Navigation – Plan du site

Silencing oneself, silencing others

Rethinking censorship comparatively [introduction]
Matei Candea
Cet article est une traduction de :
Du taire au faire taire


What is censorship? Is it different from other forms of silencing, and if so, how? Anthropology can provide some useful disturbances to familiar liberal debates about freedom of speech by expanding the cast of characters and the range of modalities of silence and expression. Yet the lessons of anthropology’s comparative explorations go beyond the blithe postmodern observation that silencing is pervasive. To speak of censorship is to mark out certain forms of silencing as illegitimate and unacceptable. An ethnographic attentiveness to the varieties of silencing must therefore keep in view the evaluative work which notions of “censorship” do for those who use them. Along the way of this comparative journey through techniques of silencing, this introduction suggests a broader observation: that the elusive distinction between censorship and legitimate silencing often relates to a sense of proportion between the silences actors impose upon themselves and those which they are thereby authorised to require of others.

Haut de page

Entrées d’index

Haut de page

Texte intégral

1What is censorship? Is it different from other forms of silencing, and if so how? For some, censorship is a transgression against a speaker imagined as essentially free, and must be denounced, whether it is displayed in plain sight in totalitarian regimes or hiding in the secret folds of Facebook’s algorithms; for others, censorship is an unduly negative name for the inevitable (albeit never innocent) cuts and limitations which define and shape genres and styles of expression, be they academic, artistic or everyday. On this latter view, expression is always entangled with power, and policing the bounds of the sayable is just another field of political struggle—a struggle which takes place on campuses, in courtrooms or online. These are sometimes seen as two fundamentally opposed philosophical, political and epistemic visions: a “modern” or “liberal” vision of censorship, opposed to a “postmodern” or “relativist” one (Bunn 2015; Darnton 2015). Yet anthropologists in particular are well placed to see that these are still arguments from a common ground, which draw from a shared stock of Euro-American philosophical and political concepts and controversies.

2To this discussion, anthropology can provide some useful disturbances by expanding the cast of characters and the range of modalities of silence and expression. Ethnographic accounts disrupt familiar assumptions about communication and agency by portraying contexts in which words cure (Lévi-Strauss 1949); in which silent songs transform the states of mind of addressees who cannot hear them (Taylor 2017); in which claims to understand another’s intention constitute a grave misstep (Stasch 2008); and in which confession becomes a form of competitive gift-giving, a flow of words which confessors find they have to struggle against rather than coax out of penitents (Rafael 1987).

3Yet the lessons of anthropology’s comparative explorations go beyond the blithe postmodern observation that silencing is pervasive. An ethnographic attentiveness to the varieties of silencing keeps in view the ways in which “censorship” is an evaluative as well as a descriptive term. To speak of censorship is to mark out certain forms of silencing as illegitimate and unacceptable. Those evaluative distinctions cannot be dissolved by analytical fiat, and they call for an attention to the ways in which acts of silencing can be authorised or critiqued: for instance, when Chinese firewalls meet Muslim ethics of civility (Wang, this volume); when a history of slavery becomes a public secret in rural Madagascar (Somda, this volume); or when Nepalese villagers passionately resist the silencing of traditional rituals by Maoist revolutionaries (de Sales 2009). Not to mention the many ways in which anthropologists themselves have struggled with the political and epistemic stakes of hearing silence as silence, and of the proper ways in which silence should be spoken of—if at all.

4In a productive return, these reflections also help to recast familiar debates about censorship themselves in an ethnographic light, not as the conceptual frame for any discussion of silencing, but as a particular variety of the concern with the powers of silence. Along the way of this comparative journey through techniques of silencing, this introduction suggests a broader observation: that the elusive distinction between censorship and legitimate silencing often relates to a sense of proportion between the silences actors imposes upon themselves and those which they are thereby authorised to require of others.

Censorship and the liberal imagination

5Censorship is a central concept in modern liberal political vocabulary, and like many such modern concepts it is borrowed self-consciously from classical antiquity—in this case from the Roman institution whose roots lay in the census and moral appraisal of citizens (Dury 1995). The specificity of the modern liberal notion of censorship stands out against the enormous variety of ways in which contemporary Euro-Americans refer positively to the joyful, productive or restful benefits of silence—often imagined somewhat nostalgically as the lost goods of an earlier age (e.g. Corbin 2018). By contrast, censorship denotes bad, restrictive silencing, wrongfully imposed upon an unwilling agent.

6There is a distinctive cultural logic to the liberal vision of censorship which deserves unpacking. It relies for its coherence on two distinctions. The first, as literary theorist Stanley Fish (1994) has famously argued, is that speech—or, more broadly, expression—can be set apart as a special form of conduct which is essentially about conveying contents (meanings, ideas, opinions, ideologies, artistic intuitions). In this way, the liberal imaginary of censorship tends to reduce a range of forms (music, films, images, actions) to versions or analogues of “speech” in this restricted sense of the conveyance of contents. The distinctness of expression in turn grounds the distinctness of censorship: power can block individuals’ actions and restrict their freedoms in various ways, but to speak of censorship is to claim that there is something distinctive about restricting freedom of expression. The second key distinction—outlined by legal philosopher Frederick Schauer (2006: 150) in his attempt to describe the ontology of censorship implied by liberal accounts—is between the uncoerced initial preferences of a would-be speaker and the subsequent coercive intervention of the censor. Censorship implies the possibility of an ideally free, autonomous speaker. It also implies that there is a specific and identifiable location from which censorship operates: a specific individual, or more commonly a group, who has the power to intervene in and limit the expression of the censored agent.

7In some obvious senses, censorship as defined above is a historical and ethnographic reality: there have been and continue to be collectives and individuals who think of themselves as censors, who understand their task as that of intervening in the communicative preferences of others, on behalf of a ruling authority. But censorship thus defined, or rather its elimination, is also an element of a certain liberal democratic mythos. Liberal democracy understands its legitimacy in matters of communication by opposition to a certain figure of censorship now relegated to a pre-democratic past, or to a non-democratic elsewhere.

8Concomitantly, an important feature of the political culture of liberal democracies is a sense of perpetual vigilance against the reintroduction of censorship in new forms: in the work of legal advisers inside publishing companies; in the ways in which experts (be they museum curators, librarians or academics) select and manage public knowledge; in the macro-structuring forces of the market; in the demands of representatives of particular religious or minority groups; or in the collective pressures of opinionated comment on social media platforms. Anthropologists have occasionally participated in this work of collective vigilance by seeking to “unmask” the new censors hiding in seemingly anodyne places, and the small slippages which take us—softly, softly—from official state-sanctioned censorship, all the way to “self-censorship” (Bruyère & Touillier-Feyrabend 2006).

Other silences

9Often, however, as I noted above, anthropological contributions to the topic of censorship have engaged with very different problematics, at some remove from the questions explored in the previous section. Contributors such as Bayly show the particular edge which ethnography and close association with particular people give in assessing the sophisticated mechanisms through which communication and silence can animate relationships between the state and its citizens, even when no obvious censorship is in play. In her account, Vietnamese “mobilisation posters” occasion silence in her interlocutors, and yet this is no straightforward instance of “self-censorship”. Rather the silence is an answer to the sense that these posters are not messages but crude, ugly tools demanding an immediate, “robotic” response—they seek to constitute viewers as speaking subjects of a particularly docile kind. In return, Bayly’s interlocutors exercised silence as an act of moral will and intellectual distinction.

10Conversely, Rappaport’s contribution shows how a combination of ethnographic, musicological and historical acumen can help us to hear silences even where quite the opposite seems to prevail. Rappaport focuses on songs which the Toraja in Sulawesi sing quite willingly, while nevertheless reporting that these songs are banned. The author unravels the context and history of this seeming paradox: the songs were once at the heart of rituals of fertility and sexual licence, which missionary Christianity suppressed. Once the rituals had disappeared, the songs could eventually resurface, and even be staged as “cultural artefacts”. This decontextualisation left the words intact yet strangely silent, just as it muted the rhythmic and musical qualities which had animated vital, desiring bodies. What was silenced here was not the content of the song but its effect.

11The anthropological archive is also full of accounts of societies in which

“the possession, articulation and circulation of knowledge are key to the definition of social distinctions relating to age, gender and leadership and to the existence of hierarchies and the distribution of power. Censorship, the control of behaviour and speech, is thus an integral feature of these societies, as are remarkable independence and freedom of choice in different measure (Marcus 2006: 229)”.

12Studies of secrecy surrounding initiation in particular (see Cros & Stoichita, Colleyn, this volume) unsettle the classic vision of censorship outlined in the previous section, by portraying enforced silence as necessary, functional and productive of a social world in which freedom and autonomy are nevertheless also present. Such settings are not animated by the classic tension between the power of censorship, on the one hand, and, on the other, an individual seeking to express their opinion or construct their identity. When speech is no longer simply a vector of opinions but rather “the instrument through which a man can become the property of another man” (Colleyn), teaching the virtues of silence takes on a very different resonance, invokes other materialities—“s’alourdir la bouche” (Colleyn), “avoir une feuille sur la langue” (Cros & Stoichita). It is not simply the nature of what is silenced (speech, images, postures, personal names) but the entire set of relations surrounding personhood and expression which are reconfigured.

13Implicitly or explicitly, such accounts of the productive, structuring effects of silencing provide a comparative counterpoint to visions of censorship outlined in the previous section. Yet, like all “frontal comparisons” (comparisons aligned across an “us v. Them” axis, cf. Candea 2018), these are disturbing not only because of the differences they highlight, but also on account of the familiarities they entail. Thus, on the one hand, as George Marcus (2006) has pointed out, there is a risk of such accounts feeding an exoticist imaginary, providing Western readers with a vision of “primitive” or “traditional” societies with their pervasive and productive silences as, precisely, the inverse of liberal “modern” ones in which such features have been superseded, for better or worse.

14Yet as Cros and Stoichita’s contribution makes clear, no such neat lines can be drawn—the stakes of silence in Lobi initiation entangle the anthropologist too. Her usual epistemological practices (documenting, revealing, describing) have to be reconfigured since she has herself partaken of and been transformed by initiation. The resulting graphic representation in these pages seeks to combine the imperatives of telling and not telling, showing and not showing. More broadly, clear-cut frontal binaries belie the complexity of a connected world in which, as we saw above, Christian missionaries or Maoist revolutionaries seek to silence the voice of local rituals while indigenous groups make claims for the protection of their intangible cultural property (Marcus 2006). Frontal comparisons, furthermore, tend to silence the obvious facts that not all “Westerners” are “liberal”, nor are all anthropologists “Western”—they silence them in order to articulate a more or less critical or productive contrast (Candea 2018). Indeed, frontal comparison’s only—and in itself non-negligible—value in this case is the way it cuts the liberal vision of censorship down to size, rendering it as no more nor less than one logic, one ethnographic object, amongst others.

15Beyond this basic relativisation, however, the most disruptive effect of anthropological accounts of “other” silences lies less in their exoticising potential than in their ability to point towards similarities and continuities. Thus the reader of these pages can move relatively effortlessly from accounts of the importance of silencing in initiation rituals, to Somda’s intriguing description of the ways in which the memory of slavery is silenced in Anôsy (Madagascar). Slavery remains what Michael Taussig (1999) has termed a “public secret” in Tanôsy society through the effect of everyday avoidances, interactions and implications—a “censure sans réels censeurs”. Somda draws on Zempléni’s (1996) account of the relational dispositif—the peculiar “savoir taire”—of initiation secrecy, arguing that Tanôsy society as a whole operates as a scaled-up version of the same interlocking dynamics.

16Following that shift, the reader will have little trouble recognising some similar dynamics in Le Caisne’s harrowing account of the power of social configurations to extinguish meaningful speech about a widely reported and horrific case of incest in a French village. She documents in precise detail the ways in which small verbal cues, changes of subject, refusals to speak and surprisingly bluff admissions (“everyone knew!”) combine to extinguish the sense that there is anything to talk about, in order to re-establish a kind of unbroken banality and normality. Le Caisne’s intricate account of the effect of these moves on her own near-inability to address the subject as an ethnographer recalls the complex interweaving of the unspoken into anthropological accounts of initiation: except that here, on home ground, the ethnographer sees it as her duty—however difficult this proves to be in practice—to speak, to puncture the skin of social secrecy.

17Moving across these cases, broad continuities (or at least contiguities) in the mechanisms of silencing are crosscut by radical shifts in the moral positioning of the anthropologist in relation to what is described. From cultural coherence and continuity as a positive good which the anthropologist seeks to preserve by the maintenance of select silences, via a discomfort about the tensions such consensus hides, through to a frank denunciation of a collective silence which covered and in some respect allowed an unspeakable crime.

18Nevertheless, from the perspective of the liberal definition of censorship outlined in the previous section, this effortless ethnographic slide across the mechanisms of what one might call “social silencing” is disruptive because it makes the logic of censorship uncomfortably familiar. For behind the contiguities lies the shadow of a broader generalisation: the relatively commonplace thought that censorship is, in the end, a pervasive, unavoidable and banal aspect of all social life.

Censorship undefined

  • 1 See Cody (2011) for a review of these arguments.

19In spotlighting these questions, anthropologists are in tune with a broader shift over the past few decades which has sometimes been described as “New Censorship Theory” (Bunn 2015). New Censorship Theory is usually associated with a range of scholars at the intersection of philosophy and the social sciences, such as Michel Foucault (1976), Pierre Bourdieu (2002) or Judith Butler (1997), but it builds, as Bunn perceptively notes, on a much older tradition within Western political theory, namely Marxist critiques of the liberal public sphere. Marx and his followers have long critiqued the liberal claim that the lifting of state censorship opens up a zone of free and unimpeded communication. Many, they noted, are still de facto excluded from this public sphere mainly, but not only, along lines of class, education and wealth; nor is this supposedly free “marketplace of ideas” without of its own subtler determinations, silences and exclusions.⁠1 While challenging Marxism in many respects, the cultural theorists of the late twentieth century have radically expanded these critical insights into a deconstruction of the two key distinctions outlined in the first section as integral to the cultural logic of liberal visions of censorship: the distinctiveness of communicative action, on the one hand, and the separation between free agents and external coercion, on the other.

20Consider, firstly, the distinction between communication and other types of action. In various settings and contexts, these authors have tracked the ways in which, always and everywhere, discourse is structured, shaped and constituted by power, and conversely they have built in various ways on philosopher of language J.L. Austin’s (1975) insight that words have real effects in the world. Extending the point beyond words to expression more generally, they undermined the thought that communicative action might be radically set apart from other classes of action. Of course, liberal legal regimes which otherwise eschew censorship have frequently been content to silence speech when it is not defined as really speech, but as something else—incitement to violence or discrimination or libel, for instance. The same goes for images and other expressive forms: these can legitimately be silenced as soon as they are no longer envisioned merely as analogues of “speech” but as incitative actions. When the members of the French film classification board studied by Esquerre (this volume) decide to make films inaccessible to certain audiences, these decisions are not grounded in the identification of illegitimate opinions or points of view, but in a set of concerns about the effects of images of sex and violence on the psychology or behaviour of young viewers. These moves play on the distinction between communication and action, moving certain entities from one side to the other of that boundary—rather than challenging the distinction itself.

21Much more challenging was the thought that expression itself, as meaningful action, might have silencing effects: the claim, central to much New Censorship Theory, that some forms of expression directly interfere with other actors’ ability to articulate themselves in their own terms. Thus, for instance, philosopher Rae Langton (2006) has argued that the repeated imagery of domination in much mainstream pornography warps and twists the ways in which women’s words can be heard and understood in real-life communication about sexual desire—a process which can in some instances make a woman’s “no!” inaudible as really meaning “no!”. As Schauer (2006) notes, critiques such as Langton’s are in one sense very much internal to the liberal imaginary of censorship, particularly in the way they still figure (some kinds of) silencing as “bad”; and yet they deconstruct the liberal distinction between free speaker and restrictive censor from the inside, by showing how some expression indirectly silences other expression. If pornography is itself a form of (indirect) censorship, these arguments ask, then on what grounds could one defend it from censorship?

22Unpicking one distinction leads to unpicking the other. Showing that the very ability of words to signify—let alone the ability of any speaker or writer to express themselves in a particular genre, style or register—relies upon a whole range of prior speech events also challenges the thought that one might identify a moment before the intervention of the censor, a subject of enunciation who is prior to social or linguistic determination. Put otherwise, this is the realisation that censorship is not a merely a negative, limiting force. Like power more generally in the Foucauldian view, censorship is productive: it doesn’t simply block but also elicits and enables expression on a variety of scales. Censorship is present at the very heart of communication because “the language of others creates rules that make language possible precisely by making some language impossible” (Schauer 2006: 153). More generally, censorship and self-censorship can form an intrinsic part of the creative process of authors, guiding stylistic choices and producing whole effects of genre, as Nora Gilbert (2013) has argued for Victorian novels and twentieth-century Hollywood films. Brkovic (this volume) explores a different aspect of the “productive” nature of censorship playing out in the work of Yugoslav filmmakers of the socialist period. Censors engaged in complex negotiations with authors, shaping rather than simply banning films. Furthermore, censorship practices could create a “buzz” around a movie which increased its visibility both locally and internationally—to the extent that some filmmakers today complain wistfully that the end of state censorship has also destroyed the cultural creativity of post-Yugoslav cinema. Hoek (this volume) shows a different facet of the productivity of censorial activity by tracing the interplay of regulation and representation in the official documents produced by the Bangladeshi Censor Board in the 1980s. While the Censor Board sought to control film societies by requiring the production of an intrusive and extensive amount of regulatory documentation, Hoek argues that these societies smuggled complex messages about the social value of art films into the margins of these bureaucratic documents themselves, condensing them into the logos and iconography with which they chose to represent themselves.

23The combined effect of the critiques and shifts in perspective associated with New Censorship Theory was to reveal censorship as an intrinsic and ineradicable aspect of meaning making. On this view, “[t]o be for or against censorship as such is to assume a freedom that no one has. Censorship is” (Holquist 1994: 16).

Censorship, expertise and the arts of silence

24Why, then, does censorship nevertheless persist as a term of denunciation and complaint? How can we explain the passionate and powerful resistances that it occasions? A cynic would reduce the problem to the bare mechanisms of a power struggle over public expression as a limited good. If censorship is everywhere, then they might conclude, with Stanley Fish (1994: 111), that “someone is always going to be restricted next, and it is your job to make sure that the someone isn’t you”. Yet there is an alternative to this view, which pays more attention to the nature of what is expressed when actors complain about illegitimate silencing—beyond everyone’s self-interested arguments for “the protection of speech they want heard and the regulation of speech they want silenced” (Fish 1994: 110). Paradoxically, this alternative starts from what might seem like a further flattening and deconstruction of the boundaries between censorship and other kinds of silencing: the insight, drawn from close accounts of what censors actually do, that censors are a particular type of “experts”.

25Anthropologists such as Dominic Boyer (2003) or William Mazzarella (2013), and ethnographically minded historians such as Robert Darnton (2015), in their studies of state censors in Communist East Germany, colonial and postcolonial India or pre-revolutionary France, have challenged the flat, stereotypical portrayal of state censors as mindless bureaucrats, simply erasing content at the behest of a totalitarian state. The censors who emerge from these accounts are far more complex figures. Their skills and expertise are shown to be comparable and indeed often continuous with those of other intellectual producers—they are themselves often authors, critics or actors. Their practice is shown to include not simply erasure and silencing, but often also editorship and curation of content, in what was sometimes a collaborative, albeit unequal, process with the authors of the work. Finally, we are given a sense of the ways in which some at least of these censors understood themselves to be involved in important, valuable and meaningful work—“a loving labor of cultural artisanry” (Boyer 2003: 538). Censorship was not so different after all, Boyer claims, from

the compromises to individual vision forced by the “peer review” system in the humanities and social sciences that mediates the great majority of academic publications. Of course, one may simply argue that peer review and related “gatekeeper” practices are matters of safeguarding professional standards and not genuine “censorship” (denoting ideologically oriented interdiction). But, does academic professionalism itself not also involve the socialization of individual authorship, the definition of the parameters of legitimate intellectual activity, the cultivation of generic “disciplinary” standards of research methods, interpretation, and representation, and so on? (Boyer 2003: 513–514)

26This has all the appearances of another relativist levelling, and Boyer himself seems to foresee this difficulty when he rather hopefully claims that his arguments “need not dull the teeth of outcry against the political abuse of intellectual labor” (2003: 540). He does not, however, give much of a sense of how this critical demarcation might be retained or reimagined.

27To do this, one might borrow Schauer’s (2006) observation that describing censors as experts also gives a way to talk about—and question—the legitimacy of particular groups to determine content. For instance, Schauer notes that many people in the twenty-first-century USA do not balk at librarians selecting which books to stock, yet would balk at the government (or even citizen pressure groups) dictating particular deselections to librarians. And of course this particular demarcation is not fixed, but a focus of ongoing struggles over legitimacy, as contemporary debates over decolonising the curriculum, for instance, attest (e.g. Hage 2017). Censorship, Schauer concludes,

is inevitable, necessary and desirable—but it is less inevitable, necessary or desirable that this group rather than that group do it. … The language of censorship is thus the language of professionalism, the language of expertise, the language of institutional competence, the language of separation of powers. It can also be, more maliciously, the language of turf. (2006: 162)

28There is a subtle but crucial difference between Boyer’s and Schauer’s positions. While Boyer relativises the liberal idea that censorship is necessarily illegitimate by comparing it to journal editorship, Schauer notes that censors, editors and other experts are always prone to being denounced as illegitimate. The point is not simply to level censorship with other types of expertise, but to show that expertise (including censorial expertise) is a critically and often hotly contested category.

On proportionality

29Censorship would thus be the name we give to illegitimate expertise. This observation takes us one step away from the relativist flattening of censorship into the unbounded category of “silencing”. But this in turn raises the question: is there any regularity to the ways in which this illegitimacy is established and contested? A comparative reading of three contributions to this special issue will help us to suggest a hypothesis.

30Esquerre describes the deliberations of members of the French film classification board. Sometimes described by others as censors, the members of the board staunchly resist the label, and consider themselves “classifiers”, whereas Esquerre prefers the term “suppressors”, in order to distinguish his analytical language from that of his interlocutors. What they are quite clearly, however, is “experts” of a particular kind: a collection of persons imagined as knowledgeable in different ways on account of their age or profession. A claim to expertise is demonstrated also in the regularity of their modes of interpretation, which Esquerre details in his piece. Call them censors, classifiers or “suppressors”, the members of the board exemplify a particular intersection of forms of silencing which constitutes expertise. On the one hand, the board constitutes the locus of legitimate “content-determining authority” in respect of films in France. On the other, it aspires to this legitimacy and attains it only through a set of self-imposed cuts and silences. To be legitimate, these classifiers must not speak of certain things (such as the directors’ intentions), or not speak in certain ways (such as through “value judgements”).

31The articles by Aymes and Provost take us further away from the more obvious terrains of censorship, and yet they both demonstrate the same structural interrelation at the heart of expertise and institutional competence between self-imposed silences and authoritative speech, the kind of speech which can ground the legitimate silencing of others. Aymes unpicks the layers and cross-cutting modes of silencing embedded in a nineteenth-century Ottoman document: a report written by the Grand Vizir to the Sultan, concerning a diplomatic incident involving the Russian ambassador and the seizure of a putatively fraudulent Russian boat. As in the historical accounts of censorship in pre-revolutionary France (Darnton 2015) or early modern Britain (Shuger 2006), what needs to be silenced in this delicate political setting is not any particular message or ideological content, but primarily talk as it impinges on reputation, honour and trust. At the heart of these moves lies what Aymes describes as a commitment to “silence contradiction”, to cut through the possibility of adverse speech by the production of incontrovertible evidence. And the document itself must demonstrate to the Sultan that this has been done and done well. Crucially, the Grand Vizir’s institutional competence is demonstrated partly through what the document itself doesn’t say. Provost’s exploration of the work of coroners in contemporary New Delhi takes us to a different setting in which experts find themselves reporting, in writing, to a higher authority, simultaneously demonstrating their legitimacy through self-imposed silences—for instance, in regard to interpretations and inferences about causes of death—and drawing on this legitimacy to impose their reading of the case—to “silence contradiction”. The documents coroners produce ought thus to be read, Provost urges, as the effects of a tense pragmatic play with the silences imposed by professionalism.

32Examining censorship through the prism of expertise, professionalism and institutional competence thus highlights the meeting point in which self-imposed silences come to authorise the legitimate silencing of competing views or particular contents. The legitimacy of particular actors’ silencing moves is premised in part on their own adherence to conventions of professional self-limitation which they can play with, perhaps, but not shrug off. There is thus a crucial relation between the way collectives of experts are silenced and the way they silence others.

  • 2 The historian cited by Brkovic, according to whom Yugoslav filmmakers might respect Yugoslav censo (...)

33This suggests a potential answer to the fraught question of how actors demarcate “the social negotiation of accredited knowledge” from “the political abuse of intellectual labour” (Boyer 2003): the ability to sustain certain forms of institutional silencing as legitimate may well depend in part on the felt sense of proportionality between the silences and limits these institutions impose upon their own experts, and the silences and limits which these experts are thereby authorized to impose upon others. Perhaps, then (pace Boyer), the difference between the widespread (albeit, of course, not uncontested) acceptance of the principle of academic peer review and the widespread dissatisfaction with state socialist censorship during and after the fall of those regimes points to something more than “a kind of a convenient fulcrum of intellectual counter-distinction” (Boyer 2003: 512). Perhaps the latter’s claims to cultural expertise were judged by many to be less robust than the former’s in part because of a felt disproportion between what these censors required of others and what they required of themselves. Critiques of censorship often imply this sort of evaluation: What standards do these censors adhere to? What forms of professional conduct are they accountable to? What stringent forms of education and training have they had to undertake in order to get to their position?2

34On the horizon of these concerns lie the censorial activities of those wholly unaccountable non-human agents—algorithms, firewalls, filters and other semi-autonomous devices—which populate the networked spaces of collective expression. These non-human censors are a world-wide phenomenon, although Wang (this volume) documents one particular version of them in her account of resistance and self-censorship in a digital Islamic space embedded in China’s censorship system. The complex online “assemblage of surveillance, blockade, leakage, diversion, friction, and disjuncture” developed by the Chinese government points towards extensions of censorial expertise into machinic automatisms. In distinguishing his own online “moderating” activity (guanli)—in pursuit of “democratic enlightenment” and informed by “basic civility and Islamic ethical conduct”—from such state censorship (shencha), Wang’s interlocutor might be seen to echo the theme of a proportionality between self-restraint and the silencing of others.

  • 3 I am grateful to Emmanuel de Vienne for suggesting I render explicit the formalist nature of the h (...)

35In sum, this hypothesis suggests a formalist way out of the “undefinition” of censorship. Instead of distinguishing censorship from “good” silencing based on their contents or actors, one might seek instead to identify the recurrent forms entailed in actors’ own distinctions between legitimate and unacceptable censorship. The relationship of proportionality between silencing oneself and silencing others is one such recurrent form of legitimation. There are others. One might mention, for instance, the reversibility between censor and censored (today’s censored might—de jure or de factobe tomorrow’s censor) which in some contexts founds the experience of belonging to a community of “peers”.3

Coda: On non-censorship

36This observation leads us back towards an ethnographic reconsideration of the cultural logic of liberal attitudes to censorship from which we began. Perhaps there is more there, after all, than two philosophically untenable distinctions (recall: the distinction between action and communication, on the one hand, and between free agent and restrictive censor, on the other). The idea that one might perfectly draw those distinctions is rather the mark of a libertarian utopia (or dystopia) of a world without social convention. A liberal antipathy to censorship need not, after all, go quite that far. It need simply imply a concern with drawing the line between legitimate and illegitimate silencing. As I suggested in the previous section, one way in which such lines are drawn is through an attention to the proportionality between the silences one requires of oneself and those one requires of others.

37And yet this vision also draws our attention to an empirical fact which should after all seem quite strange to anyone who has fully understood the message of New Censorship Theory: that censorship is everywhere. The stubborn empirical fact is that in some settings, some forms of expression which self-consciously break with social convention, which are profoundly unpalatable to a great majority of the population, which may well be harmful to individuals or dangerous to the social order, and which the state has the ability to silence, are nevertheless not silenced. Heywood (this volume) examines one such case: the surprisingly sparing way in which Italian courts have been choosing to apply a law which bans fascist propaganda. The pervasive consensus that censorship is everywhere turns such cases of liberal non-censorship into an ethnographic problem in need of explanation.


38The writing of this article was supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement 683033).

Haut de page


Quand dire, c’est faire, trad. Gilles Lane, Paris, Éditions du Seuil.

« La feuille sur la langue. Pragmatique du secret initiatique », Cahiers gabonais d’anthropologie no 17, p. 1938-1953.

Questions de sociologie, Paris, Éditions de Minuit.

« Censorship as a Vocation. The Institutions, Practices, and Cultural Logic of Media Control in the German Democratic Republic », Comparative Studies in Society and History no 45/3, p. 511-545.

« La censure et ses masques », Ethnologie française no 36/1, p. 5-9.

« Reimagining Repression. New Censorship Theory and After », History and Theory no 54, p. 25-44.

Le pouvoir des mots. Politique du performatif, trad. Charlotte Nordmann, Paris, Amsterdam.

Comparison in Anthropology. The Impossible Method, Cambridge, Cambridge University Press.

« Publics and Politics », Annual Review of Anthropology no 40/1, p. 37-52.

Histoire du silence. De la Renaissance à nos jours, Paris, Flammarion.

Censors at Work. How States Shaped Literature, New York, W. W. Norton & Company.

La censure. La prédication silencieuse, Paris, Publisud.

There’s No Such Thing as Free Speech, and It’s a Good Thing Too, New York & Oxford, Oxford University Press.

Histoire de la sexualité, t. I : La volonté de savoir, Paris, Gallimard.

Better Left Unsaid. Victorian Novels, Hays Code Films, and the Benefits of Censorship, Stanford, Stanford University Press.

« “Anthropology is a White Colonialist Project” Can’t Be the End of the Conversation », Media Diversified, en ligne : [dernier accès, septembre 2019].

« Introduction. Corrupt Originals: The Paradox of Censorship », PMLA no 109/1, p. 14-25.

« Subordination, Silence, and Pornography’s Authority », in Robert C. Post (éd.), Censorship and Silencing. Practices of Cultural Regulation, Los Angeles, Getty Publications, p. 261-284.

« L’efficacité symbolique », Revue de l’histoire des religions no 135/1, p. 5-27.

« Censorship in the Heart of Difference. Cultural Property, Indigenous Peoples’ Movements, and Challenges to Western Liberal Thought », in Robert C. Post (éd.), Censorship and Silencing. Practices of Cultural Regulation, Los Angeles, Getty Publications, p. 221-246.

Censorium. Cinema and the Open Edge of Mass Publicity, Durham & Londres, Duke University Press.

« Confession, Conversion, and Reciprocity in Early Tagalog Colonial Society », Comparative Studies
in Society and History
no 29/2, p. 320-339.

« From Ancestral Conflicts to Local Empowerment. Two Narratives from a Nepalese Community », Dialectical Anthropology no 33, p. 365-381.

« The Ontology of Censorship », in Robert C. Post (éd.), Censorship and Silencing. Practices of Cultural Regulation, Los Angeles, Getty Publications, p. 147-168.

« Civility and Censorship in Early Modern England », in Robert C. Post (éd.), Censorship and Silencing. Practices of Cultural Regulation, Los Angeles, Getty Publications, p. 89-110.

« Knowing Minds is a Matter of Authority. Political Dimensions of Opacity Statements in Korowai Moral Psychology », Anthropological Quarterly no 81/2, p. 443-453.

Defacement. Public Secrecy and the Labor of the Negative, Stanford, Stanford University Press.

« L’art d’infléchir les âmes. Les chants anent des Jivaro achuar comme techniques d’apparentement », Terrain no 68, p. 46-67.

« Savoir taire. Du secret et de l’intrusion ethnologique dans la vie des autres », Gradhiva no 20, p. 23-42.

Haut de page


1 See Cody (2011) for a review of these arguments.

2 The historian cited by Brkovic, according to whom Yugoslav filmmakers might respect Yugoslav censors because they were part of a shared cultural elite who know what Expressionism is and were familiar with “the biography of Marinetti”, provides a clear instance of this logic: where censorship was less critically viewed, it was precisely because censors were recognised by the censored themselves as “peers”.

3 I am grateful to Emmanuel de Vienne for suggesting I render explicit the formalist nature of the hypothesis.

Haut de page

Pour citer cet article

Référence électronique

Matei Candea, « Silencing oneself, silencing others », Terrain [En ligne], 72 | novembre 2019, mis en ligne le 21 novembre 2019, consulté le 24 janvier 2020. URL : ; DOI : 10.4000/terrain.18773

Haut de page


Matei Candea

University of Cambridge

Haut de page
  • Logo INSHS
  • Logo Université Paris Lumière
  • Logo Université Paris Ouest
  • Logo MAE - Maison Archéologie Ethnologie
  • Logo FMSH-Diffusion
  • OpenEdition Journals