CAIRN-INT.INFO : International Edition

Introduction: going beyond technological determinism

1Following the Iranian protests in 2009, the ousting of the Tunisian and Egyptian presidents in early 2011 breathed new life into the question of the internet’s role in social movements operating under authoritarian regimes. These movements are often referred to as “Twitter” or “Facebook” revolutions, protests for which the internet can appear as the principal source. This perspective reflects that of a number of works published over the last decade, which were characterized by technological determinism and strongly influenced by an ideal vision of democratization. More specifically, many case studies have been devoted to examples of “resistance” [1] to authoritarian regimes in different countries. These studies have documented the use of the internet by dissident groups and organizations [2] who create internet sites or distribution lists to broadcast their ideas at the national and international level, to connect with other activist networks, and/or to coordinate the organization of protests. Elsewhere around the world, internet users have expressed their indignation about various events or joined movements which are not generally directed against their respective regimes, but which nevertheless put authorities in a difficult position and force them to respond favorably to their citizens. [3] Chinese internet users did this when they posted indignant messages online in response to arbitrary policing and unfair trials. [4] In Iran, the consequences of September 11th are discussed and western books and films otherwise outlawed by the regime are downloaded. [5] In Saudi Arabia or in Egypt, police blunders or even the condescending behavior of princes and leaders can create veritable scandals online. [6]

2For the most optimistic authors, by braving censorship and forcing the authorities in question to moderate their repressive politics, these online actions chip away little by little at the inflexibility of authoritarian regimes: they thereby constitute the basis of the gradual but inevitable transformation of political regimes. Part of this orientation can be explained by the fact that a great number of these studies were financed by organizations directly or indirectly interested in promoting democracy, such as the Rand Corporation, the Red Cross and the Carnegie Endowment for International Peace.

3However, the diverse studies available have demonstrated that the effectiveness of online movements led by political dissidents is quite relative. Although activists for democracy generally succeed in broadcasting their ideas to a particular international audience, online actions do not always suffice to obtain results from the relevant authorities. For example, the Saffron [7] revolution in Burma produced significant online actions, without forcing local authorities to concede anything. The same is true for the Falun Gong [8] sect and Tibetan separatists [9] in China. From this point of view, we can see that internet-led actions have had varying results in the different countries that experienced protests during the winter of 2011, including Libya, Morocco and China. It would be equally interesting to retrace the timeline of events in Tunisia and in Egypt to highlight their respective differences.

4When online calls to protest do not bear fruit, these relative failures are often attributed to successful censorship. [10] Such censorship can take various forms, be they technical (URL or keyword filters), legal (holding service providers accountable), or social, by relying on social expectations and self-censorship. This is why Geoffrey Taubman [11] and Christopher Hughes concluded, for example, “that the internet will not democratize China”. [12] Nevertheless, it is relatively easy to evade such censorship by using ironic language [13] or ambiguous meanings, and examples of important messages published using these strategies are abundant. Even if it must obviously be taken into consideration, censorship alone cannot explain the weakness of online protests against certain political regimes, just as successfully bypassing said censorship is not the sole factor explaining the emergence of online social movements.

5The goal of this article is to propose a theoretical framework allowing us to explain the logic behind the formation of these online movements, while avoiding the pitfalls of technological determinism. To begin, we will briefly review the existing literature in order to demonstrate that many authors have been unable to free themselves from a dualist conception of “society” as opposed to the “state”, or even to nuance this dichotomy. To go beyond this obstacle, I will propose an interactionist theoretical approach using the Goffmanian notion of “framing”. This concept, which has already been used many times to analyze social movements, has never before been proposed as a means of understanding online movements in authoritarian contexts. However, its contribution seems potentially useful, as it allows for a nuanced understanding of the part internet users themselves play in limiting public expression, and at times in freeing it, without ignoring the restrictions to which they are subject or the evasive measures they utilize. Finally, I will illustrate this hypothesis with a series of extended interviews with Chinese internet users, which investigate how their understanding of public order drives them to online self-censorship, without prohibiting critical expression in the spaces they deem the most appropriate, and, at the same time, without barring them from engaging in arguments interpreting the contours of this public order. Through its representations of internet users, this article will shed new light on how online movements are focused and directed, while often exceeding the limits established by the authorities.

A persistent Manichean vision of state/society opposition

6The new wave of online protest movements has often been analyzed as a form of “empowerment of civil society” in opposition to the state. It is true that authorities are addressed in a very specific manner by online movements and that the former are consequently forced to take them into account. Nevertheless, the use of the term “civil society” poses a number of problems.

7On the one hand, this term can encompass very diverse actors and initiatives which are not always entirely representative of all internet users. In some cases, “civil society” designates very broadly an entire society apparently seeking to “liberate” itself, but which is prevented from doing so by censorship. [14] Other authors have used this term to refer to the specific actions of special interest groups or NGOs, who coordinate with each other via distribution lists and launch politically charged discussions on web forums and blogs. [15] The internet is equally useful for creating informal networks or for expressing arguments about very specific subjects, as some lawyers do on their blogs, [16] but this strategy does not necessarily enable the creation of lasting alliances. For example, Markku Lonkila notes that the anti-militarist movement in Russia [17] remains very fragmented and poorly organized.

8Much work has also been dedicated to describing the personality and ideas of famous bloggers, who may be politically engaged to varying degrees. Marc Lynch has proposed three categories of bloggers: activists, “middlemen”, who connect to the foreign and especially western blogosphere ( “bridge blogging”) and bloggers from the “public sphere” who, without necessarily being political, discuss current events that have an occasionally political dimension. [18] Another widely studied category of bloggers is that of “citizen journalists”, [19] who devote themselves to filling in the gaps left by traditional media, by transmitting witness testimonials or investigating injustices and scandals.

9These individuals, or sometimes organizations, have an undeniable influence on the formation of certain movements. It is therefore important to understand their strategies and motives. Nevertheless, it is still very difficult to evaluate the exact nature and size of the public that follows them. In most of the countries in question, internet users represent a small minority of the larger population. They are generally young, urban, educated and financially well-off. Among these users, those who choose to express themselves are even less numerous. It is thus impossible to assume that their ideas are representative of the multiple points of view of all internet users and, even less that they are representative of the population at large.

10Additionally, blogger identification, in addition to the different categories outlined above, usually has as a main criterion the degree of politicization of those in question. However, the work produced by these bloggers has most often been of a hybrid nature, as Naila Hamdy has emphasized. [20] As events develop, some blogs shift from one category to another, and the posts published stem more often from a spontaneous reaction than a structured protest movement. Similarly, some rather influential bloggers do not respond to the expectations of external observers and refuse to get involved in political debates, as was the case of certain Russian poets during the 2007 elections. [21]

11We touch here upon a second important limit with regard to the notion of “civil society”. This concept generally implies the idea of separation with and opposition to the state. From this point of view, the positioning of a great number of individuals online is entirely ambivalent. In fact, the strategies put in place by internet users to defend their causes often consist of using official discourse for their own purposes, thereby presenting various state actors with their own contradictions, while still ostensibly displaying unwavering support for the regime. [22] Some bloggers also benefit from the space for interpretation that exists at the core of official doctrines to hold extremely varied political positions. [23]

12In fact, what motivates internet users living in the different countries in question can be very dissimilar. Tools of coordination and online publication can serve to defend all sorts of causes, sometimes far removed from the defense of democracy. For example, it is alarming that the internet is also an important vector of communication for terrorist networks, [24] Jihadists, [25] and even neo-Nazis and the Taliban, [26] who all defend very different agendas. However, this category represents a small fraction of the many uses of the internet, which, in most countries, is primarily for entertainment purposes, with most internet users displaying a certain indifference to politics (in which they are not very different from other countries). This apparent political apathy is frequently attributed either to local political culture [27] or to a form of postmodern egotism seen to be typical of the younger generation of internet users. [28] Recent revolutions in the Arab world have nevertheless shown that these arguments do not hold water.

13Consequently, some researchers believe that it is useless to search for obvious signs of politicization and that it is more fruitful to pay attention to signs indicating the gradual birth of civic-mindedness among internet users. Karsten Giese thus emphasizes that, like all discourse, the online universe developed by internet users is permeated with representations and values. Their thematic or aesthetic choices illustrate different life views, and as such can be considered to be the product of identity construction. [29] Others have shown that, given the demographic profile of internet users, this construction is often tied to the values of a more or less privileged “middle class” or of the youth that constitutes the majority of internet users in numerous developing countries. [30] In certain cases, the representations conveyed contribute to the challenging of a number of hegemonic discourses, as Merlyna Lim has demonstrated in the case of Indonesia. [31]

14The proliferation of blogs where internet users can express and exchange personal opinions thus draws attention to the importance of conversations and horizontal connections, which can be interpreted as the gradual development of “civic-mindedness” [32] or the birth of a “public sphere” ultimately favorable to the political participation of citizens. [33] At the very least, an online “public opinion” seems to exist, which can, on occasion, assert itself when faced with the authorities. [34]

15There is certainly much to explore in this domain, but existing work has remained very normative, and is not always devoid of a certain kind of social evolutionism. In effect, the image of a “nascent” or “embryonic” public sphere is imbued with the idea that internet users are gradually going to adopt behaviors more in line with tenets of political participation, and as such will “emancipate” themselves from the state, if not confront it in the end.

16Nevertheless, the criteria defining this participation can differ considerably from one study to another. Moreover, this approach inevitably leads to the recognition of the imperfect practices of internet users with regard to such normative expectations. Specific attention is, for example, directed at the deliberative quality of political discussions on forums. And yet the rules of debate often vary greatly from one arena to another and they often prove disappointing in terms of debate. [35] However, the news is peppered with anecdotes which give online discussions a disorderly and sometimes even violent image. In the violence of “mobs” (spontaneous online mobilizations) and the narcissism of bloggers, some see troubling developments. [36]

17These disappointing observations have even provided fodder for a number of alarmist political discussions, which authorities in the concerned countries have taken advantage of to justify increased control of online publications. For example, in China, inspired by the troubling consequences of phenomena such as the collective harassment of individuals whose conduct is regarded as immoral (the so-called “human flesh search engine”), some researchers close to the authorities work on procedures for measuring and predicting public opinion, as well as techniques for manipulating or “guiding” opinions. [37]

18So, whether their focus is on means of control or evading control, on detailing the strategies of activists to broadcast their stories, or on evaluating the deliberative qualities of online discussions, existing studies are very often marked by a normative vision of political participation and a relatively Manichean vision of relations between individuals and the state. This has often led to a recognition of the ambivalent or paradoxical character of the behavior of internet users, and in any case has generally portrayed the internet as responsible for all policy changes; this to the detriment of a more refined analysis of these internet phenomena.

19A potentially more fruitful approach would be to acknowledge the complex nature of the formation of public opinion and to clarify its different dimensions. Isabelle Thierreau and Hua Linshan took steps in this direction in a case study about the “Sun Zhigang” affair in China, showing that the success of a movement built around an injustice relies mainly on its ability to mobilize several segments of society who can identify with the victim. This includes not only defending a cause which resonates with widely-shared moral beliefs, but also finding support in the press, as well as from opinion leaders and official institutions. [38] For his part, Marc Lynch has drawn attention to the importance of political opportunities for the success of causes backed by the Arab blogosphere. According to him, bloggers occupy a central place at the heart of the “public sphere”, in the sense that they help to frame narratives, but this role is played in close interdependence with the press that proposes an agenda and relays these narratives. [39]

20These studies invite us to consider the internet not as an ideal space for deliberation, but as a place, like other medias, [40] where complex and conflicting interactions play out between a multitude of actors, among whom may figure authorities (from all different levels) and journalists, as well as different types of internet users: bloggers, of course, but also occasional contributors and readers. Yang Guobin has proposed the notion of “multi-interactionism” to explain these different relationships, which he documents rather painstakingly in the case of China. Nevertheless, he does not depart from the idea of a “civil society” in opposition to the state and likewise tends to presuppose that the aspirations of non-state actors (journalists, NGOs, ordinary internet users) are democratic in nature, which cannot be simply assumed. [41]

21In particular, in most existing studies, there seems to be a relative lack of information about the motivations of ordinary internet users who come to join the ranks of online movements. In fact, most studies have emphasized the visible aspects of such movements, taking as their object the themes of the causes defended, the strategies of the most engaged protagonists, or the contents of online discussions. Consequently, it would be interesting to focus on the less visible dimension of online movements: namely, their audience, which is caught in the tug-of-war between the different actors involved and which constitutes the strength of a movement in the eyes of the authorities. In fact, this pressure is often deduced from the number of page hits, or from cursory reactions on forums, such as simple agreement or onomatopoeia, which have nothing to do with explicit and argued positions.

The notion of framing online speech

22To address this line of reasoning, the Goffmanian notion of “framing” [42] and the work it has inspired in the domain of the analysis of social movements can constitute a very fruitful theoretical response. As Jean-Gabriel Contamin argues, “by ‘frame analysis’, we refer to a collection of works which [emphasize] that a movement does not depend so much on the availability of resources, the opening of political opportunities, or on a cost-benefit calculation, as it depends on the manner in which these different elements are perceived. Against the omnipotence of structural models of explanation, it is a matter of taking into account the subjective and the ideal components of participation (bringing ideas back in), the work of construing meaning engaged in by all the stakeholders of a collective movement.” [43]

23Many works have thus focused on the themes of social movements and the justifications offered by their actors during protests; they have also granted capital importance to the role of representation. In particular, David A. Snow has emphasized the fact that social movements require a great number of individuals to share the same interpretation and qualification of an event (as unjust, for example) as well as perceive the same possibilities for remedying it. Snow terms this “frame alignment”. [44] This phenomenon explains the importance of the storytelling work done by the “movements” entrepreneurs”, as well as the central position of the media in broadcasting these stories. It may also explain the success or the failure of some movements as a function of how much the reasons cited for action resonated with one or multiple communities.

24As Robert Benford reminds us, it is not sufficient to describe the “frames” at a given moment, but on the contrary, it is important to show the dynamism of their construction, in order not to reduce them to a structure which individuals are unable to influence. In effect, the frames are created through the interactions between individuals and accordingly, are always the object of dispute and conflict, over the course of which they are reinterpreted and renegotiated. In particular, this implies considering the different and unequal positions of individuals, but also the emotional dimensions of their engagement in this process. [45]

25In an attempt to further nuance these studies, Daniel Cefai has also proposed returning to Goffman’s original texts. [46] By observing situations “in the process of creating themselves”, one can grasp the subtlety of this notion of “frames”, which cannot be reduced to a set of reference points that people draw upon to make sense of various situations. The “frames” are produced during the interactions themselves and this process can lead to misunderstandings, “dissonances”, conflicts of interpretation or acts of “reframing”. There is also a theatrical dimension to this process. Actors play “roles” on the different “stages” where they develop. [47] They can play a “double game”, lie or joke – in brief, engage in all sorts of “procedures” which allow participants in the interaction to understand, or to think that they understand “what is happening here”.

26This plasticity of “frames” is particularly crucial in the case of political crises. Although he doesn’t directly use this term, Michel Dobry has demonstrated that each “move” made can modify the logic of the participants’ situation. This affects the intelligibility of the situation and the pertinence of the participants’ calculations, which can explain the extreme unpredictability of crisis situations. [48]

27Although it includes a cultural dimension, the term “frames” also allows us to avoid one of the major risks associated with the notion of culture, which, when reified, prevents us from grasping the misunderstandings, conflicts of representations, double games, in short, the plastic and thus often conflicting character of the meaning we assign to situations. Whereas the notion of culture sometimes serves as a veil for the “paradoxes” of an individual’s behavior, the notion of “frames” allows, on the contrary, for one to enter into the complexity of those same paradoxes by showing how different actors draw upon varied cultural repertoires, how they can interpret them in different ways and adopt or discard them over the course of their interactions.

The participation of individuals in domination

28Daniel Cefaï has shown that this “has consequences for understanding authority and legitimacy”. Classic political science notions like “material or symbolic force, the norm and the law, interest and contract, deliberation and consensus”, prove insufficient to describe this process of the coproduction of a legitimate order and to comprehend the manner in which individuals come to “submit to a political authority”. “Authority and legitimacy are not comprehensible if one only considers structural analyses: what one believes to be true, good, right and are all equally at play in the situated game of conceding rights and assuming power”. Moreover, Dobry himself emphasizes that the notion of “loyalty” proposed by Albert O. Hirschman does not suffice to explain the recourse to support for authority. [49] He describes a “harnessing of calculations” which “assumes nothing except the fact that for the members of a given sector – no matter what they want or believe – they cannot act differently, in their relevant activities, than calculate in function of their sector’s social logic. They are, in a certain sense, trapped by this logic”. [50]

29Therefore, this position assumes an interactionist vision of power, very different from that which can be read between the lines in the majority of the aforementioned works on online movements. Power here is seen as a relationship in which both the dominant and the dominated participate. [51] It is therefore not so much a matter of understanding people’s strategies of resistance or transgression when confronted with the state, as it is about grasping the conditions of adherence to public order for various categories of individuals, their daily participation in the definition, maintenance, and negotiation of rules and norms, or in some cases, their bypassing or contesting of said rules and norms. As such, as Beatrice Hibou highlights, [52] it is possible to “approach the question of domination in authoritarian or totalitarian regimes by analyzing the processes of legitimation” and their “justifications”. [53]

30It is upon this particular aspect that I would like to focus here. In effect, the analysis of frames as applied to online movements, as described above, constitutes a vast program of research, in which many aspects merit exploration. It would be interesting, for example, to engage in a detailed analysis of online blog or forum dialogues, or to review the press’s editorial choices. However, it seems that the least studied, and in my eyes the most intriguing, angle consists of exploring framing as it is perceived and exercised by “ordinary” internet users during their everyday use of online opinion sites. I propose, therefore, to study their mechanisms of speaking, from self-censorship to the most virulent protests. In other words, I will focus on describing the themes and the language of speaking out, [54] of indignation [55] and, in some cases, of social mobilization. I adopt an approach already tested in many works, [56] but which has not yet been used to study the specific case of the formation of online movements under authoritarian regimes.

31From this perspective, the “frames”, such as individuals perceive them, could be more simply defined as follows: they allow internet users to know what can be said and what cannot be said online, and how, where, and when one can say things as a function of who one is. It is a matter of questioning social norms and shared representations, but also ones which are disputed or renegotiated by different categories of web surfers. This notion allows us to simultaneously explain the constraints which limit internet users’ ability to speak out and the opportunities they can take advantage of to defend sometimes very politically sensitive causes. More specifically, this includes censorship, but also politeness, indignation and transgression. This idea favors representation and social norms in light of their constricting, but also fluctuating and uncertain characteristics, and does so without diminishing distance and criticism, role-playing and stratagems.

China : choice of terrain and methodology

32The material I am using comes from a study done in Beijing in 2006 and 2007. As China simultaneously has the largest population of internet users in the world, an exponential number of online movements, and one of the most sophisticated structures of censorship, it is an unparalleled example of the paradoxes of the development of the internet under an authoritarian regime.

33The proposed theoretical approach, which involves a line of questioning exploring usage, values and representation, invites the use of a qualitative methodology. [57] Observation of events in situ, “in real time”, in a cybercafé or in the homes of the interviewees and commentated by the participants, would no doubt have been the ideal choice in other contexts, but was too delicate in this case. First of all, such observation presented great methodological and deontological problems (how to be present at the right time? could one take excerpts from private instant message conversations?). Secondly, taking into consideration the political context and the nature of my interviews, my presence around the respondents could have led those questioned to strongly self-censor. With such subject matter, it is sometimes more fruitful to present the themes in question as general questions, rather than inquiring into the personal practices of those concerned. In the end, the most valuable material seemed to be the opinions of internet users about their own practices, about the manner in which they did or did not disguise themselves, and about how they justified their behavior.

34For this reason I held a series of extended interviews with 50 internet users in Beijing. Given the main purpose of my study, I sought not to disqualify those who consulted websites without actually commenting, and therefore ruled out the use of an online survey. For the same reason, it was not possible to recruit respondents through an online ad, which would have overly biased the sample. Consequently, I used a survey company to assemble a sample of individuals meeting a series of pre-established criteria. [58] Respondents were thus selected for their intensive use of the internet (consultation without actually contributing being sufficient) and for their association with the most common web surfer profile: between twenty and thirty years old, urban and active. The profiles of respondents in this group were also as diverse as possible in terms of age, gender, professional occupation (secretary, chef, graphic artist, etc.), wages (from 0 to in excess of 2000 euros a month [59]) and places of residence.

35Over the course of these interviews, face to face and in Chinese, which generally lasted two hours, internet users were questioned about their paths in life, their aspirations, about what it meant to them to surf the web, to express themselves or to react to others’ opinions. In particular, they were questioned about the constraints imposed upon them, the manner in which they perceived or justified those constraints, which aspects of these constraints seemed the most important to them, and the possibility of them breaking certain rules. [60]

36The choice of Beijing and the fact that the sample was limited to one particular profile were decisions essentially motivated by practicality. This study does not therefore presume to study a population representative of all of Chinese internet users, but simply to show the value of adopting “framing” as an approach to studying a specific public.

Chinese internet users and the drive towards modernity

37Nevertheless, the characteristics of this group are close to those of the most prevalent internet user profile in China. At the time of the study, there were 338 million web surfers in China, [61] mostly concentrated in large cities. They were generally younger and more educated than the average Chinese citizen, and they enjoyed relatively higher income. On the basis of their profiles and their adoption of new technologies, these web surfers are representative of the generation born after the beginning of economic reforms, referred to in China as the “post 80” generation, and which also corresponds to the first generation of only children. By their presence on the web alone, they have a tendency to consider themselves as the pioneers of Chinese “modernity” and its values.


“If someone is modern, they are obligated to be on the internet.” (Deng, F. 28 years old, 2 years of university, housewife, 18 June 2007) “The internet allows for information to be broadcast faster and this then enhances culture and that ultimately makes human culture progress.” (Zhu, H. 23 years old, engineering student, 17 November 2006)

39They thus echo the discourse on the “modernization” of society and “the improvement of the quality of the population” [62] which is ubiquitous in China. This discourse was first created by the Party in official propaganda, which promised modernity in exchange for social and political stability. [63] This argument can be found in official press editorials, in the public relations campaigns accompanying public policies (urbanism, education, major projects, etc.). It is also relayed in public speeches, which frequently depict the ideal “middle class” family – which owns its own house and car, with a child and often a dog. This ideal family is highly educated and “civilized”. This discourse accompanies campaigns against “archaic” behaviors, such as spitting in public places. Beyond the advertising iconography, adopting this “modern” lifestyle and its associated values functions in most of society as a sign of distinction. In the context of growing social stratification, any distance from this norm contributes to the justification of numerous forms of discrimination towards groups such as peasants or migrant workers. This discourse is so pervasive within Chinese society that such people sometimes consider themselves to be “backwards” or “not very civilized”. [64]

40The role of the media in broadcasting this discourse is of central importance. It uis conveyed via Chinese, as well as (pirated) American television shows, and through magazines which endlessly illustrate the characteristics of an ideal urban style (women’s magazines, home decoration, cars, technology, etc.). While all Chinese people cannot presume to adopt the typical attributes of this lifestyle, the internet allows a growing number of them to feed their imaginations and to appropriate these norms for themselves. Blogs and forums thus play a very important role, because they house innumerable discussions about what the modern Chinese lifestyle should be, allowing participants and readers to draw connections with the preoccupations of their daily lives. [65]

41Moreover, the way in which respondents understand “modernity” serves as a basis for the “framing” of their storytelling, in the sense that it serves as a guide to determine how they can express themselves online.

Framing speaking on the Chinese web

42It is evident that the internet users questioned did not define the rules imposed on them. In particular, they respected the constraints imposed by censorship, of which they were keenly aware. Most of the respondents described it quite accurately, or at the very least outlined its main principles:


“If we break the laws of the country or if we say things on the internet which it’s forbidden to publish, if someone realizes it then it’s removed, and I think that there are many regulators and they are very efficient.” (Pan, F, 24 years old, office employee, 500 euros a month, 15 May 2007) “All sites are monitored by the government, and it’s the same for administrators, if someone realizes that there are a lot of negative things on their site, including things not favorable to the government, the government can force the administrator to remove them, that’s happened to me many times.” (Tian, H, 27 years old, 2 years of university, employed in a foreign business, 350 euros a month, 20 June 2007)

44Despite this very lucid awareness of censorship, the limits of the control exercised by the authorities are not always very clear. Neither the exact criteria of censorship, nor the consequences and the risks incurred are universally well understood. Web surfers must rely on their own intuition to understand these rules, or sometimes wait for other web surfers or moderators to warn them. For example, many respondents mentioned having been reprimanded by other web users when they expressed themselves in too aggressive or too critical a manner. The definition of norms is thus done by trial and error and collectively, depending on what people have written. The respect of these rules is guaranteed by social control as much if not more than by the actual or supposed surveillance to which people are submitted.

45Behavior from one person to the next can therefore vary greatly. Tian, who was the most political internet user in our sample, and who held the most critical position towards the government, considered that he risked nothing more than seeing his contributions removed. Conversely, the great majority of those interviewed adopted much more prudent behavior: they self-censored.


“You need to control yourself, if you express your own opinions on the internet. Where the government can exercise a certain amount of control, you cannot say certain things, and you shouldn’t just say anything because the internet has no limits.” (Yang, F, 32 years old, zhuanke, [66] automotive industry employee, 320 euros a month, 26 July 2007). “According to Chinese law, even if there is no firm limit, we might not be punished, but we also might be punished. For these kinds of things, nothing is clear, the consequences are not clear. It’s a question of personal opinion, and as an individual, it is necessary to accept the consequences of your opinions.” (Lin, H, 28 years old, zhuanke, sports coach, between 500 and 1000 euros a month, 24 October 2006)

47What interests me here is not so much the phenomenon of self-censorship itself as the subjective processes behind it, which are seen particularly clearly in the last quotation. Faced with no firm guidelines, web surfers are forced to incorporate the criteria of political control into a system of norms which is coherent enough to be operational when they want to speak out online. Here, the question of significance is central: web surfers use a “frame” in the sense that they construct, individually and collectively, an array of explanations which give meaning to the different constraints imposed on them.

48In so doing, they can draw on a number of argumentative repertoires. Certain internet users mentioned Chinese culture, which implies an unconditional adherence to the political regime. Many others adopted the vocabulary used in official ideology, which encourages citizens to support social order and thus political conformity, in the name of building a prosperous and modern society. Thus, faced with annoyances as diverse as disorder, vulgarity, fighting, and also censorship, web surfers value what they call “responsible” or “civilized” behavior, which is indispensable for guaranteeing the continued existence of different internet sites; consequently, insults, pornographic images and anti-governmental posts (among other things) are discouraged.


“I find that limits are necessary, and I also find that they even have advantages. If we sent too many things, without them being managed, it would be chaos.” (Liu, F, 21 years old, benke, [67] 4th year finance student, 16 November 2006)

50This definition of chaos does not just mean an outbreak of vulgarity or personal attacks online. It also includes certain forms of critical and political expression. For example, Zhang is relatively critical in terms of the government’s politics, while Zen supports it wholeheartedly. Nevertheless, both think that public confrontation of ideas is futile, because the interests of different societal categories are fundamentally contradictory. For them, bringing these contradictions out in the open would be irresponsible, “uncivil” behavior because that would ruin the foundations of living together. [68]


“No, we cannot necessarily talk about everything, I find that certain people can represent a group, that is to say that they represent a fixed group, like a professional occupation, maybe some people can represent manual laborers, and others can represent farmers, and others can represent white collar workers, but those who speak in the name of white-collar workers cannot represent citizens from lower classes in the society, that’s for sure, and citizens from lower classes cannot represent the upper classes like white-collar workers, and it is certain that they cannot represent each other.” (Zhang, H, 26 years old, benke, decorator, around 400 euros a month, 5 November 2006) “If I start to violently debate things with many people, I don’t believe that after the argument we could still come together and address shared issues peacefully.” (Zeng, H, 25 years old, benke, medical worker, 800 euros a month, 6 June 2007)

52It is important to understand to what extent the idea of “chaos” and the concept of civility, used here, are tied to the modernization discourse that is omnipresent in China today. To be a “modern” web surfer is to control one’s speech and to express oneself in a “civilized” manner, in order to avoid provoking any form of disorder. All deviant expressions, that is to say vulgar, conflictual or critical speech is thus systematically attributed to the assumed lack of education among certain categories of the population and is consequently disqualified. This idea is particularly evident in the comments made by Wu, one of the richest respondents in our sample :


“So people who post in forums today are of rather poor quality. […] We need to elevate the quality of the nation. – What does that mean? – For example, if someone starts a new site, they are obligated to control it very strictly. […] I find that internet sites hurt themselves when they don’t take their responsibilities seriously. In the end, the site is dirty, and as soon as it is dirty, a certain class of people will certainly no longer come to the site. They will think that what they see doesn’t apply to them, no? […] In cybercafés, some people chat, some play games, and the rest are there to insult people, yes or no? Their cultural levels are low, they don’t even have a middle-school diploma.” (Wu, H, 28 years old, benke, hotel sales and marketing, more than 1000 euros a month, 17 October 2006).

54In this discourse, it is the very existence of spaces of expression open to everyone which is threatened by the behavior of less educated and thus “less responsible” web surfers. This last quote is from a respondent who wasn’t at all critical of official discourse. Nevertheless, this manner of conceptualizing public order online was shared by the majority of the respondents, including individuals like Zhang, who were relatively critical of the government.


“[Ten years ago] on the internet, there was no one, there wasn’t this, well these rather lower class people, because they couldn’t afford the cost of the internet and social expenses. At that time, I found that people discussed things in a rather pure manner, because there were very few people, perhaps there wasn’t this kind of deception you’re talking about.” (Zhang, H. 26 years old, already quoted)

56In short, this suggests, in a rather iconoclastic manner, that certain Chinese internet users curb themselves, or at least apply certain limits to their freedom of expression online. They thus engage in “framing” of a sort, which consists of circumscribing speech within the limits they judge to be necessary to maintain public order online. In so doing, they align themselves with the official discourse which constantly warns them against the dangers of an uncontrolled internet. On the one hand, it is possible to see this as a protective reflex meant to guarantee the continuing existence of spaces of expression that have become very precious to them. On the other hand, it is remarkable that the principles of justification they use support the construction of a very conflicted and hierarchical society. Therefore, social control and self-censorship can be analyzed as the elements of a strategy of distinction among web surfers, many of whom claim to be members of the “middle class”.

Roles, scenes and hidden transcripts

57It is nevertheless important to approach the framing of speaking publicly in a nuanced manner. As Jean-Noel Ferrie has shown, in the case of Egypt, [69] we can consider these norms of speaking out (“civility”) as conventions, as ways of behaving oneself in public which allow interactions to happen smoothly. This does not necessarily imply the adoption of these same norms in other situations. Web surfers can adopt very anti-establishment opinions in certain circumstances, as long as they express themselves appropriately and on the right platforms. Criticism is thus not erased, but redirected towards spaces believed to be private, whether that means they have a smaller audience or a less official status.

58Here, the theatrical metaphors proposed by Goffman [70] are useful to try and capture the nuances that this public dimension of speaking out implies. Goffman effectively suggests that a theatrical dimension exists in interactions between individuals. People play roles, they adopt fronts, [71] and they work on the image of themselves that they want to show. These roles can change depending on the stage where the individuals find themselves, on the nature of their interlocutors, and on the types of relationships that connect them. This suggests that in speaking out in an online space, web users play a role which they deem to be the most appropriate to the situation. This is exactly the vocabulary used by the young man below:


“The internet is like a stage. A stage? – Yes, it is like a stage, people play different roles, a little like a TV series, or a play. Whatever you say, whether it’s true or false, it’s nothing but the true psychological reaction and expression of someone in at a certain moment in a given context. It’s neither true nor false.” (Lin, H, 28 years old, already quoted)

60Essentially, there exists a representative dimension in the manner in which web surfers justify the limits they put on themselves. For example, this might be a question of expressing oneself “as Chinese” which, in the next quote, amounts to giving a positive image of one’s country:


“For example, when there are [annoying] things, of course I personally think that the government has a problem, and that you shouldn’t limit people’s freedom. But when I look at things as a Chinese person, I hope that this type of thing doesn’t happen too often, meaning things that are not good for the country. We should not make too much of it because if everyone pays attention to this aspect, I think that all in all it won’t have a good influence.” (Cai, F, 30 years old, secretary in a factory, 450 euros a month, 25 July 2007).

62As this quote shows, we should not exclude the hypothesis that web surfers keep a certain distance with respect to the roles they play on different platforms. If Chinese web surfers do not allow themselves to speak about very sensitive subjects on the most public online platforms, that does not mean that they are disinterested. Rather, they move these discussions to spaces which seem to them to be more private, where they can apply more lenient rules to their expression. These reflections explain the initial observation of the supposed apathy of Chinese internet users. As in the case of the American associations observed by Nina Eliasoph, [72] the relative absence of political questions in public discussions does not always signify real indifference. Rather, it is often indicative of a public code of conduct, a way of “playing the game”, which rejects conflict and public criticism.

63And yet, the norms imposed can vary considerably depending on the type of “stage” concerned and the “position” [73] or the status of the people involved in the action. The web is never considered to be a single, monolithic space where the same norms of behavior always apply. It is composed of a number of large portals, which are relatively controlled public showcases, but also of a string of blogs and discussion threads, many of which are a fortiori considered to be personal and private sites by their users, despite their theoretical visibility. These aren’t directed at the public at large, but more often at friends or those in the know.


“In general, small forums are pretty lax. We use them mostly to chat with friends. There might be 20 people on the site, always the same. However, on more commercial forums, everything is standardized; people go to the largest sections. – What do you mean by standardized? – Standardized, that means larger forums. They are more controlled, more specialized. They are less personal.” (Zhang, H, 26 years old, already quoted.)

65In such niches, information can circulate more freely, without being controlled by the government or users themselves. From this point of view, the ambiguity of numerous web pages allows for an extraordinary circulation of information, and the majority of web surfers interviewed are pleased about this.


“Therefore we often suppress information, and then the same information will not be discussed from the same position. We look at things from a certain angle; we analyze them from a certain angle. It’s national interest that counts the most. But on the internet, there is a lot of other information, which comes from the SINA portal, or from any blog or review, that lets you hear different things”. (Xu, H, 33 years old, construction worker, 200 euros a month, 24 July 2007)

67In addition, it is possible to show that, even in the most public forms of online speech, framing can provide the resources required to elaborate the publication strategies of antiestablishment ideas. In effect, the same points of reference used by certain people to determine their limits are used by others to circumvent censorship in a strategic manner. It is necessary to take into consideration the “tactical” aspect of respecting social norms. [74] For example, when they have a sensitive message to post, web users often choose more restrained spaces or ambiguous formulations. Zhang, who once had to defend a cause which concerned him personally, recounts his strategy as follows:


“Everything depends on how you present your comments and everything depends on the subject. Depending on the level of freedom of the subject, we choose the way to say it and also the place.” (Zhang, H, 26 years old, already quoted)

69These strategies are largely used by web users who are heavily engaged in the circumvention of norms of public expression. Humor, irony, plays on words, and allusion are so many “hidden transcripts” [75] which allow individuals to avoid authoritarian control and reprobation from other web users.


Do people often talk about politics? – Yes, a lot. But they talk about it indirectly. – Do you have an example? [laughter] – Uhm… [clears his throat]. It’s difficult. For example, in every country, people make jokes about their leaders”. (Ren, H, 25 years old, benke, marketing employee, more than 1000 euros a month, 1st August 2006)

71Additionally, satirical content in the form of comic strips and videos cut across the barriers of censorship and of self-censorship, thanks to their informal nature. In such videos, mascots are given indecent names like Caonima, the “horse of grass and mud”, which sounds just like “fuck your mom”. Such techniques are technically difficult to detect and interpret, and thus to censor.

72As Ashley Esarey and Xiao Qiang have highlighted, these tactics are a means of “speaking the truth” with each other but also in front of the authorities. [76] Moreover, the provocative tone of their increasingly virulent satirical content seems to indicate that the acceptability of censorship is in decline. This does not, however, necessarily contradict the dimensions of the framing of speaking out online as described above. In fact, Chinese internet users mock censorship by showing its archaic nature, and by revealing how it restricts the ordinary activities of modern Chinese citizens, even when they are just looking to be entertained. Censorship becomes all the more intolerable when it ultimately denies their ability to present themselves as “responsible” by exercising their own regulation of public speech. [77]

73This highlights the fact that “framing”, which often contributes to limiting the free speech of internet users, may on occasion become a rhetorical tool to justify breaching the limits set up by the authorities.

Battles of representation

74In this light, it is important to emphasize the fact that normative frames are easily blurred, constantly changing and subject to misunderstandings and permanent renegotiations. Despite an apparent consensus on the necessary modernization of the country, numerous debates arise every day around the meaning of this modernization and its implications for daily life. When a problem arises, each protagonist can propose an interpretation of the norms and of their meaning, and attempt to get others to agree with them in order to win the day. Thus we can view a proportion of online movements as fights over representation, during which different actors employ storytelling tactics.

75In this context, the official discourse, in which the government defends values such as transparency, honesty and the good management of public resources, provides arguments which web users can exploit to air their grievances. Every new official doctrine thus becomes a new opportunity which some web surfers never fail to exploit. Authorities often find themselves trapped by the very same discourse they engendered. [78]

76Zhang helped me to better understand this dimension. He has a Labrador retriever, which is illegal because a municipal regulation bans dogs larger than 35 cm inside the city. He decided to use the internet to make his situation known and plead the case of dogs in the court of public opinion. To this end, he decided to present his message in the most consensual manner possible, by playing on the different meanings of the concept of public interest.


“First, you can’t compete with the government’s policies, no. What we have to do is to explain to ordinary people around us why we think that raising a dog won’t bother other people’s lives, and why we think that, even for things like rabies, the question of control is not really about the dogs, it should be about the people, and why we think that what people say about dogs in regard to pollution, in reality it’s a problem of city management, not dogs. When I said this, I just basically said that we can discuss this with our friends, our neighbors, and we can tell them that people who raise dogs have rules, and that we would not disturb the public interest.” (Zhang, H, 26 years old, already quoted)

78By mentioning his willingness to preserve the public interest, this web user operated within a frame, formulating his problem using the vocabulary of the authorities, in the hopes of increasing the chances of his speech being both heard by other web users and authorized by those who censor.

79This space for interpretation at the heart of official discourse even gives rise to negotiations with the people in charge of the censorship. The law obliges blog and forum moderators to take responsibility for the contents published on their sites. This means that moderators must hire workers to monitor what’s published, and it’s at this level that the majority of censorship occurs. Zhang can thus discuss with them how to interpret these rules.


“Also, since I spend a lot of time on the internet, and I often go on forums, I have a lot of friends online. Moderators are very often forced to reflect on the personal and collective points of view, for them it’s a kind of pressure. People like us are often, are surely the backbone of a forum’s activities, thus our views often get a lot of attention, and if they don’t include politics, if they don’t target governmental departments, if they’re only about the activities of the forum, we can say first of all ‘I didn’t do anything violent, I didn’t say anything violent’. Under such conditions, our speech is not in principle limited, no one limits us in an inflexible manner, in principle at least”.

81This approach to negotiation is thought out and rests on a very calculating analysis of the authorities’ agenda and the propaganda techniques used to impose it.


“There are still things we can discuss, […] because in reality this control depends on the specific timing. There is a certain kind of coordination with the government. When we don’t coordinate with the government, let’s take an example; today for instance it’s very complicated to have a dog in Beijing. Right now, the government does not allow stories talking about the advantages of house pets to appear in the press, like the idea that dogs are our friends, and all sorts of articles that might discuss this type of question. Why? Because right now they want everyone to know about the dangers of rabies.”

83As he emphasizes, circumstances and timing are very important. In the case of the 2008 Olympic Games, the question of dogs took on a particular dimension: for the authorities, it was about eradicating dogs likely to spread rabies, which resulted in the euthanasia of tens of thousands of so-called “stray” dogs. It was nevertheless necessary to avoid prompting the indignation of citizens championing the cause of domestic animals and who had already organized protests in 2006. This specific case opened a window of opportunity allowing for more open discussions on this subject, which at times exploited the contradictions of this policy with the promise of a “modern” lifestyle.

84This quote also shows the importance of the role of the media in web users’ understanding of a situation. The press generally channels the official political position to the population at large, thus indicating the sensitivity of a subject at any given time.

85Nevertheless, the press is also in an ambivalent position. As Zhang’s negotiations with moderators would seem to indicate, the main objective of website editors is to retain the web users who participate the most, since the content they produce attracts more traffic. Similarly, print media seeks to increase profitability by attracting new readers, which drives them to discuss increasingly sensitive subjects. In this context, an increase in online conversations gives journalists the opportunity to publish otherwise unpublishable articles, since it prevents the authorities from covering up these matters. [79] As such, internet users, journalists and the authorities (and others still) send each other signals which help to shape how speech is framed on both sides. This reciprocity works like an echo chamber and amplifies certain themes which can sometimes then lead to real online movements, often in a relatively unexpected manner.

86Nevertheless, we should not forget that the same phenomenon also creates a “spiral of silence”, [80] which means that a rarely discussed subject will be very rapidly drowned out by other, more popular topics, or confined to places not often visited. [81] The framing which drives this has numerous dimensions. The directives of the Party at the national or local level have a relatively structuring effect which prevents the most sensitive subjects from being transmitted. In addition, the growing commercialization of newspapers has driven the media to increasingly orient themselves towards subjects deemed interesting by their wealthier audiences. [82] Similarly, the fact that the most favored social classes are the best represented among Chinese internet users induces a kind of social control which does not encourage the more marginal to speak out, in line with the image of a “modern” Chinese society. Social stratification, which has never been as important in China as it is today, thus proves to be a crucial dimension for understanding online movements in this country.


88* *

89Because it takes into account the representations and reasons for speaking out online, the notion of “framing” offers a new tool to understand the constraints imposed on web users. In an authoritarian context, framing does not elude the question of censorship, but it situates it within a larger mechanism of social control. Unlike earlier works on this subject, my approach has attempted to consider how internet users might participate in activities of limitation, by considering power in a relational sense. On one hand, this has allowed me to relativize the effectiveness of state surveillance stricto sensu, by showing the existence of spaces and forms of expression where critical discourse can take place. On the other hand, it has permitted us to understand the emergence of movements in a dynamic manner, by demonstrating that the norms defining what can and cannot be said are sometimes quite fluid, subject to change and a fortiori capable of becoming the object of a battle of representation.

90In China, the idea of “modernity” crystallizes a large part of these battles of representation. This idea, as it is presented in Party doctrine, represents a form of compromise, in whose name the population is invited to not publically question the political regime, in exchange for a stability that is favorable to the economic development of the country.

91The ultimate problem for the government is demonstrating that it can keep its promises by not only guaranteeing the improvement of the quality of life, but also by guaranteeing the ability to attain a more “modern” lifestyle. It is not just a question of maintaining a double-digit growth level, even if that is important: it is also a matter of proposing a constantly evolving narrative of what this “modern” lifestyle should mean, and publicizing policies consistent with this speech.

92Internet access and the possibility of expressing oneself in an entertaining, creative and even critical way are thus essential attributes of this “modernity”. Consequently, criticisms of “old” censorship are becoming more and more virulent. [83] At the same time, these spaces of expression serve as new venues of protest open to Chinese internet users, in a country where other channels of public expression remain almost inaccessible. This represents a considerable obstacle for the authorities who must now keep track of such protests. In a sense, the tolerance of the authorities towards this relative increase in freedom of expression is the condition required to maintain their own legitimacy. [84]

93Still, this increase in the freedom of expression has its limits, which are not only imposed by authorities but also to a certain extent by internet users themselves. The prevalent discourse on modernization is inseparable from an unprecedented growth in social stratification in China. [85] This stratification has created a fear of instability and chaos, which plays a determining role in the social control exercised by web surfers. For the moment, I have only been able to study a category of internet users whose living conditions are relatively comfortable. [86] For them, the last 30 years have brought very real yet still fragile improvements. All mentioned a fear of being left behind in an extremely competitive environment where social insurance systems are poor. The care that those surveyed took to distinguish themselves from other groups (farmers, migrant workers, the less educated, etc.) was also obvious. This translates into the marginalization of certain themes or the delegitimization of certain types of speech, in particular when most critical of the regime. In other words, internet users are not always the most fervent champions of freedom of expression for everyone.

94It is therefore insufficient to raise the question of the “digital divide” in the form of internet access, but one must first and foremost be able to understand the ability of different groups to rally concerned individuals to their causes. Different categories of the population are not equal in this high-stakes game, whether in terms of technical competence, cultural capital, rhetorical resources or even political opportunities. For the moment, it would therefore seem that in China, the greatest room for maneuvering is awarded to individuals hailing from the most favored categories, who are also the most likely to support the notion of political stability. [87]

95Consequently, from the point of view of the political regime, the most dissident stances are not necessarily the most dangerous in the short term, as they are generally removed by censorship or by the effects of framing public expression. In a broader sense, this logic leaves little place for pluralism. However, subjects that preoccupy the largest and most influential categories of internet users can genuinely challenge the regime, to the extent that they erode the credibility of the tacit agreement on which adhesion to these categories rests. Access to housing, jobs for young graduates, and, increasingly, environmental questions are some of the subjects which particularly illustrate the contradictions of the discourse about the modernization of the country.

96Here we thus observe the fragility of an equilibrium that rests on shifting pillars such as the credibility of the modernization discourse, or the perceived risk of political instability by groups of very diverse actors interacting with one another. The speed with which scandals can emerge on the internet leads us to believe that different elements of this framing, and with them their participants” calculations, can change in a very short space of time.

97These remarks support the idea that that which is essential is found neither in cultural explanations nor in simple technical determinism. It is not a question of negating the material properties of forums and blogs, which, among other things, have granted internet users new venues for protests, have sped up the process of the formation of movements, and above all, have awarded them a national audience. Nevertheless, these dimensions cannot replace an analysis of the political context of the societies in question; that is to say, more generally, they cannot replace the traditional tools of political sociology. [88]


  • [1]
    See, for example, the case of Singapore: Tan Suqin, “Resistance on the internet: a study of the Singapore case”, masters dissertation, Singapore, National University of Singapore, 2006.
  • [2]
    Michael Chase, James Mulvenon, You’ve Got Dissent ! Chinese Dissident Use of the Internet and Beijing’s Counter-Strategies (Santa Monica: Rand Corporation, 2002); David F. Ronfeldt, The Zapatista “Social Netwar” in Mexico (New York: Rand Corporation, 1998); Volodymyr Lysenko, Kevin Desouza, “Cyberprotest in contemporary Russia: The cases of and”, Technological Forecasting & Social Change, 77(7), 2010, 1179-93.
  • [3]
    See Shanthi Kalathil, Taylor C. Boas, Open Networks, Closed Regimes. The Impact of the Internet on Author
    itarian Rule (Washington D.C.: Carnegie Endowment for International Peace, 2002).
  • [4]
    On the Chinese case, see Pierre Haski, Internet et la Chine (Paris: Seuil, 2008). See also blogs such as “China Media Project” (<>), “China Digital Times” (<>), or “East South West North” (<>).
  • [5]
    On the Iranian case, see Nasrin Alavi, We are Iran (London: Portobello Books, 2006); Babak Rahimi, “The politics of the internet in Iran”, in Mehdi Semati (ed.), Media, Culture and Society in Iran. Living with Globalization and the Islamic State (London: Routledge, 2008), 37-55.
  • [6]
    Bruce Elting et al., Mapping the Arabic Blogosphere. Politics, Culture, and Dissent, Internet and Democracy Case Studies (Cambridge: Harvard University Press, Berkman Center for Internet and Society, 2009).
  • [7]
    Creighton Powell Davis, “The internet as a source of political change in Egypt and Saudi Arabia”, Al Noor, 1(1), 2008, 32-41.
  • [8]
    Stephen O’Leary, “Falun Gong and the internet”, Online Journalism Review, 2000, accessed on 13 July 2010 on <>.
  • [9]
    John Bray, “Tibet, democracy and the internet bazaar”, Democratization, 7(1), 2000, 157-73.
  • [10]
    On this subject, the most complete overview is by Ronald Deibert et al., Access Controlled (Cambridge: MIT Press, 2010). The website “Opennet Initiative” gives an analysis by country.Online
  • [11]
    Geoffry Taubman, “A not-so World Wide Web: The internet, China and the challenges to nondemocratic rule”, Political Communication, 15(2), 1998, 255-73.
  • [12]
    Christopher Hughes, “Pourquoi Internet ne démocratisera pas la Chine”, Critique internationale, 15, 2002, 85-104.Online
  • [13]
    Ashley Esarey, Xiao Qiang, “Political expression in the Chinese blogosphere: below the radar”, Asian Survey, 48(5), 2008, 52-77.
  • [14]
    BJ McDuffie, “Civil society and the public sphere in Syria: How internet surveillance affects online participation”, Journal of International Politics and Economics, 27(2), 2009, 40-6.Online
  • [15]
    Yang Guobin, “The co-evolution of the internet and civil society in China”, Asian Survey, 43(3), 2003, 405-22.
  • [16]
    Anna Zyw, “Les avocats chinois, promoteurs d’un réseau juridique virtuel”, Hermès, 55, 2009, 65-70.Online
  • [17]
    Markku Lonkila, “The internet and anti-military activism in Russia”, Europe-Asia Studies, 60(7), 2008, 1125-49.Online
  • [18]
    Marc Lynch, “Blogging the new Arab public”, Arab Media & Society, 1, 2007, accessed 13 July 2010 on <>.
  • [19]
    David Bandurski, Martin Hala (eds), Investigative Journalism in China. Eight Cases in Chinese Watchdog Journalism (Hong Kong: Hong Kong University Press, 2010).
  • [20]
    Naila Hamdy, “Arab citizen journalism in action: challenging mainstream media, authorities and media laws”, Westminster Papers in Communication and Culture, 6(1), 2009, 92-112.
  • [21]
    Floriana Fossato, “Discussion: is Runet the last adaptation tool?”, The Russian Cyberspace Journal, 1(1), 2009, <>.
  • [22]
    Tai Zixue, The Internet in China. Cyberspace and Civil Society (London: Routledge, 2006).
  • [23]
    Zhou Yongming, “Living on the cyber border: Minjian political writers in Chinese cyberspace”, Current Anthropology, 46(5), 2005, 779-803.
  • [24]
    Gabriel Weimann, Terror on the Internet. The New Arena, the New Challenges (Washington: US Institute of Peace Press, 2006).Online
  • [25]
    Wael Adhami, “The strategic importance of the internet for armed insurgent groups in modern warfare”, International Review of the Red Cross, 89 (868), 2007, 857-878.
  • [26]
    Peter Chroust, “Neo-Nazis and Taliban On-Line: Anti-Modern Political Movements and Modern Media”, Democratization, 7 (1), 2000, 102-118.
  • [27]
    Dominique Colomb, Médias et communication en Chine, Paris: L’Harmattan, 2008.Online
  • [28]
    Masserat Amir-Ebrahimi, “La jeunesse iranienne dans le miroir du blog”, Les Cahiers de l’Orient, 79, 2005, 43-56; Yangzi Sima, Peter Pugsley, “The rise of a ‘me culture’ in postsocialist China”, The International Communication Gazette, 72(3), 2010, 287-306; Ellen Rutten, “More than a poet? Why Russian writers didn’t blog on the 2008 elections”, The Russian Cyberspace Journal, 1(1), 2009, 25-30.
  • [29]
    Karsten Giese, “Virtual identity in the Chinese internet”, in Peter Ho, Randy Kluver, Kenneth C. C. Yang (eds), Asia Encounters the Internet (London: Routledge, 2003), 207-09.Online
  • [30]
    N. Alavi, We are Iran; Henrike Schmidt, Katy Teubener, Natalja Konradova (eds), Control + Shift. Public and Private Usages of the Russian Internet (Norderstedt: Books on Demand, 2006).
  • [31]
    Merlyna Lim, “From real to virtual (and back again). Civil society, public sphere, and the internet in Indonesia”, in P. Ho, R. Kluver, K. C. C. Yang (eds),, 113-28.
  • [32]
    Rebecca MacKinnon, “Flatter world and thicker walls? Blogs, censorship and civic discourse in China”, Public Choice, 134(1), 2008, 31-46.
  • [33]
    Hung Ching-Fu, “Public discourse and ‘virtual’ political participation in the PRC: the impact of the internet”, Issues & Studies, 39(4), 2003, 1-38.Online
  • [34]
    Johan Lagerkvist, “The rise of online public opinion in the People’s Republic of China”, China: An International Journal, 3(1), 2005, 119-30; Xiao Qiang, “The rise of online public opinion and its political impact”, presented at the Chinese Internet Research Conference, University of Hong Kong, 2008, accessed 14 March 2010 on <>.
  • [35]
    Min Jiang, “Authoritarian deliberation: public deliberation in China”, presented at the Annual Congress of the International Communication Association (ICA), Montreal, Canada, 2008; Yeslam Al Saggaf, “The online public sphere in the Arab world: the war in Iraq on the Al Arabiya website”, Journal of Computer-Mediated Communication, 12, 2006, 311-34.
  • [36]
    Liu Chang, “Les narcissiques et les mobs: deux styles extrêmes parmi les internautes chinois”, Hermès, 55, 2009, 47-54.
  • [37]
    Zhang Zhibin, Wang Zhanping, “Lun tu fa shijian wangluo yulun de yindao fangfa”, Qingbao tansuo [Information Research], 8, 2007, 47-9.
  • [38]
    Isabelle Thireau, Linshan Hua, “De l’épreuve publique à la reconnaissance d’un public: le scandale Sun Zhigang”, Politix, 3(71), 2005, 137-64.
  • [39]
    M. Lynch, “Blogging the new Arab public”.
  • [40]
    Here I draw on the work of Érik Neveu, “Médias, mouvements sociaux, espaces publics”, Réseaux, 98, 1999, 17-85.
  • [41]
    Yang Guobin, The Power of the Internet in China. Citizen Activism Online (New York: Columbia University Press, 2009), 12: “Part of the project of this book is to reveal the democratic aspirations of the common people in the ways in which they produce, receive, and respond to critical, contentious discourses on the Internet”.
  • [42]
    Erving Goffman, Les cadres de l’expérience (Paris: Minuit, 1991); Façons de parler (Paris: Minuit, 1987).
  • [43]
    Jean-Gabriel Contamin, “Analyse des cadres”, in Olivier Fillieule, Lilian Mathieu, Cécile Péchu (eds), Dictionnaire des mouvements sociaux (Paris: Presses de Sciences Po, 2009), 38.
  • [44]
    David A. Snow et al., “Frame alignment processes, micromobilization, and movement participation”, American Sociological Review, 51(4), 1986, 464-81.
  • [45]
    Robert Benford, “An insider’s critique of the social movement framing perspective”, Sociological Inquiry, 67(4), 1997, 409-30.
  • [46]
    Daniel Cefaï, Pourquoi se mobilise-t-on? Les théories de l’action collective (Paris: La Découverte, 2007): on framing, 557; on procedures, 563; on authority, 573-4.Online
  • [47]
    Erving Goffman, La mise en scène de la vie quotidienne (Paris: Minuit, 1973).
  • [48]
    Michel Dobry, Sociologie des crises politiques. La dynamique des mobilisations multisectorielles (Paris: Presses de Sciences Po, 3rd edn, 2009), 34.
  • [49]
    Albert O. Hirschman, Exit, Voice and Loyalty (Cambridge: Harvard University Press, 1970).
  • [50]
    M. Dobry, Sociologie des crises politiques…, 103, original emphasis.
  • [51]
    We find here the heritage of Étienne de La Boétie, Discours de la servitude volontaire (Paris: Gallimard, 2008), as well as that of Max Weber, Économie et société (Paris: Pocket, 1995), 285: “All real dominant relationships have at a minimum a desire to obey and by consequence, an interest, external or internal, to obey”.
  • [52]
    Béatrice Hibou, Anatomie politique de la domination (Paris: La Découverte, 2011), 23-26.
  • [53]
    Luc Boltanski, Laurent Thévenot, De la justification. Les économies de la grandeur (Paris: Gallimard, 1991).
  • [54]
    Dominique Cardon, Jean-Philippe Heurtin, Cyril Lemieux (eds), “Parler en public (2). Dispositifs contemporains”, Politix, 31(8), 1995.
  • [55]
    Dominique Cardon, Jean-Philippe Heurtin, “La critique en régime d’impuissance. Une lecture des indignations des auditeurs de France-Inter”, dans Bastien François, Érik Neveu (eds), Espaces publics mosaïques. Acteurs, arènes et rhétoriques des débats publics contemporains (Rennes: Presses Universitaires de Rennes, 1999), 85-119.
  • [56]
    For an excellent overview, see B. Hibou, Anatomie politique de la domination.
  • [57]
    For a review of methodological problems of online qualitative studies, see Christine Hine, Virtual ethnography (London: Sage, 2000).
  • [58]
    My own networks were deemed insufficient to create this sample: not only did they have a tendency to present me with people who did not correspond to the desired profiles, but the connections between respondents led them to significantly censor themselves. I chose the lesser of two evils.
  • [59]
    In 2007, the average salary for urban areas in Beijing was 45,823 RMB/yr., or 3,818 RMB/month (around 382 euros), and the annual revenue per resident was 21,988 RMB, or 1,832 RMB/month (around 183 euros). Source: National Bureau of Statistics of China, 2008 China Statistical Yearbook (Beijing: Publications of the National Bureau of Statistics of China, 2008), 327.
  • [60]
    Interviews took place in cafes or in the homes of the subjects (in particular in the case of Zhang) and were recorded. The study was presented as a study about the daily usage and habits of internet users: frequency of site visits, keeping a blog, centers of interest, choice of online activities, etc. This information was relatively factual, thus offering the opportunity to ask follow-up questions about motivations: why speak out, why talk about one subject more than another, where to do it, with whom? It was only subsequently that the most sensitive aspects were alluded to: what limits are put on speaking out, how are they explained or justified by the respondents? In certain cases, self-censorship in the interview was evident and could be analyzed in a performative manner. In the majority of cases, however, even the most sensitive questions were discussed rather openly: it’s not about criticizing the system, but giving the keys of interpretation to a foreign “student” who doesn’t understand anything !
  • [61]
    Today there are 457 million. Source: CNNIC, January 2011, <>.
  • [62]
    On the idea of “quality”, see Andrew Kipnis, “Suzhi: a keyword approach”, The China Quarterly, 186(1), 2006, 295-313. More generally, see David Goodman, The New Rich in China. Future Rulers, Present Lives (London: Routledge, 2008).
  • [63]
    Which is a classical model of legitimation in authoritarian regimes. See B. Hibou, Anatomie politique de la domination, 15ff; “Modernité et technocratisation”.
  • [64]
    On the appropriation of norms by dominated categories, see Lisa Rofel, Desiring China. Experiments in Neoliberalism, Sexuality, and Public Culture (Durham: Duke University Press, 2007).
  • [65]
    On this point, see Séverine Arsène, “Quand les internautes chinois prennent la parole: les nouveaux repères de l’identité”, Hermès, 55, 2009, 17-22.Online
  • [66]
    Professional two-year diploma undertaken after the end of secondary education.Online
  • [67]
    Equivalent of a bachelor’s degree, or four years of university study.
  • [68]
    Note that certain respondents “admit” that they indulge in online debates, but that they associate this with a sort of vice or temptation which they wouldn’t otherwise allow themselves.
  • [69]
    Jean-Noël Ferrié, Le régime de la civilité en Égypte. Public et réislamisation (Paris: Éditions du CNRS, 2004).
  • [70]
    E. Goffman, La mise en scène…
  • [71]
    Or “face-work”: Goffman explicitly uses this notion often referenced in China.
  • [72]
    Nina Eliasoph, Avoiding Politics. How Americans Produce Apathy in Everyday Life (Cambridge: Cambridge University Press, 1998).
  • [73]
    E. Goffman, Façons de parler…, specifically the chapter “La conférence”, 167ff.Online
  • [74]
    Michel de Certeau, L’invention du quotidien (Paris: Gallimard, 1990).
  • [75]
    James Scott, Domination and the Arts of Resistance. Hidden Transcripts (New Haven: Yale University Press, 1990).
  • [76]
    A. Esarey, X. Qiang, “Political expression in the blogosphere…”.
  • [77]
    Séverine Arsène, “La satire, ou la ringardisation de la censure sur le web chinois”, Le Kiosque, 2010, <>.
  • [78]
    This phenomenon was notably identified by Jason M. K. Lyall, “Pocket protests: rhetorical coercion and the micropolitics of collective action in semiauthoritarian regimes”, World Politics, 58(3), 2006, 378-412.
  • [79]
    Johan Lagerkvist, “In the crossfire of demands. Chinese news portals between propaganda and the public”, in Jens Damm, Simona Thomas (eds), Chinese Cyberspaces. Technological Changes and Political Effects (London: Routledge, 2006), 42-63; Lee Chin-Chuan, Power, Money, and Media. Communication Patterns and Bureaucratic Control in Cultural China (Evanston: Northwestern University Press, 2000).Online
  • [80]
    Elisabeth Noelle-Neumann, The Spiral of Silence. Public Opinion, Our Social Skin (Chicago: The University of Chicago Press, 1993).
  • [81]
    Which does not prevent individuals or groups from using these spaces of expression with limited audiences for activist ends, as many associations do. This is also not necessarily indicative of public opinion.
  • [82]
    Zhao Yuezhi, Communication in China. Political Economy, Power, and Conflict (Lanham: Rowman & Littlefield, 2008).
  • [83]
    S. Arsène, “La satire…”
  • [84]
    To draw a parallel with the Egyptian and Tunisian cases, we might even ask ourselves to what extent the total cutting off of telecommunication networks actually prompted public protests for this same reason.
  • [85]
    In effect, the encouragement of individual enrichment is the flipside of the progressive dismantling of most institutions which took charge of individuals (including their work units). Above all, the maintenance of hukou, an identity document which links each resident to a locality, creates a de facto inequality between rural and urban citizens, and deprives migrant workers of many rights. The doctrine that justified this increase in inequality by the fact that “some must be the first to get rich” is henceforth offset by new Party slogans about the necessity of creating a “harmonious society”, which proves that inequality is an issue at even the highest level of government.
  • [86]
    It would nevertheless be necessary to reproduce this survey with other groups in order to confront their points of view and shed more light on the framing of speaking out in other places.
  • [87]
    We might also envision other sources of division, such as generational conflict, in particular in countries which have recently experienced a demographic transition.
  • [88]
    The author would like to thank Marie-Laure Geoffray, Amélie Le Renard and the anonymous editors at RFSP, whose advice greatly contributed to improving this article.


This article explores the usefulness of the notion of ‘frames’ (Goffman) to analyze online mobilizations in authoritarian countries. As most of the existing research is based on the idea that citizens are opposed to the state, it cannot explain the relative weakness of political dissent online, apart from as a function of censorship. The notion of ‘frames’ enables one to solve this apparent paradox in a nuanced way. Indeed it highlights the participation of individuals in the definition and maintaining of social order, as well as such subtle phenomena as misunderstandings, double meanings and conflicts of interpretation, or the dramatic dimensions of mobilizations. This idea is tested through a series of in-depth interviews in China.

Séverine Arsène
Séverine Arsène holds a PhD in Political Science from Sciences Po Paris and is currently a postdoctoral fellow at the University of Georgetown in Washington D.C. (Yahoo! Fellow in residence). She is the author of a dissertation entitled Speaking out by Chinese internet Users, which has been published in revised form as Internet et politique en Chine (Paris: Karthala, 2011). Her work centers on cultural practices, speaking out and online protest movements (Université de Lille, UFR Infocom, rue Vincent Auriol, 5900 Roubaix )
Translated from French by 
Katherine Schnakenberg
This is the latest publication of the author on cairn.
Uploaded on on 03/03/2014
Distribution électronique pour Presses de Sciences Po © Presses de Sciences Po. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait