense of clarity and forcefulness of heart and speech, as well as with an intensified sense of hostility towards enemies. By internalising the point of view of a hostile and fierce spirit, the vision-seeker can now clearly see others through the point of view of an enemy, that is, as prey. Yet the content of the vision remains shrouded in secrecy, and the vision-seeker’s own heart (the seat of perception and intentionality) becomes veiled to the gaze of others: while the vision-seeker gains clarity and self-awareness, he simultaneously becomes opaque to others.

In Timor-Leste, we find another example of how the clarity and truth associated with metahuman agency transforms into opacity in everyday human dealings. In her article, Bovensiepen describes how individuals and groups (called ‘houses’) assert authority by delegating intention to the ancestors, which include deceased humans, totemic animals, and sites in the landscape. Ritual speakers typically invoke such ancestral beings in origin accounts to legitimise and contest the asymmetric political relations between emplaced house groups. Origin accounts are considered to be ‘true words’ handed down directly from the ancestors, but by selectively revealing or obscuring them in speech performances, individual ritual speakers can adapt them flexibly to new political or ritual situations.

In both cases, the powers of seeing attributed to metahumans, and conversely, the opacity reigning among ordinary humans, gives rise to highly dynamic political systems: in Timor-Leste, the ability to keep something hidden, or to partially reveal it at crucial moments, is a source of power and social flexibility; in Amazonia, the ability to become opaque to others becomes the engine of real existing egalitarianism.

The forces of nature, including the processes of life, the inevitability of death, and the weather, are not liable to full human control and remain ultimately uncertain, even if humans strive to manipulate them, as in the last two cases, or simply avoid them, as is common in Melanesia and some parts of the Kalahari (see contributions by Stasch, Laws, and Widlok in this volume). Whether metahumans have to be avoided, confronted, or manipulated, the relations between humans and metahumans in non-centralised societies are never set once and for all: in some cases, like in Amazonia, beings are constantly slipping in and out of relations, appropriating and being appropriated within new bonds, as when one loses an arutam and must acquire a new one; while in others, such as in the Kalahari or Timor-Leste, metahumans provide various possibilities for political action, depending on how they are mobilised and interpreted at pivotal moments.

Crucially, therefore, we argue that metahumans do not constitute a stable centre or unified vantage point from which to watch over and control all the creatures of the world or even a select sample: in Amazonia, not even species masters or owners have total control over their creatures (Fausto 2008). An important source of evidence in this regard comes from the role of mediators between the world of humans and the world of metahumans.

In many cases, cosmic mediators are endowed with extraordinary skills to see or act on the world, thereby affording powers of legibility over the human world. Yet as our contributions show, such cosmic mediators (typically shamans and sorcerers) can only establish partial legibility. When they appropriate metahuman agency, they may acquire powers of visibility but never omniscience; similarly, the judiciary powers and the rules they create do not amount to absolutist or general moral codes. Instead, such metahuman powers generally remain partial, temporary, and unstable, as no one can harness enough resources to predict and control the intentions of everyone at any one point. As Laws’ study of metahuman agency in the Kalahari reveals, ancestors and shamans are crucial for the government of opacity among Jú|’hoàn speakers living in the Kalahari: only they can know the intentions of others, specifically their kin. Ancestors observe the intentional states of their living relatives, and shamans mediate between the living and the dead to bring these states into the open, thereby governing the opacity that living relatives fear controlling. Rather than a general condition of transparency, however, such knowledge is apparent only in specific situations of illness and crisis, when ancestors and shamans intervene. The ancestors are thus never capricious: they strike only their own relatives with sickness, and only when they see the pain caused by suspicions, doubt, and resentment among relatives. Similarly, shamans who become temporarily all-seeing, all-hearing, and all-powerful by acquiring the capacities of ancestors use those capacities only to reveal the specific circumstances that motivate the ancestors to make their relatives ill in the first place. For all the extraordinariness of these acts, neither ancestors nor shamans care to make populations legible in their entirety; instead they are bound by specific bonds and capacities, intervening only when specific kin experience emotional pain.

Tools of social legibility in non-centralised societies are primarily practices of deixis that are not fixed to permanent symbolic ‘signposts’. Building on the distinction between deictic and symbolic fields, Thomas Widlok discusses the importance of context in relationship to the fundamental act of human pointing and emphasises how it is different from the fixed ‘signposts’ that are crucial to centralisation and state formation. While signposts may be expected to be either there or not, human pointing is subject to much more nuance and debate as it depends on the positionality of interactants. The deictic field in non-centralised societies relies on forms of co-presence which are never fixed into standard representations that transcend contextual interactions. As long as humans do not care to establish permanent signposts, human practice – which includes political practice – takes place in particular deictic fields in which participants differ in terms of perceptibility, proximity, memory, anticipation, prior talk and knowledge of each other, and access to each other’s intentions. Whilst opacity can be reduced over time, it never disappears completely as no situation is ever entirely predictable.

Together these contributions show that there are ways of creating common ground, comparing perspectives, and holding others accountable that do not lead to transparency, univocal control, and centralisation. Visionary quests, shamanism, ritual discourse, and everyday deixis can indicate what others feel and think, and thus make others ‘readable’, in locally situated and relationally limited ways. Marked by constitutive ambiguity and dynamism, these forms of social legibility subvert the verticalisation of relationships necessary to attain the degree of centralisation characteristic of state legibility. What then is the impact of projects of state legibility on the politics of intentions?

State Legibility and the Correlates of Governing Opacity

Since Seeing like a State (1998), James Scott has argued that states need to make populations legible so as to be able to achieve the fundamental objectives of statecraft, which are taxation, conscription, and pacification. He describes and analyses a number of means that radically simplify and map social reality to create state legibility. We have mentioned already scientific forestry, and other examples are permanent family surnames, urban planning, collectivist agriculture, and sedentarisation. In his most recent book, Against the Grain (2017), Scott extends his study of state legibility to the first states in the historical record, specifically in Mesopotamia. He demonstrates that these early states relied on state legibility just like modern states did, even though they used different means to create it: aside from sedentism and writing, it was in particular the cultivation of millets and cereal grains that served this purpose. These particular grains are relatively visible and easy to quantify, harvest, and store, and for the same reasons can be managed relatively easily by the authorities. Tubers are the opposite: they are less visible and quantifiable, and more difficult to harvest – and hence these were generally the crops of choice for the ‘escape populations’ outside the reach of the states (Scott 2009).

In Scott’s argument, state officials tasked with creating and maintaining state legibility generally tend to neglect the practical knowledge (what he calls the ‘metis’) of the commoners. For the same reason, many state projects that over-achieve ultimately fail. Yet, what he does not consider directly is how, exactly, both the knowledge of state officials and of ordinary people changes in the process. States do not so much know people, as utilise tools of legibility to make them equivalent and comparable as aggregates, while at the same time individualising them. Such tools can be quite basic – for instance, sedentarism, writing, and agriculture – or more intrusive ways of inscription, e.g. forcing people into grid structures. Modern states go beyond the basic tools that Scott identified, and establish normalising discourses to govern populations by problematising individual behaviour. In Foucault’s classic outline (1991), ‘governmentality’ is not simply rule by restriction and suppression, but rather a positive rule by proxy, a form of power that acts through the autonomy of the individual, by shaping hopes and desires of subjects: for example, an individual willing to confess his sins, a pupil eager to work, a patient wishing to recover, etc. ‘Bio-power’ and ‘governmentality’ are theoretical constructs premised on the statistical and scientific knowledge of populations, including the bodies, minds, and potential intentions of subjects (note the importance of ‘technologies of self’ and the normalising force of ‘types of knowledge’). They rely, therefore, on very advanced forms of population knowledge and the public availability of minds. Our focus in this special issue is not on these advanced forms of governmentality (which presuppose mass literacy and normalising science), but instead on the most fundamental forms of ‘bio-power’ and how they rely on tools of legibility.

We make the following proposition: with the increase of methods for creating state legibility in a given population, the question of how to make other minds legible and publicly available for inspection will become increasingly prominent in the same population. Indeed, the means to create state legibility that James Scott has described are fundamentally tools only for the management of information about people, and not tools to make intentions and private states legible or publicly available. But it is not a coincidence that the technologies that increase state legibility (e.g. census, register, bureaucracy) also typically lead to a systematisation and public availability of private mental states (which are then fully exploited in the normalising institutions of the modern state): lists, for instance, provide a tool to hold people accountable and thus facilitate a cross-examination of their actions, intentions, and minds (see Steinmüller, this volume).

So far, we have shown that the ability to read other minds is universal but knowledge about mental states is often publicly de-emphasised in non-hierarchical and decentralised societies. In such societies, in fact, an opacity of mind prevails; as we have shown above, not even metahumans (or those who master relations with them) can make others fully legible or control their intentions. Even when there are asymmetric relations between metahumans and humans, these do not translate into a unification of perspectives; a unification and centralisation of perspectives, however, would be necessary to create state legibility. We now turn to the correlations between state legibility and the public availability of intentions.

The creation of state legibility is a mutual project enacted both by those who govern and by those who are governed, even though the mutuality of this project is typically under-acknowledged by those in government. The purpose of state legibility is to make the particular contours of any given population and the natural environment visible and recognisable to a centre. The primary function of these projects is that of a map: they reduce all the multifarious and chaotic features of the world to a few indexical signs that provide the information necessary to classify and structure the environment, and then to manage and exploit it. These signs are – quite literally – signposts, that is, material objects that point to a specific and limited symbolic field; as such, they are fundamentally different from the ephemeral and situated act of pointing (see Widlok, this volume). The creation of state legibility thus produces a crucial difference: the difference between the simplified version of a population that is legible, and the diversity of the population in question that remains illegible. On the side of those in government there is now a constant need to refine the measurements, scale, and scope of their lenses, whereas on the side of the governed, there is a need to consider the consequences of being seen and simplified.

The contributors to this special issue propose that if the population that is made legible and governed is a human population, there will be a growing need to know and influence the intentions of those in government. Through the means to create state legibility – e.g. the grain fields of the peasants, the permanent family surnames in the census, or the grid structure of urban settlements – the government apparatus acquires the knowledge necessary to manage a population. Together with the imbalance in the control of the tools of legibility (e.g. census, register, maps), there emerges an imbalance of knowledge and imagination: the government knows enough about a population to manage it, while the population, in principle, doesn’t know much about the intentions of the government. The government can read the population, but only its superficial contours (e.g. name, place, movement), and not the precise intentions of each member of the population. In fact, the knowledge the government needs of a population is necessarily superficial. Too much knowledge about each single subject would be contrary to the purposes of government, such as taxation, conscription, and pacification: the problem is similar to the map of the empire that is as big as the empire, and therefore useless (Borges 1975). The epistemological violence that projects of state legibility (such as maps) exert thus facilitates physical violence wielded on a population. In this vein, Bovensiepen describes a predatory mode of governance relying on state legibility, which is opposed to (though not necessarily incompatible with) the mimetic governance that relied on mutual imitation and adjustment.

From the perspective of the population, the situation is inverted: the population cannot read the government and its representatives, but because of the potential impact government officials have on people’s lives, people try to understand the intentions of officials. What is created, then, is what Graeber (2012) calls ‘a lopsided structure of the imagination’: the unintended consequence of projects of state legibility is therefore the instigation of projects of mind legibility by the population.

These ‘lopsided structures of the imagination’ are typical of situations where there are huge differences in power: for instance, when women are more used to anticipating men’s perspectives rather than the reverse; or when ordinary people try to understand the intentions of government officials, while officials remain oblivious to the intentions of ordinary people. Graeber (2012) extends this insight to a general theory of bureaucracy, defined as organisations that limit empathy and enable violence: ‘dead zones of the imagination’ that are based on not interpreting the intentions of others. It is important to bear in mind that a prime mover in the tendency toward public mind reading might be the indifference of superiors toward the intentions of inferiors. This indifference, often accompanied by an aversion to explicit mind reading and a denial of empathy, in turn sets in motion a chain reaction toward public mind reading. Perhaps the most obvious example is military command, and Steinmüller provides examples of the impact of military relations on everyday exchanges in the villages of the Wa State of Myanmar. Once the fundamental inequality of commander and soldier is established, including a fundamental disparity of perspectives, commanders rarely put themselves in the shoes of soldiers, while soldiers increasingly feel the need to interpret the intentions of their commanders.

Such fundamental disparities are typically experienced once taxation, commodity exchange, and permanent government are introduced. In his contribution to this issue, Stasch demonstrates that since the Korowai have settled in villages and unsurmountable differences in wealth and power are imposed onto their everyday life, capacities for anticipating the intentions of powerful outsiders have become crucially important to their survival and well-being. Concomitantly, the Korowai living in villages welcome intra-group authority relations whereby ‘village heads’ tell subordinates what to do, as for example in the coordination of communal labour, in order to participate in the distribution of money transfers as part of community-driven development in West Papua. The outcome here is that while government officials often have hazy knowledge of concrete conditions amongst the governed, Korowai have gone out of their way to make themselves legible to officials to participate in development: they open clearings, compile lists of village residents, and lobby intermediary officials for seed resources.

While open speculation about the intentions of others and embrace of authority appears to violate a pre-existent ethos of egalitarian autonomy, Stasch proposes that these new attitudes are largely driven by old Korowai egalitarian principles. Patterns of transactional self-deprecation and of imitation (wishing to ‘be like’ others) previously served to level inequalities and were part of a general cultivation of opacity. But now, the same patterns are repurposed and help Korowai to make sense of projects of state legibility, practices of public mind reading, and new relations of authority. The Korowai case demonstrates the importance of situating opacity doctrines within the practical context of everyday life. When this context changes – as Stasch describes it did due to the emergence of a local ‘welfare state’ – the same patterns may end up working as pathways into subordination and towards state legibility.

The expansion of projects of state legibility, however, does not only create the necessity to understand the intentions of those who govern. It also creates new kinds of relationships between the governed, those who are the objects of state legibility. Amongst them, projects of state legibility create a need to make each other legible: for example, to co-exist in relation to centralised or externally imposed standards, to deal with the scalability of conflict in bounded areas, and to cooperate under changed conditions, such as the need to manage increasingly formalised covenants, public monies, and infrastructure. At their most basic, projects of state legibility are based on concentrations of humans, plants, and animals that James Scott has called ‘late-Neolithic multispecies resettlement camps’ (2017: 18). While such camps made state-building possible, they also provided ideal conditions for crowding and disease. Humans thus had to deal with new forms of drudgery that were part of living in crowded conditions: Scott emphasises that forced labour was essential to state building, and so were the large barbarian ‘escape formations’ around the earliest states. But aside from submission and escape, one essential corollary of living inside legible and crowded camps is the creation of new collective and shared means to make intentions public, so as to manipulate the conditions for co-existence. This is fundamentally due to problems of scale and emerging divisions of labour and can be seen clearly in new forms of managing internal conflict that deal with the unpredictability of others: who turns up to work, what to do about free-riders, how to deal with dissenters, aggressors, and threats to the public good?

For instance, aside from the hierarchical relations with powerful superiors, Steinmüller and Buitron (in their respective contributions) also identify an emergent tendency to speculate publicly about the intentions of others who live in the same ‘grid-like’ situation. In a continuum of dealing with the intentions of others, Steinmüller, for instance, points out that projects of state legibility propel a transition from ambivalent to unidirectional mentions of intentions, from gestures to ritualised forms to direct verbalisation. Similarly, Buitron illustrates how newly organised village assemblies exhibit a shift towards a decontextualised morality that promotes public accountability centred on individual intentionality.

The concentration of humans (animals and plants), and higher levels of interdependence between humans, creates pressure to render intentions public and legible – even when there is no direct request by the agents of the state to do so. Similar processes are observable when modern nation-states enforce policies of sedentarisation through rural settlement or village formation among previously mobile forest-dwellers. Such ‘efforts at domestication’ (Scott 1998: 184) aim to make mobile and peripheral populations legible – visible and clear – while moulding their landscape to the state’s techniques of observation, for example by encouraging the relocation of these populations to sedentary settlements. But while this is happening, new nucleated settlements and villages, understood as a specific mode of social organisation related to processes of state formation, trigger similar efforts among those who find themselves resettled. For instance, like many peoples who have experienced sedentarisation around the world, the Shuar of Amazonia (Buitron 2016: 122–153, 284–285; and contribution in this volume) and the Korowai of West Papua associate villages with a nonviolent ethos, which require that they cultivate inner harmony and a willingness to subject their intentions to public scrutiny or to disavow ‘anger, violence in favour of a new dispensation of “good thoughts”’ (Stasch 2013: 560–561).

Among our contributors, Buitron in particular focuses on the ways in which speech registers in new contexts of public interaction evidence more intrusive forms of accountability. She argues that for the Amazonian Shuar, coping with conflict in a situation of heightened proximity and visibility in new sedentary villages has resulted in new forms of peaceful speech-making, in which significant emphasis is placed on individual accountability and the capacity of new collective bodies to single out and emit judgements on individuals. The public speech of assemblies is premised on recently introduced ideas of peaceful subordination to village regulations and has an intellectual affinity with how Christian Shuar frame their obedience to God’s power, which, like community bylaws, stands above everyone and increases the possibility and scope of shamanic legibility.

New practices of legibility – such as the public assembly (Buitron) and corvée labour (Steinmüller) – thus facilitate the creation of accountability and ultimately establish a tendency toward public mind reading. The same correlation is true, a forteriori, for the institutions that are built on these practices, including the courts, schools, and government offices of contemporary nation-states. All of them require making the world to some extent legible from a central perspective; the centralisation of perspectives is generally accompanied by the establishment of abstract rules sanctioned by the centre (and often willingly embraced from the margins), against which ordinary everyday action is then constantly compared. World religions and imperial states thus commonly expand institutions of legibility against a new horizon of universal validity and a new space of interiority. The aspiration to universal validity and the promotion of new forms of interiority have been identified respectively in the transition from mythos to logos in ancient Greece (Vernant 2006), the beginning of modern theatre in Renaissance Italy and Spain (Egginton 2002), modern Protestantism (Keane 2002), and modernity as the spread of second-order observation (Luhmann 1998). Pace all these assertations of epochal change, claims to universalism and interiority are possibly typical of every cultural counter-movement against a civilisation backed by military power; movements that might later become established as the legitimising ideals of civilisational rule. If Protestant Christianity, for instance, started off as a ‘cultural revolution’ putting universal validity and interior relations against the powers of the Catholic Church, later these innovations became themselves associated with the hegemonic powers of church institutions and government. As such, when powerful institutions push toward universalism and interiority they frequently meet resistance from social groups not used to the same language of intentions and ways of intention management. This may be the reason why Christian practices such as confession, which rely on the assumption of the existence of an inner self that is accessible to introspection (Cary 2000) are so problematic in certain societies of the Pacific (Robbins 2008 for the Urapmin; Schieffelin 2008 for the Bosavi). By the same token, it helps us understand why introspection and free mental expressiveness are cast as morally dubious activity in the same contexts (Stasch 2008 for the Korowai; Throop 2008: 417 for the Yap).

Similar dynamics are at play once a legal system is created. The law is always a ‘concomitant of centralizing processes’ (Roberts 2005: 13), and modern legal systems fundamentally rely on the assumption that intentions can be made legible. Where we find law or legalistic reasoning, we generally find a vocabulary that implies a ‘theory of mind which allows the apportioning of responsibility’ (Dresch & Scheele 2015: 23). This is particularly the case in the psychologised theory of responsibility that has been foundation to criminal law in the West at least since the nineteenth and twentieth centuries (Lacey 2001). But ideas of intent also appear to be widespread in non-western legal institutions. They emerge in early Islamic law, and classical Chinese law ‘distinguished formally among premeditated, intentional, and accidental killing’ (Dresch & Scheele 2015: 23).

All this does not mean, however, that once intentions are ‘discovered’ and reckoned to be important, they will be managed in the same manner everywhere. Far from it – intentions matter in entirely different ways to universalist revolutions and to legal discourses, for instance. There are numerous forms of creating social legibility and state legibility, connected to different types of accountability, and resulting in different regimes of intention management. Addressing intentions as ‘minds’ and making them available in the public domain is one possibility, but here too there exist many different options of how to conduct public mind reading. Doctrines of transparency, therefore, need to be studied in context for they create as many problems as they solve – just like doctrines of opacity. Our contributors rise to the challenge on both accounts, and present a broad range of possible regimes of intention management and possible modalities of governing opacity.

Conclusion

In the simplest possible terms, the core proposal of this collection is that the cultivation of opacity in decentralised societies is opposed to the promotion of transparency in state societies. Our contributors deal with various aspects of this proposal and specify it further, for instance by discussing the rules of metahuman governance, the importance of pointing and signposts, and the emergence of decontextualised forms of morality and a ‘lop-sided imagination’ in centralised government. They demonstrate how different 'regimes of intention management' rely on tools of legibility, and their inherent potential to create accountablity, which we have outlined in this introduction. Those two concepts, regimes of intention management and tools of legibility, have allowed us to identify a series of correlations between the government of opacity and the political imagination:

The government of opacity largely depends on how legibility is established. There are numerous tools through which others can be made legible, as well as institutions that allow for ‘intention management’; even the most fundamental rules and roles fulfil such a function. Symbols, which are used universally by humans, may become tools of legibility, as may ritual. But in many societies, such tools and institutions remain fundamentally deictic and opacity-enhancing, and only a few of them become fixed signposts that allow for a centralisation of perspectives. If this happens, and state legibility is established, it has a series of consequences for the government of opacity: crucially, state legibility increases the need to reveal and adjudicate on the intentions of others, and ultimately produces a tendency toward public mind reading.

Tools and media of legibility provide means to hold people accountable. Legibility does not necessarily imply accountability, accountability does not necessarily involve intentions, and intention management is not the same as mind reading – but, as our contributors show, there are many examples for each correlation: from legibility to accountability, from acts to intentions, and from intentions to public mind reading. The association between accountability and mind reading also appears to work backwards: public mind reading increases the extent to which individuals become accountable to others. To hide intentions, on the contrary, generally helps to deflect accountability. Cultivating the opacity of mind, therefore, can maintain a situation of political autonomy by distributing responsibility across different actors. Our emphasis is on ‘can’: the opacity of mind is neither a necessary condition of, nor causally related with, political autonomy. As we have also shown, the opposition between the cultivation of opacity in decentralised societies and the promotion of transparency in centralised societies should not be overdrawn. With the increase of transparency, new demands for opacity emerge: for instance, in totalitarian societies that promote maximum transparency and aim at maximum conformity, there is a corresponding need to conceal intentions and separate interiority from the public performance of consent and obedience. This is but an extreme case of the general human capacity to create different institutions of accountability and legibility capable of governing opacity – a capacity that is essential to the ways in which politics are imagined and practiced anywhere.

Acknowledgements

The ideas collected in this special issue were first conceived at the workshop ‘State and Mind Legibility’ in June 2019 at LSE. The participants at this workshop helped us a lot in refining conjectures and intimations into concepts. David Graeber was our discussant and offered an inspiring set of comments on the problem of opacity; his absence remains an incomprehensible absurdity, but his spirit lives on. It was a great pleasure working with Judith Bovensiepen, Megan Laws, Rupert Stasch and Thomas Widlok on this special issue, and we are grateful for your patience and inspiration. Thanks also go to James Laidlaw for commenting on an earlier version of the introduction, to Hallie Wells for copy-editing, and to Nils Bubandt for editorial oversight.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The writing of this article was partially supported by the European Research Council [grant number 715725].

Notes

1 Here we draw on studies in environmental psychology where legibility is always understood within a social perspective; that is, spatial cognition is always influenced by social meanings. In this sense, legibility is not only a function of space or form – the physical elements of a situation – but also of the ways of understanding the relationships between individuals and surroundings (Ramadier & Moser 1998).

2 We prefer not to use the term ‘egalitarian’ to avoid importing a unified view of equality; see Buitron and Steinmüller 2020.

3 Sahlins followed anthropological convention when he described decentralised social arrangements without relations of command and obedience as ‘egalitarian’. We have argued elsewhere (Buitron & Steinmüller 2020) that this convention misconstrues the combination of autonomy, mutuality, and incommensurability that characterise the same arrangements. We continue to use the terms ‘egalitarianism’ and ‘egalitarian’ in quotation marks and when summarising the arguments of others, but with the caveat that what really is meant is the combination of autonomy, mutuality, and incommensurability, or what otherwise might be called ‘real existing egalitarianism’.

4 See Buitron and Steinmüller 2021, where we engage in much more detail with the proposal of the Original Political Society.

5 In some cases, opacity reflects the problem of hiding one’s intentions from others, while in others it corresponds to the asymmetry between ‘verbalizing my own thoughts and verbalizing yours’ (Keane 2016: 128).

6 For the purposes of this special issue, we understand responsibility to be concerned with the causation and the moral imputation of an act to an agent (or various others), and accountability as the secondary demand for a response from the responsible actor. Accountability is based on responsibility, but includes the obligatory response of the agent that requires the capacity to answer and the capacity to be questioned.

7 Duranti (2015: 11–20, 233–235) groups the theories of three prominent philosophers, H.P. Grice, John Austin, and John Searle under the ‘Standard Theory’ of intentionality.

8 In fact, even access to our own thoughts is interpretative, and hence uncertain and opaque, as Carruthers (2011) points out; see also Moran 2001.

9 A considerable amount of social knowledge stems from habit, such as learning from repeated interactions with the same people. This allows inference not based on signs but on tacit knowledge of how other people tend to behave, which may bypass even automatic, unconscious mind reading (Astuti 2015; Duranti 2015: 238).

10 That is to say, they are concerned not with the knowability of inner thoughts per se – which would be an eminently psychological problem – but rather with people’s power to hide such thoughts, their capacity for concealment, or what amounts to a ‘practical, moral, and even political problem’ (Keane 2008: 477; see also Rumsey 2013: 337).

11 Similarly, Hollan and Throop (2008: 393) argue that the aversion of quasi-first-person knowledge of the other, which is often felt as intrusive and shameful for the receiver, is ‘one of the most common and important forms of empathic communication we find anywhere’. This kind of basic empathy (to avoid presuming first-person knowledge of someone else) might be essential to practice self-sufficiency and emotional control (see also Briggs 2008).

12 This is particularly the case in formal occasions in the Pacific where speakers seek to restore harmony, so speech is kept deliberately opaque and indirect to avoid public offenses and retaliation (Duranti 2015: 44; Besnier 1989; Brenneis & Myers 1984; Watson-Gegeo & White 1990; Throop 2010: ch. 5).

13 In a similar vein, Bloch argues that morality is experienced ‘less as a matter of individual choice and more as a matter of submission and recognition of the presence of others who penetrate you’ (2013: 14), see also Bloch (2004).

14 See Bloch (2013: 15ff.) for another account of the porosity of individual boundaries within the social fabric, or in his terms, the possibility of interpenetrating each other’s bodies, a phenomenon he also attributes to Theory of Mind. The thrust of their arguments is however diametrically opposed: while Luhrmann emphasises the difference between various local forms of theory of mind, Bloch’s discussion of theory of mind aims at the human condition encompassing all local differences.

15 Parsing through what parts of persons (bodies, souls, or a combination of the two) are susceptible to revelation is, of course, not just a matter of epistemology but also of ontological difference – that is, what persons are and how they are made. In Melanesia, as Robbins argues, minds, not bodies, are considered unbreachable, so relationality does not require intersubjectivity, but instead results from intercorporality (2011: 313). In Amazonia, by contrast, the presumed sameness and intercourse between souls turns bodies into a more persistent and anxious focus of differentiation. Understandably, points of view, and ultimately the specific intentionality of subjects, lie in the body. Regardless of where the locus of agency and intentionality sits in each case, in both regions there is a common emphasis and concern with radical forms of alienation from one’s thoughts, for example when the person becomes the victim of sorcery, predation, love magic, etc. (see Londoño Sulkin 2012: 48–49 for Amazonia, and Stasch 2011: 450–452 for a similar phenomenon in Melanesia).

16 Bloch (2013: 25–27, 37) for example argues that religion and similar phenomena rely on the human-specific capacity to imagine other worlds, an adaptation that also sustains other aspects of human sociality, such as the representation of essentialised roles, statuses, and groups, which he calls ‘the transcendental social’.