‘Firing cannons to kill mosquitoes’: Controlling ‘virtual streets’ and the ‘image of the state’ in Bangladesh

Abstract

This article examines the historical, social and political legacies of the Information and Communication Technology Act (ICT Act) (2006–2018, amended in 2013) and the Digital Security Act (DSA) (2018–) in the Bangladeshi state’s attempt to control the ‘virtual streets’ of Bangladesh. The application of ICT and DSA has become an increasingly visible and controversial means to provide the spectacle of a state that extends disciplinary power and governmentality into proliferating online spaces—akin to ‘firing cannons to kill mosquitoes’. We use the lens of Tim Mitchell’s structural-effect (1991, The American Political Science Review 85(1), 77–96) to understand the state beyond the frameworks of its salience or elusiveness, arguing that the criminalisation of online speech has enabled the creation of ‘digital vigilantes’ who are predominantly the powerful, the sycophants, and the multitude of attention seekers who are driven by their personal contestations and ambitions. The legal outcomes, however, have been more ambiguous and uncertain—but the effect is to produce fear as an ‘environment’ (Virilio 2012, The Administration of Fear. Cambridge, MA, The MIT Press) through frozen/suspended charges with the potential to be redeployed in different contexts.

I Introduction

On 18 November 2018, Mr Chowdhury—a self-proclaimed ‘die-hard supporter’ of the prime minister of Bangladesh—sent a legal notice to the cinema at the Jamuna Future Park shopping mall, which was screening the newly released biopic: Hasina: A Daughter’s Tale (Khan 2018). ‘Tale’ in the title was misspelt as ‘tail’ and was corrected later. Mr Chowdhury felt that the spelling mistake was humiliating and demanded the Blockbuster cinema correct the error and apologise publicly or face a $90 million (£70 million) lawsuit. In online exchanges, opinion was split about the lawsuit: ‘How should I react - laugh or cry? Tale becomes tail which is definitely defamatory’ said one Facebook user, while another responded by saying: ‘It’s just a typo, bro. Apparently, you are using cannons to fight mosquitoes’ (cited in BBC NEWS, 2018).

This incident highlights the prevalence of events/actions where individuals feel that the state has been humiliated, and they need to take steps to redress it in the context of the Section 57 of the Information and Communication Technology Act (ICT Act 2006). First enacted by the Bangladesh Nationalist Party (BNP) in 2006, the Act was amended and made more draconian by the Awami League (AL) government, culminating as the harsher Digital Security Act (DSA) passed on 8 October 2018 under the framework of Digital Bangladesh. ¹ The aims of Section 57 are:

If any person deliberately publishes or transmits or causes to be published or transmitted in the website or in any other electronic form any material which is false and obscene and if anyone sees, hears or reads it having regard to all relevant circumstances, its effect is such as to influence the reader to become dishonest or corrupt, or causes to deteriorate or creates possibility to deteriorate law and order, prejudice the image of the state or person or causes to hurt or may hurt religious belief or instigate against any person or organisation, then this activity will be regarded as an offence. ²

In short, section 57 criminalises electronic content, which is deemed to be fake and obscene, defamatory, hurts religious sentiments (dhormiyo onubhuti te aghat kora), could contribute to the deterioration of law and order, could ‘instigate’ another person to perform harmful acts or could tarnish the ‘image’ of the state (rashtrer bhabmurti khunno kora). Both ICT and now DSA have been invoked in numerous events from 2006 to 2019. Rights activists and journalists have been critical of section 57 from the initial stages of its development; experts say the provision goes against people’s right to freedom of expression and free speech, and it contains vague wording, allowing its misuse against journalists and social media users. In the case of the misspelling of the film, the DSA was not invoked, as it was a mistake, but a lawsuit was threatened.

The phrase mosha marte kaman daga—firing cannons to kill mosquitoes—refers to a disproportionate response to a problem. Such an overblown response was also at work against the protests relating to the road safety movement (Schulz and Kuttig, ‘Introduction’ of this issue) in July–August 2018. Under the ICT Act, 86 people from over Bangladesh (including photographer and activist Shahidul Alam, leader of the student’s federation Maruf, University student Asif, actor Nawshaba and many others) were arrested and remanded between 3 and 15 August, 2018, in relation to the road safety movement.

On 29 July 2018, two students, Abdul Karim and Dia Khanom Mim, of Shaheed Ramiz Uddin Cantonment School, were run down by a private bus in Dhaka (Lacy 2018). These road deaths were just two among the many that occur on a regular basis and are usually seen as one of the hazards of travelling in Bangladesh by road (Prothom Alo 2018). But this incident brought into focus the corruption that is viewed to be rife in the transportation industry in Bangladesh (see Kuttig in this issue) and the irresponsibility of some drivers. The killing of the two students galvanised school children to take to the streets, the road safety protests highlighting an everyday problem that affects all sectors of society in Bangladesh. Students carried banners in their peaceful protests declaring: ‘We don’t want Digital Bangladesh. We want SAFE Bangladesh’. In other words: ‘We don’t want to be promised ambitious visions of a secure and developed future for Bangladesh if the most basic forms of everyday security and safety in the present are neglected or ignored’.

This article seeks to examine the Bangladeshi state schemes to make its population ‘legible’ (Scott 1999), its implementation of the Information and Communication Technology Act (ICT Act) (2005/2006–2018, amended in 2013) and its development into the Digital Security Act (DSA) on 8 October 2018 under the framework of digital Bangladesh. We explore the decade-long social and political implications of these new techniques of governmentality on social media (as outlined in the Introduction, in this article) and political practices in Bangladesh. We argue that the criminalisation of online speech has enabled the creation of digital vigilantes who are predominantly the already ‘powerful’ along with a vast multitude of sycophants and attention seekers who may not necessarily have much power but are motivated by their personal contestations and aspirations of being recognised and rewarded for providing services in terms of spotting and highlighting dissident or critical voices, which might also serve their own local rivalries. In these instances, their personal ambition is the critical driver rather than any enduring loyalty to the political figures.

The spectacle of an all-powerful digital state attempts to silence discussion of the more mundane social and political power relations that produce and maintain a multitude of unrelenting and pervasive everyday insecurities, inequalities and injustices. There is a need to think through the concept of digital vigilantes and the way we theorise their relationship to the Bangladeshi state and in other similar contexts like in the spectacular, non-spectacular and everyday violence and vigilance (online and offline) used to shut down the protests relating to the Citizen Amendment Bill in India from November 2019. Hence, instead of the state being ‘blurred’ (Gupta 1995), or the debates over characterisations of the Bangladeshi state as a ‘patriarchy’ (Hassan 2006), or a patron state (Lewis 2011) or having a ‘party–state effect’ (Schulz this issue), we examine what Timothy Mitchell (1991) describes as the ‘state-effect’ that is deployed to produce itself as a ubiquitous source of disciplinary power, as pervasive and unrelenting as anything imagined to exist in ‘surveillance societies’ or in ‘surveillance capitalism’ (Zuboff 2019).

These highly visible spectacles often become far more ambiguous and uncertain as they unfold through digital legal mechanisms and are implemented by state agents in a haphazard and unpredictable way. The article explores how laws/rules laid out by legislators in state institutions both express and generate anxieties, producing fear as an environment (Virilio 2012). We also show that the attempts to curb online criticism of the state and individuals highlight the fear authorities have about the iterability of online speech. In the first and second sections of the article, we examine the arrests in August 2018, followed by an analysis of the manifestations of these digital laws. In the third section, we examine the historical legacies of ICT/DSA evident in Section 295A since 2003. While new forms of digital governmentality, bureaucracy and surveillance are pivotal in creating the imaginations of an omnipresent state, we also want to argue that the criminalisation of online speech highlights the vulnerability of the seemingly omniscient state.

The article is based on ethnographic research and interviews carried out among middle-class activists and bloggers in Dhaka and beyond from 2003 to 2008 on Article 295A and again from August 2018 to October 2019 on ICT Act and DSA along with the examination of online reports, blogs and press clips. Respondents were selected from among those who were willing to speak on the issue of 295A, ICT Act, DSA and were based on a snowballing effect on the basis of suggestions from those willing to engage. As a result, the sample frame ended up being that of left-liberal activists and bloggers. At the same time, it is important to explore the potential ‘pitfalls’ of doing ethnography on online offences (as constructed by the state) as this phenomenon itself has impact on the forms of materialisations of the state and social media. In many of the accounts of the cases under ICT Act/DSA, which are occurring all across Bangladesh, the reportage is quite standardised and does not include details that can enrich ethnographic analysis. As a result, cartoons, images and films are mostly circulated in WhatsApp groups and not shared publicly with the fear of defaming the state or individuals.

Carrying out cyberethnography (Fernback 1999) on these digital laws and their manifestations has allowed us to focus on this crisis, posed new epistemological challenges and reconfigured the idea of the ‘field’ through various lateral networks within and outside Bangladesh. This method allowed us to engage and identify various respondents through online interactions whom we followed up with offline engagements, which are central to cyberethnography. Our personal experience of events of August 2018, which we outline in the next section, and our long-term ethnographic engagements (Lacy 2018; Mookherjee 2015) ³ allowed us to look back historically, diachronically, analytically and contextually from accounts which might seem to be ‘anecdotal’ and ‘journalistic’.

II ‘Axing one’s own feet’: The arrests in August 2018

Policing streets, motorways and cities remains one of the most powerful illustrations of a state’s competency and ability to control territory and provide safety and security (Virilio 2006). As circulation in Dhaka city was slowed down in July 2018 (more than the usual janjot/jams), as young people took to the streets peacefully for many days, anxiety about the state’s ability to ‘keep things moving’ and control the potential for urban disruption surfaced in relation to the protests on road crashes. The idiom ‘firing cannons to kill mosquitoes’ materialised when disproportionate force was exercised by motorcycle-helmeted assailants (suspected to be ruling parties’ youth wing) who—in cahoots with the state security forces—attacked some of the protestors, photographers and journalists. These incidents were filmed, posted online—on the virtual ‘streets’—shared and widely circulated through the citizens’ own ‘vision machines’: mobile phone messaging and Facebook live posts.

Vision, Virilio argues, is revolutionised in the modern age—in particular, by the innovations that occur in the military realm and the technologies that seek to see the battlefield/enemy better (Virilio 2006). While the possibilities of surveillance and vision are transformed by military (and policing) desires, the individual can record and document events in the city with their own ‘vision machines’. The renowned photographer and activist Shahidul Alam took to the streets of his city on his bicycle with such a vision machine to visually record, document and report (via live Facebook posts) what was going on. He tweeted a picture of his camera that was smashed as he moved around the city, by armed (with sticks and machetes) helmeted assailants. As more of the videos of the protests and attacks by helmeted assailants circulated online, there emerged intense governmental unease and paranoia about the ‘virtual streets’ of social media. A digital fog descended over Internet access, making it harder to circulate images and information. ⁴

We were watching the events of Dhaka online from Kolkata as we were due to travel to Bangladesh for a research trip in August 2018. After getting assurances from friends in Dhaka, we bought our tickets to Dhaka in the afternoon of 5 August 2018 to fly out on 6 August afternoon. At 10.30 p.m. on 5 August, we saw posts online that Shahidul Alam was abducted from his home by 30–35 ‘plain clothes’ police. This occurred after Alam had posted live images of the attack on students on Facebook and had given an interview to Al Jazeera on the political conditions in Bangladesh. With elections on the horizon, his comments clearly irritated key politicians in the administration. Rahnuma Ahmed, Shahidul’s partner and a fellow anthropologist, has been a long-term friend and so the news about Shahidul was extremely distressing. As we landed in Dhaka, we heard that along with the government crackdown on peaceful protests by school children, the police had arrested 22 university students who were protesting in support of the movement. They were also remanded and denied bail for several weeks. Shahidul had not been made to ‘disappear’ (goom) ⁵ but was detained by the DB—the detective branch—of Bangladesh in connection with allegations of having committed an offence under section 57 of ICT. A DB officer filed the case against Alam as he was deemed to have expressed defamatory, false news, tarnishing the image of the state in his Al Jazeera interview.

After being picked up, and before being remanded, Alam had allegedly been tortured. He was first presented in court on 6 August barefooted, his body held by police officers. After the legal hearings, he was sent to jail under the ICT. His bail hearing was set for 11 September, but he was refused bail five times without any reasons being given. This incident shocked civil society in Bangladesh. A public discourse emerged in government-affiliated media and among civil society intellectuals that deemed him to be an ‘agent’ of various countries and an anti-liberation element, as one of his uncles—his maternal grandmother’s brother—was a collaborator during the war of 1971. What remained unaddressed in this discourse is that his other uncle was one of the founding members of AL (Ahasan 2018). While some started making demands that Alam should be released, most of the buddhijibis (intellectuals), usually regarded as the voice of conscience, remained mute. Caricatures were circulating in WhatsApp groups of the State authorities holding up Alam as an example and telling a group of silent, huddled familiar buddhijibis: buddhi maraile bhainga dimu (if you use your intelligence, I will break you). There were rumours of videos being released through WhatsApp and Facebook, implicating other prominent people in the arts, publishing and journalism—which might have been another reason for only a few to raise their voices in protest. ⁶

There was, however, a quick dissemination of this news in the international press, and it grew into a global campaign, involving international organisations and individuals (no governments however) ⁷ seeking justice for Shahidul. These organisations included the United Nations (UN) Human Rights Council, a large group of Nobel laureates, PEN International, Human Rights Watch, Professor Gayatri Chakravorty Spivak, various photographers, writers, academics and cultural activists from across the world and South Asia. Amnesty International declared him to be a Prisoner of Conscience (Amnesty International UK 2018, 7 August). In spite—and maybe because—of this international movement to demand the release of Shahidul, the Bangladeshi government remained resolute to keep Shahidul in jail and finally released him only on 20 November 2018, just 1 month before the controversial elections on 30 December 2018. ⁸

It is important to reflect on the paradoxical nature of this phenomenon. The image of the state is believed to be tarnished in the face of any non-adulatory international publicity. The imposition of bans, pre-emptive censorship, abductions, deterrent incarcerations also legitimises and strengthens the negative reputation seemingly tarnished through such publicity. At the same time, as the state attempts to produce a spectacle of increasingly granular and local power, surveillance and policing, an uncontrollable global spectacle emerges of a state that appears increasingly unstable and unpredictable, driven by individualised vindictiveness that overwhelms other geo-political strategic considerations, a state obsessed with the harm done to the ‘image of the state’ but now publicly self-harming.

The attack and arrest of students and Alam by the government in the run up to the December elections of 2018 was surprising. Some felt that with this incident, the state ‘was axing its own feet’—carrying out an action, which is harmful to oneself. Not only had the government fulfilled its promise to set up the International Crimes Tribunal (albeit controversial but considered as significant by a large proportion of the electorate in Bangladesh), ⁹ it had also executed well-known collaborators of the war of 1971 (which had predominantly popular support in the country), talked convincingly of development for the country (as it aims to be a middle-income country by its 50th anniversary of independence in 2021) and had been lauded internationally for giving shelter to the thousands of Rohingya refugees fleeing persecution by the Myanmar government. At the same time, various questions were being raised about the government in relation to the democratic deficit, human rights violations relating to disappearances and repression of opposition party leaders and protests as shown through the protest movements. The elaboration on the various cases filed under ICT/DSA will highlight the operation of the digital laws.

III Manifestations of digital laws

Digital life in Bangladesh started in 1996 with the advent of the Internet in the country. The usage of the technology had risen by 517.3 per cent around 2012. ¹⁰ In 2015, the nation had over 120 million mobile subscribers and 43 million Internet subscribers (Rahman 2015). As Bangla fonts and keyboards came to be developed in 2005, community blogging picked up in Bangladesh, and bloggers became a growing presence. In 2006, the Information and Communication Technology Act was enacted by the BNP–Jamaat government. The maximum punishment for offences under section 57 was a 10-year imprisonment and a fine of BDT 10 million. Police had to seek permission from the authorities concerned to file a case and arrest any person under the law. Between 2006, when the law was first enacted by the BNP government, and 2013, when it was amended by the AL government, police data show that while there were 426 complaints, only a few resulted in arrests or prosecution (Bergman 2019). The huge rise in cases occurred after 2013. The laws were changed to make it much easier to arrest without warrant on any of the six broad charges of Section 57 just after the filing of a case under the ICT Act; to make it more difficult to get bail; and to increase penalties to draconian levels (minimum of 7 years and a maximum of 10/14 years). ¹¹

During the British colonial period, a provision was incorporated into the Criminal Procedure Code (CrPC), empowering courts to issue direct arrest warrant against persons, including journalists, writers and publishers of any books or newspapers, if they wrote or said anything defamatory. Under the Penal Code of 1860, one could be punished with imprisonment for a maximum of 2 years, or with a fine, or with both, for defaming others. These colonial-era penal codes have been retained in most South Asian countries, including Bangladesh, only to be invoked at crucial junctures. Here, the amended, more punitive version, has criminalised online speech and a range of acts. The trajectory of the ICT/DSA cases until September 2019 would also help us comprehend the implications of the possibility for Bangladeshi citizens to criminalise a range of online speech and acts, to ‘legally’ charge/remand/fine fellow citizens for ‘posting’, ‘liking’ and ‘sharing’ opinions on social media, which might tarnish the image of the state/person.

In 2011, charges of defamation were brought by a local AL politician in Mymensingh against the microfinance pioneer Muhammad Yunus over comments he made about Bangladeshi politicians in a 2007 media interview. Yunus had to appear in court, pay a fine and he soon was removed from Grameen Bank, and investigations into his financial matters were started by the government authorities. He is said to have angered the government for backing the military-backed interim government (2006–2008) and trying to set up his own political party in 2008. The draconian amendment of the ICT law in October 2013 followed soon after the Shahbagh protests, demanding the controversial death penalty of collaborators of the war of 1971 as the latter had enjoyed political impunity for the last 40 years.

These death penalties also resulted in the resurrection and strengthening of Hefazat e Islam, a group based in a qawmi madrassa (orthodox Islamic school) that has been holding counterdemonstrations and rallies in Bangladesh, demanding the execution of so-called atheist bloggers and stopping the Shahbagh protests. Many of the participants at these rallies were supposedly young, male and poor students of the madrassas. On 5 May 2013, Hefazat held a rally in the commercial centre of Dhaka. When a human rights organisation published reports on these demonstrators being killed by security forces, its director and secretary were charged under ICT and denied bail.

The debate and demand for the repeal of Section 57 intensified, following the arrest of journalist Probir Sikdar in 2015. The veteran journalist was arrested and sent to jail after he posted a status on Facebook, expressing fear that his life was in danger and that an AL minister, a businessman and a fugitive war criminal, should be held responsible if he were killed. Following the post, an AL leader filed a criminal case against Probir in Faridpur under Section 57 of the ICT act for ‘tarnishing the image’ of the minister. In September 2015, Mohan Kumar Mondal and his colleague were arrested in Satkhira after an AL activist filed a case alleging that a Facebook post by Mondal had hurt the religious beliefs of Muslims. The post criticised Saudi Arabia’s security arrangements during the Haj that led to a deathly stampede killing hundreds. The men were detained for 2 months before the Cyber Tribunal finally granted them bail on 29 November 2015.

Many of those who complained to the police are explicitly linked to political parties. In August 2016, Rashedul Isam Raju, General Secretary of AL’s student wing at Rajshahi University complained to the police about the Facebook posts supposedly made by Dilip Roy, a student from a leftist opposition party. Roy remained in custody for 3 months before the High Court granted him bail. Raju also organised a Facebook Page event (Kuttig and Suykens forthcoming) to elaborate on the charge he brought against Roy. The editor of Daily Star, Mahfuz Anam was charged with defamation and treason in more than 80 cases filed by AL workers in 2016, with damages sought exceeding US$8 billion. They were filed in various lower courts across Bangladesh, forcing him to run around the country seeking bail, and he has had to obtain periodic ‘stays’ on them ever since. This means these charges could be revived if Anam was felt to exercise too much press freedom. Prothom Alo, another critical national newspaper, has faced more than 100 criminal charges and, along with Daily Star, is being deliberately starved of advertising revenue with advertisers pressured to not support them (Reuters 2018, 13 December).

ICT charges were brought on people from all walks of life. In April 2017, Monirul Islam, a rubber plantation worker from Srimongol, was arrested for ‘defaming the country’s prime minister’ by liking and sharing a Facebook post that included a caricature of the prime minister. A case was filed against him for condemning Islam under Section 57 of the Information and Communication Technology Act, describing his actions as a ‘betrayal to the country’. Islam was denied bail and detained for over 3 months before the High Court issued papers for his release. The author of the original Facebook post went into hiding for fear of his own arrest (Freize 2018, 14 August).

In May 2017, Nipun Chandra Das and Sanjoy Banerjee, local correspondent of The Asian Age and Dainik Janakantha were sued by the secretary of the local AL, following their reports on the torture of a Hindu woman by a local gang who attempted to grab her land (Saha 2017: 26). In July 2018, a professor of Sociology in Chittagong University expressed his support for the student’s road safety and quota reform movements and protested against the assaults on these students by ruling party members on social media. On 13 July, a day after the government told parliament it was impossible to accede to some student demands, he posted two words on Facebook ‘Hasi Na’ (meaning ‘not laughing’). Soon AL activists filed a complaint against him under the ICT Act, accusing him of defaming the Prime Minister Sheikh Hasina. The academic was arrested and jailed for more than a month before he got bail, and if he loses the case, he faces up to 14 years in prison.

On the night of 4 August 2018, police arrested Nusrat Jahan Sonia, a 25-year-old primary school teacher, in a rural area south of Dhaka. They said she had ‘written something anti-state’, arrested her and seized her mobile phone and laptop. Seven months pregnant, she was held for nearly 2 weeks for ‘spreading rumours’ under section 57 because she shared a Facebook post that appealed for peace during the August 2018 road safety movement. According to a family member, she has been suspended from her job at a government school.

By exploring the nature of the cases filed, the convictions, the process of enforcement or. more specifically, the non-conviction of these laws, status of the cases post-acquittal and the reasons for the charges being brought, we get a vivid idea of state action in this domain. Once a Cyber Tribunal dedicated to dealing with offences under the ICT Act was established, the number of complaints to the police, arrests and prosecutions soared. Hundreds, including several journalists, have been accused under section 57 for criticising the government, political leaders and others. The Daily Star has found that 11 cases have been filed against 21 journalists, since March 2017, related to news reports (Adhikary 2017). A total of 260 cases were filed between March 2017 and the first week of June 2017 (Adhikary 2017). In the first 3 months of 2018, police submitted 282 charge sheets with Cyber Tribunal officials and 1,271 charge sheets between 2013 and April 2018. Most involve charges of criticism of the government, defamation or offending religious sentiments, tarnishing the image of the state, while the rest are allegations against men publishing intimate photographs of women without their consent.

On 8 April 2018, after student protests, a police officer filed a complaint referring to 43 ‘provocative’ Facebook posts, which ‘many have liked and commented on’ that has, as a result, ‘created a situation which could potentially harm society and create chaos’(ibid.). Similarly, in August 2018, 28 twitter handles and accounts were charged. It is important to note that for all the visibility surrounding these cases, there are relatively few convictions resulting from prosecutions. In September 2017, Md Nazrul Islam Shamim, special public prosecutor of the Cyber Tribunal, told The Dhaka Tribune that 65–70 per cent of cases filed under section 57 cannot be proved in court. ‘Some cases are totally fabricated and are filed to harass people’, he said. Shamim’s comment is reflective of the much broader issue of ‘false cases’ that has been used as an instrument by both the state and influential individuals. There are also concerns that some of the accused—like the aforementioned respondent Mondal—had their accounts ‘hacked’.

In terms of the enforcement and, in fact, non-conviction of the laws, in the first 3 months of 2018, eight of the nine cases where trials were concluded were acquitted. This phenomenon of few convictions of the ICT/DSA cases and the concomitant harassment implicit in the process is similar to the ‘logic of non-enforcement’ described by Berger (this issue) or, more specifically, the logic of non-conviction. His article investigates the ways in which state and non-state law becomes intertwined in practices of conflict resolution in rural Chittagong. According to his logic of non-enforcement, people in rural Bangladesh frequently appeal to state courts—but not to get binding and enforceable verdicts. The threat of the state courts is supposed to alter the dynamics of non-state justice institutions in their favour. In the ICT cases, it is evident that gaining conviction in court is not necessarily the critical objective. The very fact of being charged results in a chain reaction of harassment and social sanctions, some of which may be held in reserve for suitable future occasions. In these instances, the law is being enforced, and the underlying objectives are being attained in effect, even if there is no conviction. Accordingly, the crucial point would appear to be non-conviction rather than non-enforcement.

It is important to note that similar to the ICT/DSA cases, which are primarily brought in by the powerful, rarely can the poor bring about a shalish ¹² in rural Bangladesh. We are comparing the phenomenon of shalish with the manifestation of these Acts as highlighted in Berger’s article but not necessarily of their legal and social consequences. So, in the case of ICT/DSA cases, while there are few convictions, the process of remand and arrest seeks to suppress dissent and carries social consequences. Individuals face the constant threat of being arrested, held in pretrial detention, subjected to expensive criminal trials, fines and imprisonment, as well as the social stigma associated with having a criminal record. In addition, such treatment may chill free speech. ‘The government has reassured the public of their commitment to freedom of speech’, the Dhaka Tribune, deemed to be a pro-government newspaper, said in a September 2017 editorial. ‘Then why does section 57 continue to be a tool of harassment?’ Above all, even after individuals are acquitted on ICT Act/DSA, charges are not dropped with the potential to be revived in case the individual does not conform to being non-dissenting. But, here, it is the accusation that is fundamental to the spectacle of state power over the individual in the online sphere.

Exploring the various cases, it seems the main intent with which it is being used is to threaten and frighten people into inaction and silence. As Saha (2017: 26) notes: the legal tool of ICT was actively used by a dominant political party as a threat to file cases against its opposition as a means of harassment through the paying of fines and imprisonment. Powerful leaders have also used it to stifle criticism. In instances where stories of corruption and self-aggrandisement have been leaked, the ICT has been invoked to stop these accounts from circulating.

Most of the time, section 57 has been used by the established, rich and powerful (AL politicians and their acolytes, judges, companies, local chairmen, priests, madrassah heads) at local and national levels with contacts among the police to stop investigations by journalists (quite a few instances of Hindu journalists), editors of online portals, of accusations of corruption and attempts to highlight the misuse of power by the powerful. Social media users have also been booked under section 57 for publishing online newspaper reports or liking/sharing others posts. ¹³ In August 2018, activists talked about fear as an environment as they tried to stop leaving a digital trace by not liking, sharing or commenting on online posts about social and political issues, leaders and influential people (Shakhawat 2017). In 2019, a similar restraint was exercised by Indians who are critical of their government’s actions of revoking section 370 in Jammu and Kashmir.

As Kuttig and Sukyens (2020) have illustrated, student politics—which historically has been significant in key political events—in Bangladesh is used as a means for developing careers in business and government and as a resource for political parties for a range of tasks, from large-scale political events to hartals. The ‘visibility’ that students cultivate and perform via social media might extend through to deploying new laws against opponents and even creating a Facebook event page to publicise to the local and national echelons of the party how they have used the ICT law against those who are ‘tarnishing the image of the state’. Here, workers may be seeking visibility with their employers in the same way students seek visibility with their parties. Even the simple and mundane act of liking or commenting on a post becomes a potentially self-destructive act, an act that ‘brings’ the state into one’s most intimate environments. The tragic murder of engineering student Abrar Fahad in October 2019 brought home the extreme actions being taken by the digital vigilantes (Daily Star 2019, 7 October). Abrar was ‘grilled’ by AL student activists for his critical Facebook post on the India–Bangladesh water deal, was suspected of being involved with the Islamicist student group Shibir and thereafter beaten to death, which led to widespread protests in Bangladeshi universities.

State practices, here, set out to monitor and transform behaviour through what Foucault (2004) would describe as governmentality, the internalisation of power into one’s physical acts, movements, instincts and thought processes (the gentle click of a ‘like’). At the same time, the threat of the circulation of WhatsApp rumours and the exemplary verdicts set up with the remands of Alam and others serves as an effective deterrent for civil society intellectuals who are usually considered to be the voices of conscience in Bangladesh. The social and political phenomena emerging from the exercise of the ICT/DSA show that through this legal instrument, the state is not only a source of surveillance but enables influential citizens to feel defamed and to bring legal complaints against those who have insulted them. So, here, the role of the sovereign is carried out locally and exemplifies the sense of governmentality.

Along with the existing powerful coteries and formal agents of the state invoking this law, certain groups of self-seeking ‘citizens’ and sycophants are involved in surveillance and making accusations as local and digital vigilantes. So not only do the locally powerful use the ICT and act as the eyes of the state, it also enables powerful citizens and their acolytes to build up a local politics of patronage and fear. The threat of invoking ICT’s infamous clause 57 legitimises the party locally and nationally, while, at the same time, enables them to try to gain attention of those more powerful than them. This phenomenon illustrates Mitchell’s conception of the state effect as the state–society boundaries, which are internally deemed to be distinct, come to be based on complex power relations and power aggrandisement. The technical innovations of the ICT make the Bangladeshi state external to these local/individualised interventions to police online activities and yet through which the state effect is conjured through sycophancy.

After the detentions and imprisonments in August 2018, the DSA was enacted in October 2018. Bangladesh’s journalists are concerned about Section 32 of this act, which will treat the use of secret recordings to expose corruption and other crimes as espionage, arguing that it will restrict investigative journalism and muzzle media freedom. State representatives justify the passing of the Digital Security Act (Wazed 2019) to protect citizen data and privacy, to safeguard against false or inflammatory digital publications that incite violence ¹⁴ and to prevent anyone spreading misinformation about the liberation war and its 3 million dead. As a result, in July 2019, a Digital Security Agency has been set up with cooperation from Facebook, and the government will intervene in any content on social media from September 2019, to use ‘collected technology’ to collect data. ¹⁵

The state in upholding the DSA for its citizens is also a guarantor for not hurting the citizen’s feelings of defamation and comes to encompass those defamed because of investigations into their corruption and those offended by blasphemous rumours, images and satire. Further, it links up these state practices to the practices of other states by invoking the Holocaust denial framework to justify clauses against the dissemination of misinformation relating to the liberation war. The following section explores how the clauses of ‘hurting religious sentiments’ and ‘prejudicing the image of the state’ have been used prior to this contemporary predicament.

IV Policing of onubhuti (feelings) and bhabmurti (image): 2003–2018

The draconian nature of Bangladesh’s digital laws has its precursors in Article 295A (2003), which reinforced the presence of the state in the everyday life of Bangladeshi citizens through ICT Act/DSA. It is in this context that we need to identify the contours of how the state is seen (Scott 1999) by the state itself and by others who agree that the bhab (any state of mind or body, way of thinking or feeling,) murti (image) is made khunno (tarnished). This tarnishing of the image becomes the ground for state action on the basis of which books and films are banned (in the case of Taslima Nasreen elaborated below) and citizens are exiled and imprisoned.

The most well-known and exemplary use of the phrase dhormiyo onubhutite aghat, (hurting religious sentiments) has been used by both the Bangladeshi and West Bengal (of India) governments to ban the books and writings of a Bangladeshi physician and author—Taslima Nasrin—known for her militant views against gendered inequality. A fatwa was first issued against her, following an interview in The Statesman (9 May 1994), in which she allegedly demanded the revision of the Koran. This was preceded by the publication of her novel Lojja (Shame, Nasrin, 1993), which portrays the nightmare which befell a Hindu (Duttas) family in Bangladesh in the face of a backlash on minority Hindu communities in Bangladesh. This backlash occurred in response to the demolition of the Babri Masjid at Ayodhya in India on 6 December 1992 and the consequent killing of Muslims in India. Lojja ends with a poignant note as the Duttas leave Bangladesh for India. Lojja was banned in Bangladesh in 1993, and, subsequently, four more of Nasreen’s books were banned by the Government of Bangladesh and the Communist government of West Bengal, India, by invoking Section 295A of the Penal Code as the books were considered to ‘hurt religious sentiments’. The prosecutions under 295A made Nasreen leave Bangladesh and seek exile in Sweden.

The notion of hurting religious sentiments—dhormiyo onubhutite aghat, is a reference to onubhuti—feelings or affective states—which seem to be central to these contestations. At the same time, this affective language coincides with allusions to Nasreen and her work by a set of intellectuals and activists who would say ‘let’s not talk about her’ or let out a visceral chi chi—an expression which stands in for shame and condemnation. The banning of these books by the Bangladesh and the West Bengal governments followed court injunctions brought by a set of intellectuals in both Bangladesh and West Bengal as Nasreen wrote about how they had sexually propositioned/harassed her. The books were, however, banned because they are deemed to be an affront to religious sentiments and are considered to be ‘pornographic’ and ‘with no literary value’ (Naoreen 2004: 75).

The bans on the first two autobiographies of Taslima Nasrin became co-terminus with the BNP government’s concern to penalise activities that tarnish the image of the state. From 2001 onwards, a series of incidents seemed to confirm this concern: both the attacks and charges of sedition brought against human rights activists, highlighting the rape of minorities in Bangladesh and the Indemnity Ordinance given to army officials involved in the ‘Operation Clean-Heart’, which was put into effect in January 2002, to curb the ‘law and order’ situation in Bangladesh leading to 44 deaths. It is in this terrifying context of actions taken by the government with regard to tarnishing the image of the state that a chilling, contagious atmosphere of self-censorship enveloped the left-liberal intellectuals, feminists, students, writers and journalists who primarily saw themselves as adversaries of the BNP government and its policies.

A decade later, in 2013, when bloggers were arrested by the AL government, they were also booked under the ICT Act of 2006 as they were deemed to have hurt religious sentiments. Here, the distressing effect on secular sentiment under an AL government is acutely felt by these secular bloggers according to whom legislation on religions sentiments is deemed to be against the spirit of the Liberation war. Referring to the role of surveillance carried out by the state through section 57 of the ICT Act, they refer to it as digital ojuhat not digital projukti (digital excuse not digital development).

The ICT Act/DSA enables the state to undertake surveillance and refine governmentality by extending the eyes of the state so that all citizens who are offended can be digital vigilantes on behalf of the state, creating an alliance of the defamed with the state at its helm. It also enables a process of repetitive and explicit censorship even though it is marked by unpredictability with the possibilities of the threat of ICT Act/DSA to the offenders within a culture of deterrence produced to deter future offenders. The ICT Act/DSA and its concomitant censorship also produce ‘subjects according to explicit and implicit norms’ and ‘the production of the subject has everything to do with the regulation of speech’ (Butler 1997: 133). Butler (ibid.: 129) refers to such action as a pre-emptive censor, which attempts to regulate the sphere of public discourse by institutionalising the norms that establish what ought properly to be included there. When censorship operates prior to speech, it constitutes the norms by which the speakable (here defamable) is differentiated from the unspeakable (undefamable).

Through the presence or the absence of posts, the activists’ alliance with the Shahidul case apparently became clear. In both Nasrin and Shahidul’s case, the legal charges were intertwined with one-sided rumours about their personal lives. Hence, the state practices through the ICT Act/DSA produced a public and a subject who is prescribed by the norms of non-defamation. In determining the digitally and virtually speakable discourse and its constitutive norm, we have the operation of Mitchell’s state effect, where the state is neither salient nor elusive. The state structure, instead, appears external to society though internally bound by complex power relations. Along with the control of the norms of the digital discourse of offendability, the state practices of the ICT Act/DSA also led to the interrogation of the subjectivity of Nasreen and Alam and an effective response towards the phenomenon.

V Conclusion: Fighting with one’s own shadow

The Bangladeshi state is a transnational, virtual phenomenon produced and represented in many sites, establishing the norms of defamation through an affective form of litigiousness. The state depends on affective and spectacular forms of publicity through the deployment of phrases such as ‘hurting religious sentiments’ and ‘tarnishing the image of the state’. Their historical trajectories precede the onset of section 57 of the ICT Act (2006–2018) and continue into the DSA from 2018.

We have shown in this article how digital laws laid out by legislators in state institutions both express and produce anxieties and also how through the phenomenon of digital vigilantes, these rules are implemented in a haphazard and unpredictable way. In fact, laws here work as tools for the logic of non-conviction but with huge social fallouts for those charged under ICT Act/DSA. The retention of colonial legal frameworks in the subcontinent allows them to be revived to discipline a population. The idea of tarnishing the image of the state is very new, at least in the law—which has moved from earlier notions of sedition, requiring actual steps to overthrow the government through to this idea of mere expression being sufficient to cause harm. As a result, the non-dismissed lawsuits against those charged under ICT Act/DSA enable a continuation of control and unease and serve as a spectacle of deterrence as well as sycophancy.

As we have shown, the legal outcomes have, on occasion, been ambiguous and uncertain, but the effect was to produce an environment of fear and strategies of deterrence that aim to control and contain political debate in the public sphere. In this sense, anxiety and concern about the social relations or networks of power in Bangladesh, the links between business and political/legal power and authority, become harder to establish in the face of a state that creates the performance of an omniscient and omnipresent state. Historian Afsan Choudhury notes: ‘Hate, insults, endless allegations and occasional physical confrontations have become the staple of Bangladesh’s political culture’ (Chowdhury 2018). In terms of the pressure placed on journalists, cases of legal harassment rose from 33 in 2013 to 169 in 2017, but the use of physical force and assault as a silencing tactic decreased from 173 in 2013 to 113 in 2017 (Article 19 2018). This might reflect a move from more material forms of violence and intimidation to subtle and legal attempts at control. Already there is a growing and conspicuous silence on public criticism of the government and online and social media-based criticism of the government has evidently declined since the passage of the DSA.

The preponderance of these digital laws has been intertwined with the sacrosanct discourse of the Liberation War. As scholars continuing to work on the war for over 2 decades in Bangladesh, we argue that 1971 has become a technology of rule for the state. The release of the docudrama (with which we started the article)—Hasina: A Daughter’s Tale—a month before the elections of December 2018, focuses on the assassination of Sheikh Mujibur Rehman and his entire family. The focus in this film on the effect of that assassination on the premier and the sacrifices made by the family on behalf of the nation through its leadership in the war of liberation is an illustration of the significance 1971 has for the current Bangladeshi state to rule the country. In the DSA, digital laws not only disallow the questioning of the contested enumerative community of 1971 (‘3 million dead’), it has also made it punitive to make any online criticism of Sheikh Mujib. Given the restrictions placed by the government-inflicted digital laws, one can comprehend 1971, and work on it can only be seen within this nationalist project. This makes ethnographic work on the state even more difficult today. An activist puts the paradox succinctly in a personal communication in October 2019:

Does Bengali nationalism get nullified as roots of secularism, equality just because they are used as autocratic weapons by the state? Do we throw the baby out with the bath water as I find apparent in most modern rights discourses? Or do we explore creative ways of going back to the roots and reinventing democratic spaces?

The implementation of the ICT Act and the campaigns surrounding it have influenced the younger generation who are attempting to reinvent these democratic spaces. We were told by University students in Dhaka how they decided not only to self-censor but also subversively screen Satyajit Ray’s film Hirok Rajar Deshe (Banerjee 2013) (In the Land of the Diamond King) or stage a play based on the film as part of college events in front of visiting ministers and as a way of critiquing the state practices linked to the ICT Act. A children’s film, a musical comedy, Hirok Rajar Deshe contains a subtle and powerful critique of inverted totalitarianism. Without any use of violence on its citizens, this kind of totalitarianism is based on the political apathy of its citizens and what the film terms mogoj dholai (brainwashing). According to activists and students, in the 1980s, when this film was shown in Bangladesh at the time of the anti-military demonstrations, the words of the songs in the film were drawn overnight on the walls of the city.

While new forms of digital governmentality, bureaucracy and surveillance are pivotal in creating the imaginations of an omnipresent state, we also want to argue that the deployment of the ICT Act and the DSA highlights the vulnerability of the seemingly omniscient state. The state’s paranoia about the digital age becomes focused on using the tools of the digital age as a technique of deterrence, both the individualisation of deterrence (focused on a key, prominent figure) and the magnification of this deterrence (compared to the earlier events involving Nasrin), marking everyday life and behaviours and through the production of a national and international spectacle. As the technologies and techniques that appear to be able to challenge the state seem to multiply, the modalities of deterring the citizenry appear to proliferate and intensify, moving from an anxiety over what one can read or discuss through to what one can ‘like’ or ‘post’.

Through an obsession with tarnishing the image of the state, a new image of the state is produced: an accelerating, expanding and mutating machine that presents itself as omnipotent, omniscient and omnipresent, a machine that can make local, molecular events (such as an arrest or accusation) national or global, an element in a spectacle of deterrence; a machine that can make national events or policies feel increasingly local and personal, individualised or intimate, expanding and intensifying an environment of fear. What is alarming here is how new laws are being used to limit discussion of older, more ‘traditional’ problems (corruption, problems of governance, violations of rights, denial of democratic rights and space), and the invocation of ‘older’ problems (the hurting of religious sentiment, tarnishing the image of the state) is being deployed in attempts to control the new technologies that the state is both seduced by and terrified of—the poison and cure of Digital Bangladesh—while it continues to fight with its own shadow in its search for adversaries.