India: visible and invisible censorship
I find it a useful thought experiment to think of the number of Indians who have published in a newspaper or have had their voice or image broadcast over radio or television since India’s independence in 1947, and to compare that with the number of Indians who have published on the Web since 1995, when India’s first public-access Internet service provider started functioning. The latter number is surely larger. The Internet, as anyone who has ever experienced the wonder of going online would know, is a very different communications platform from any that has existed before. The medium enables those who have access to it an unprecedented ability to directly share their thoughts with millions of others in an instant, even while it replicates many of the inequities of other media.
The various kinds of state, corporate, and societal regulations and impositions that existed in pre-digital times continue to exist, albeit they have changed, though not necessarily for better. In this chapter, I hope to show the regulatory architecture of digital censorship in India. In particular, through the examples of the Intermediary Guidelines Rules, the ham-handed curbs on SMS and web pages in August 2012, arrests under the IT Act, and websites blocked under copyright enforcement, I shall make the argument that the most important safeguard against censorship is visibility and that we are fast losing that feature. The examples I explore shall demonstrate that public reaction to a censorship law depends less on how damaging it is (seen as how much speech can be curbed without sufficient justification and due process of law) and more on how direct it is and how visible it is.
Brief Chronology of Direct State Censorship of the Internet in India
India has had censorship of the Internet since the middle of the nineteen nineties.1 At that time the only way of accessing the Internet was through Videsh Sanchar Nigam Limited (VSNL), the state monopoly internet service provider (ISP). During this period, access to websites of certain voice-over-IP (VoIP) providers (like Vocaltec, Net2Phone, etc.) was blocked alongside VoIP itself, leading to the first case filed on Internet censorship in Indian courts, in 1998.2 VSNL argued that it had the authority to block access to regulate Internet telephony and block access to VoIP provider websites under the Indian Telegraph Act, 1885.3 It is unclear what statutory powers it was using to block access to the website of the hacker collective Cult of the Dead Cow in 1998,4 or to block access to the website of one of Pakistan’s leading newspapers, Dawn, during the Kargil war between Pakistan and India in 1999,5 even though it denied having taken such an action.6 In 2000, while a VSNL employee initially admitted having blocked e-mails from and to the ‘Middle East Socialist Network’ (MESN) mailing list,7 in an affidavit to the court in the Arun Mehta case, VSNL denied ever having blocked access to the eGroups.com website (which hosted the archives of the MESN list), but noted that “in view of the problem of spamming on the internet, temporarily the e-mail operations of egroups.com was stopped,” and later restored.8
Since 2000, the licence — provided under the Indian Telegraph Act — that ISPs in India must enter into to provide Internet services includes clauses that require the ISP to take measures to prevent “objectionable” content and “anti-national activities”,9 and take down websites that unspecified “enforcement agencies” ask them to remove.10 The Indian Telegraph Act is still in force, and it is still unclear what provision in it empowers the government to block websites.
Information Technology Act and After
In 2000, the Information Technology Act (IT Act) was passed, primarily being a law derived from the UNCITRAL Model Law on Electronic Commerce. While it contained a provision criminalizing the electronic publication of obscene materials,11 it did not provide the government the power to block websites for obscenity, or for any other reason.12 However, in 2003, the Department of Information Technology issued an executive order, citing powers under section 67 (the provision on obscenity) and section 87 (the provision on subordinate legislation), empowering the newly-created Indian Computer Emergency Response Team (CERT-In)13 to block websites,14 even though the statute itself didn’t provide the government any such powers. Extraordinarily, the Indian government accepted as much in another gazette notification, that soon followed:
As already noted there is no explicit provision in the IT Act, 2000, for blocking of websites. In fact, blocking is taken to amount to censorship. Such blocking can be challenged if it amounts to restriction of freedom of speech and expression. But websites promoting hate content, slander or defamation of others, promoting gambling, promoting racism, violence and terrorism and other such material, in addition to promoting pornography, including child pornography, and violent sex can reasonably be blocked since all such websites may not claim constitutional right of free speech. Blocking of such websites may be equated to “balanced flow of information” and not censorship.15
This presented a novel idea in Indian freedom of expression jurisprudence which has traditionally had an expansive view on what constitutes speech,16 but then has at times been equally expansionary as to what kind of speech may be rightfully restricted.17 This interpretation by the Department of Information Technology seems to indicate that there are some speech that may not count as speech itself, rather than as speech that may rightfully be restricted. Given this, they state that they do not need statutory powers to engage in blocking of websites, since blocking of websites of a certain sort does not amount to ‘blocking’. As per government, it is not a matter of rightfully restricting speech — for which to be constitutionally valid, they would need statutory authority — but instead, it is a matter of promoting a “balanced flow of information”18 — for which, seemingly, executive powers seem to suffice.
The first notable action subsequent to these notifications was when CERT-In ordered Yahoo and all Indian ISPs to block access to a mailing list with around 160 members called “Kynhun” on Yahoo Groups,19 which was being used by the Hynniewtrep National Liberation Council, a little-known proscribed separatist group from Meghalaya, to publish a newsletter called the Voice. According to one commentator, that newsletter contained articles on “how the corrupt government is building non-existent roads and public utilities (and swallowing money in the process), how this minority is being victimized and such.”20 Lacking the technical capabilities of blocking a single group, multiple ISPs blocked web access to all of Yahoo Groups instead. This made it possible to keep receiving mails from that mailing list (and other mailing lists on Yahoo Groups), but prevented all web access to Yahoo Groups. In a matter of a few weeks, the excessive blocking was rectified without any public statements by either the government or the ISPs that over-blocked.
In 2004, the U.S.-based right-wing website HinduUnity.org was blocked by Indian ISPs on orders of the Mumbai police, though at least one ISP apparently refused to, citing lack of legal authority in the Mumbai police to request such a ban.21 In 2001, it had been dropped by its American web host due to hate speech concerns.22 But none of these events gave rise to much mainstream media attention to Internet censorship. That happened for the first time in 2006, in the aftermath of train bombings in Mumbai, when the Department of Telecommunications issued orders to ISPs to block 17 domains and web pages.23 The timing gave rise to many rumours about the blocks having been occasioned by the bombings. However, by going through (the non-public) list one saw that the list included mostly obscure sites: a site arguing for Dalit separatism,24 a personal website of a right-leaning Indian American,25 little-known right-leaning American blogs which had nothing to do with India,26 a web-based SMS gateway service,27 and some domains that didn’t even exist on the day they were blocked,28 amongst others. The most notable website that was included in the list was HinduUnity.org (which, as noted earlier, had already been ordered to be blocked in 2004).29 Despite the lack of popularity or notability of those 17 sites, this secretive order was noticed by ordinary Web users because of a gigantic mistake.
Amongst the 17 sites ordered to be blocked were specific blogs and pages hosted on Blogspot.com, and Typepad.com. Instead of those particular blogs being blocked, all blogs and pages hosted on Blogspot.com, Typepad.com (and Geocities, inexplicably) were blocked. This resulted in the block being noticed by a large number of people, and garnering a larger amount of media coverage than in the past. However, the only response of the government to the media furore was that of pinning the blame on the ISPs for over-blocking,30 rather than seeking to justify the blocking of those 17 URLs, which contained perfectly legitimate websites that didn’t seem to prima facie violate any Indian laws.
The next time that these issues sprang into prominence of some sort was when the website of Savitha Bhabhi, an erotic webcomic, was blocked in 2009, just before a large amendment of the IT Act came into force.31 The anonymous UK-based author of the cartoon series outed himself and contacted lawyers in India to defend his creation, but due to pressure from his embarrassed family, he dropped the matter.
In the 2008 amendment to the IT Act (which were brought into effect in October 2009), a new provision — section 69A — was added which granted the government powers to block websites if it “is satisfied that it is necessary or expedient so to do in the interest of sovereignty and integrity of India, defence of India, security of the State, friendly relations with foreign States or public order or for preventing incitement to the commission of any cognizable offence relating to above”.
Transparency around Website Blocking
In 2011, I submitted a right to information (RTI) request about what websites had been blocked since the new law came into force. The government of India’s reply to this RTI request was groundbreaking in a way, since it was the first time the government had provided an official list of URLs that it had blocked in India.32 Even the publication of this list did not result in much mainstream media coverage. All eleven blocked URLs had been ordered to be blocked by courts — constituting direct state regulation — however the courts and the attorneys had done an amazingly shoddy job: some of the URLs were for Google search results rather than the web pages themselves, the whole of the Indymedia portals for San Francisco and Arizona instead of just the pages the High Court found to be defamatory, and similarly all of Webs.com was blocked instead of a specific URL. The rationale for most of these was not clear even after circumventing the blocks and visiting those pages which continued to exist.
Later, in May 2012, the Internet collective Anonymous released a list of URLs blocked by Reliance Communications.33 Since none of the blocks that CERT-In has ordered has been published by them, whether in the Gazette of India or on their website, this was the first time that a list of all websites blocked in India — and not just under the IT Act, since 2008 — was available in the public domain. Going through the list the same evening they were released, I found, as alleged by Anonymous, that there were more than a dozen links — mostly alluding to the involvement of a senior Reliance official, who was then in prison, in a telecommunications spectrum corruption scandal — that were blocked if one used a Reliance Communications connection but not on other ISPs.34 However, by the next morning those links were working on Reliance networks too. This leak, even though it was reported on a prominent independent political blog, as also a blog run by a mainstream news magazine, did not get much traction in the wider mainstream media.
Apart from detailing private censorship, that leak also made it clear that BuyDomains, Fabulous Domains and Sedo.co.uk — domain name marketplaces — were being blocked on orders of the Indian government. What is less clear is whether the government had any legal authority to do so.35
By far the largest category of blocked websites is entertainment and files-sharing websites. One set of those (104 domains) were blocked by an interim order of the Calcutta High Court.36 The rest of them, however, were blocked by private requests by entertainment companies subsequent to generic “John Doe” orders from courts. There is a strong case to be made that this private extension of John Doe orders is unlawful and far beyond the scope of the orders themselves.37 Further, even if one were to argue that they were lawful, there are numerous clear examples of indefensible overreach — where sites that are clearly not engaging in copyright infringement of music or films have been blocked.38 Thus, it is plain to see that perfectly lawful and non-infringing websites are being censored in the guise of copyright infringement.
Importantly, these private blocks defeat currently-available means of transparency. Thanks to the Right to Information Act, 2005, the list of blocked websites under section 69A is available to the public upon request, even if the IT Act does not require proactive publication of the list, as it should. This provides the opportunity for a constant vigil against direct state-ordered censorship, even if through less-than-ideal means. However, a right to information request would not cover the sites that were blocked through private requests by entertainment companies. For access to those, we had to count on leaks to the press and civil society organizations by industry insiders and unauthorized access to ISP servers.
So far in this brief history, I’ve covered mostly direct state censorship, and one instance of state-allowed private censorship undertaken by some entertainment companies. In the next section, I will deal with the regulations made under the intermediary liability law in India, and focus on how those regulations greatly expand the scope of state-enabled private censorship, and undermine the possibility of challenging censorship.
Indirect Censorship: Intermediary Liability
In India, section 79 of the IT Act is the provision that provides Internet intermediaries39 protection from liability for their users’ actions. Before the 2008 amendment, it covered “network service providers”, but then was expanded and re-drafted quite extensively,40 with the jailing of Avnish Bajaj, the CEO of Bazee.com, — for one of its users offering an illegally-obtained pornographic CD for sale — being a major impetus for the provision’s amendment.
On February 7th 2011, the Department of Information Technology under the Ministry of Communications and Information Technology published draft regulations under section 70 on its website (initially titled “Information Technology (Due Diligence Observed by Intermediaries Guidelines) Rules, 2011” and “Information Technology (Guidelines for Cyber Cafe) Rules, 2011”) in exercise of the powers conferred by section 87(2)(zg), IT Act, read with section 79(2). Comments were invited from the public till February 25, 2011.
The Centre for Internet and Society submitted comments noting, inter alia, that the proposed rules were ultra vires the parent statute, and that some of the provisions of the draft Intermediaries Guidelines rules were plainly unconstitutional since they enabled the government to require Internet intermediaries to remove content on grounds that were far beyond those contained in Article 19(2) of the Constitution of India,41 while the draft Cyber Cafe rules greatly encroached upon the right to privacy.42 At that point, the draft of the Intermediaries Guidelines rules allowed only an “authority mandated under the law for the time being in force” to complain to intermediaries and require them to “remove access” to the offending material.
The government not only ignored the problems that were highlighted by civil society organizations, but introduced far greater ones. The final version of the Information Technology (Intermediary Guidelines) Rules (hereinafter ‘Intermediary Guidelines’), which have been in effect since April 2011, give not only an “authority mandated under the law”, but all “affected persons”43 great powers to censor the Internet!
Policy Sting Operation
Since there is no reporting mechanism contained in the Intermediary Guidelines, there is no means of gathering information about the usage of the rules: no one, not even the government, knows how often the rules are being used, and what content is being removed. Given that, we at the Centre for Internet and Society decided to test the censorship powers of the new rules through a ‘policy sting operation’, by sending frivolous and plainly defective complaints to a number of intermediaries.44 Six out of seven intermediaries removed content, including search results listings, on the basis of the most ridiculous complaints. The people whose content was removed were not told — none wrote to us asking why we objected to their content — nor was the general public informed that the content was removed. If we hadn't kept track, it would be as though that content never existed.45 Yet, not only was what the Internet companies did legal under the Intermediary Guideline Rules, but if they had not, they would have lost the protection from being punished for the content put up by their users.46
Fundamental Problems with the Intermediary Guidelines
There are many problems with the Intermediary Guidelines, but the fundamental issues are discussed below.47
First, it shifts the burden for exemption from liability on to intermediaries. Until the Intermediary Guidelines were brought into force, an intermediary who fell within the ambit of section 79(2) of the IT Act did not have to engage in a positive act to be able to claim exemption from liability for the words and deeds of their users. However, the Intermediary Guidelines require that intermediaries publish the terms of service contained in Rule 3(2) of the Guidelines, appointing a Grievance Redressal Officer as under Rule 3(11), follow reasonable security practices as required by Rule 3(8), report “cyber security incidents” to CERT-In, and perform other such acts to be able to claim the exemption from liability. This might mean that non-Indian intermediaries who fail to publish new terms of service in accordance with the Intermediary Guidelines would automatically fall afoul of the law and could be held liable for their users’ actions in Indian courts.
Second, it seems to pin liability on intermediaries for failing to perform acts unrelated to liability. Many of the requirements of the Intermediary Guidelines have nothing to do with the speech or conduct that may give rise to liability. The question then arises if failure to perform them could result in exemption from liability being denied. For instance, if a web hosting company failed to follow reasonable security practices or failed to report a particular ‘cyber security incident’, could that result in it being liable for all the defamatory content on its servers?
Third, it denies users any chance to defend their speech. The Intermediary Guidelines require that intermediaries that receive a complaint, “shall act within thirty six hours and where applicable, work with user or owner of such information to disable such information that is in contravention of sub-rule (2)”. It leaves it unclear what “where applicable” means in this case. It doesn’t seem to categorically state that the complainees need to be informed about complaints that the intermediary receives, nor does it categorically state that the complainee should be provided a chance to defend against the complaint. As noted above, during our policy sting operation, we did not receive a single complaint from any of the parties that might have been affected by our complaints. It seems as though none of the intermediaries ever informed those who would be affected about our complaints. The Supreme Court of India has held, “[i]n considering the reasonableness of laws imposing restrictions on fundamental right, both the substantive and procedural aspects of the impugned law should be examined from the point of view of reasonableness and the test of reasonableness, wherever prescribed, should be applied to each individual statute impugned”.48 Given this, there is a strong argument to be made that a system for removal of content which does not embed core principles of natural justice such as the audi alteram partem doctrine, would fail the reasonable test of Article 19(2).49
Fourth, the Intermediary Guidelines greatly expand the grounds under which content can be deemed unlawful. The prior means of blocking a website required a person to approach a statutory authority under section 69A of the IT Act citing one of six grounds, mostly to do with national security. But with the Intermediary Guidelines, there are thirty-two distinct grounds, a great many of which are not constitutionally justifiable. For instance, disparaging speech — as long as it isn't defamatory — is not unlawful in India; however the Intermediary Guidelines list that as a category of prohibited speech. Advertisements promoting gambling are not unlawful in India — indeed, various state governments regularly take out print advertisements and put up billboards about their lottery schemes — but now all Internet intermediaries are required to remove content that are about gambling, even if it doesn’t promote it.
Fifth, the Intermediary Guidelines make the intermediaries the judge of whether any particular content is in compliance with the law or not, rather than a judicial, or even quasi-judicial, body. This relegates speech regulation to private actors. While speech regulation by private actors isn’t in itself harmful (just as speech regulation by the state by itself isn’t), private actors are generally subject to far less accountability than the state.
Sixth, the law promotes a complete lack of transparency and accountability. No public notice is required to be provided that content has been removed, nor is there any reporting mechanism provided for the government to gather information about requests from intermediaries. So even the government does not know how many requests have been made after these Guidelines have come into effect, nor what content has been removed subsequent to those requests. This means that even the RTI Act, which has proven a powerful transparency tool to pry open the government, cannot be used. It also means that even the government doesn’t have the information necessary to judge the law’s successes and failings. In essence, this allows for invisible censorship.50
In the Centre for Internet and Society’s proposed alternative to the government’s Intermediary Guidelines, we suggest that the government run an open and central takedown request monitoring system similar to the Chilling Effects Clearinghouse,51 to which all those who receive notices — under the notice-and-notice provisions we’ve advanced — would be required to contribute.52
Seventh, the differentiations between categories of intermediaries is removed. A one-size-fits-all system is followed where an e-mail provider is equated with an online newspaper, which is equated with a video upload site, which is equated with a search engine: they all have to include Rule 3(2) of the Intermediary Guidelines and its thirty-two speech restrictions in their terms of service, and they will all lose exemption from liability if they fail to comply. This is like equating the post office and a book publisher as being equivalent in terms of liability for, say, defamatory speech. This is violative of Article 14 of the Constitution, which requires that unequals not be treated equally by the law.53
Eighth, the Intermediary Guidelines don’t require a proportionality test. A DNS provider is an intermediary who can be asked to ’disable access' to a website on the basis of a single page, even though the rest of the site has nothing objectionable. Given the way the DNS system works, it is not possible for a DNS provider to selectively prohibit access to a single page. However, there is nothing in the law that would prevent such an abuse, or require the hosting provider to be contacted instead of the DNS provider in such a case.
Ninth, the Intermediary Guidelines seem to be based on a presumption of illegality of content where any allegation of unlawful content is sufficient to constitute “actual knowledge” of the content’s unlawfulness.54 In a case on defamation, the Delhi High Court held, “Rule 3(4) of the said rule provides obligation of an intermediary to remove such defamatory content within 36 hours from receipt of actual knowledge.”55 In that case the complaint to the website (Hubpages.com) contained allegations of defamation, but those allegations were held to be sufficient to constitute “actual knowledge” on the part of the website of defamation. If the Court’s interpretation is correct, the Guidelines are based on the presumption that all complaints (and resultant mandatory taking down of the content) are correct, and that the incorrectness of the takedowns can be disputed in court if the complainee ever discovers that her content has been removed/blocked, etc.56 While this was at one point the interpretation of the Department of Electronics and Information Technology, it no longer is.57
Tenth, the Intermediary Guidelines are atemporal, assuming that any content removal / block has to last forever. On the other hand, many blocks, such as those relating to copyright infringement of a sporting event, are temporal in nature. Material removed or blocked due to a temporal event end up becoming permanent.
Eleventh, governmental diktat cannot just mirror industry “best practices” without any regard to constitutional validity. The Indian government has justified the Intermediary Guidelines as, “best practices followed internationally by well-known mega corporations operating on the Internet.”58 However, that ignores the fact that speech restrictions that may be imposed by “well-known mega corporations” aren’t restricted by the Indian Constitution in the same manner as it restricts the actions of the government. Further, it ignores the fact that different corporations choose to have widely differing terms of service. Even different services provided by a single corporation may have different policies on what is acceptable on that platform.59 The Intermediary Guidelines homogenizes those terms of service and makes it mandatory upon all intermediaries to include the government-prescribed terms, regardless of the services they provide and regardless of what the intermediaries consider as acceptable speech.
Twelfth, the Intermediary Guidelines do not bar governmental actors from using it to send takedown requests. Previously governmental actors would have to comply with the requirments of section 69A of the IT Act, or approach the courts — which seemingly are bound by no limits in terms of ordering the blocking of websites. Now, if they so choose, governmental actors can choose to go for the notice-and-takedown route which provides them far greater leeway — including the ability to block content it would be unconstitutional for the government to directly block under section 69A — while also providing statutory sanctions against intermediaries who fail to comply. This means that the government can get far more material removed without turning up in transparency reports of the kind that Google, Twitter, Yahoo, Facebook, and others issue.
And lastly, there are no penalties for filing frivolous complaints of the sort that we at the Centre for Internet and Society filed, nor for filing malicious complaints. This creates a perverse incentive structure that privileges complainants over complainees — who aren’t even required to be told about the complaints, and are not required to be afforded a chance to defend themselves.
In 1984, the then-Prime Minister of India, Indira Gandhi, was forced to sue Salman Rushdie for defamation in a London court in order to ensure one sentence was expurgated from his novel Midnight’s Children. Today Gandhi wouldn’t need to win a lawsuit against publishers. She would merely have to send a complaint to websites selling the book and it would be removed from sale unless the website wants to waive its exemption from liability.
What is astounding is not that such badly drafted subordinate legislation could be put forward by the government; it is that it could be passed despite cogent and trenchant criticism being provided as part of the public consultation process, as well as those criticisms being aired prominently in newspaper op-eds and editorials.
Conclusion
While the concerns with the Intermediary Guidelines were covered by the press, they mostly ignored the nuances involved in it — such as the fact that it did not require the complainee to be told, that it could lead to undetectable and invisible censorship, and other such procedural matters.60 By contrast, the publicity provided to instances of direct state censorship has been far greater. The four instances where the press provided the most coverage for Internet censorship over the past few years were instances of direct state censorship, state-directed private censorship, and state-enabled private censorship.
Example 1: In December 2011, the Minister for Communications and Information Technology told Indiatimes, Google, Yahoo, Facebook, and Microsoft, in closed-door meetings that they should come up with a code of self-regulation using which they should pre-screen certain kinds of objectionable content, noting that the government would come up with a ‘self-regulation’ code for them if they didn’t do so on their own.61 This was leaked to the New York Times, and that led to constant coverage that month.62
Example 2: In December 2011, a journalist named Vinay Rai filed a criminal complaint against Google, Yahoo, Microsoft, and a number of other companies for hosting content that “promoted enmity between communities”, as well as for hosting obscene content,63 while former journalist named Aijaz Ashraf Qasmi filed a civil lawsuit against them. These two court cases, following soon on the heels of the government’s attempts to muzzle those companies, allegedly for similar kinds of content, led to a great deal of mainstream media coverage.64
Example 3: In the aftermath of the violence that erupted in southern Assam in July and August 2012,65 and a rumour-fuelled panic that spread in Bangalore and elsewhere as to the physical safety of residents from the north-east of India,66 the government of India placed curbs on SMSes, and over a period of four days ordered 309 specific items (those being URLs, Twitter accounts, HTML tags,67 blog posts, blogs, and a handful of websites) to be blocked.68 This was the first time that such a large number of websites and web pages were ordered to be blocked by the government, and this led to plenty of mainstream news media coverage.
Example 4: In September 2012, a little-known cartoonist named Aseem Trivedi was charged under multiple statutes, including under section 66A of the IT Act, and arrested, followed two months later by the arrest of two girls from Mumbai for posting and ‘liking’ a comment on Facebook about the city-wide bandh (general strike) observed in Mumbai after the politician Bal Thackeray’s death.69 These outrageous arrests in September and November 2012 led to widespread condemnation of section 66A of the IT Act, which penalizes the sending of offensive messages through communication services.70
Instances of state-directed censorship, like those mentioned above, which can be observed much more easily, and conform to more traditional ideas of what constitutes censorship, get a fair amount of media coverage than state-enabled private censorship through the Intermediary Guidelines or through copyright infringement claims by entertainment companies, though in actuality the latter might be far more widespread than the former and affect much greater amounts of speech, and may affect far greater range of speech.
While private actors have always been involved in speech regulation, the centrality of the role that they now occupy is something new, but is also inevitable. Those who believe that all speech regulation must be done by the state, following due process, are trying to prop up the procedural standards of a bygone world. Instead of harking back to the procedures that exist for censorship of books, and demanding that they be followed in all cases of online content, we must find new ways of countering the complete lack of transparency and accountability of private actors. We must find a way to appropriately extend the civil and political rights we enjoy against the state — which were writ when the state was the predominant actor in the silencing of speech — to act as guarantees against certain kinds of private action as well. And central to that endeavour would be the shining of light and removing the cloak of invisibility under which most forms of private censorship, whether conducted at the behest of governments, subsequent to enabling laws, or otherwise, occur. Not doing so immediately will undoubtedly make it more difficult to counter this brave new world of invisible censorship that we are transitioning into.
Footnotes
The most detailed overview of this history is presented in a monograph produced by Raman Jit Singh Chima as part of his Sarai fellowship. Raman Jit Singh Chima, The Regulation of the Internet With Relation to Speech and Expression by the Indian State (2008), http://dx.doi.org/10.2139/ssrn.1237262. For a shorter history, see Shivam Vij, Internet Censorship in India Has a Long, Murky Past, Sunday Guardian (Dec. 11, 2011), http://www.sunday-guardian.com/technologic/internet-censorship-in-india-has-a-long-murky-past.↩︎
Arun Mehta v. Videsh Sanchar Nigam Ltd., Writ Petition (Civil) No. 4732 of 1998 (New Delhi), on file with the author.↩︎
Arun Mehta, Status of VSNL Censorship of IP-Telephony Sites, India GII, http://members.tripod.com/~india_gii/statusof.htm (last updated Aug. 9, 2001), archived at https://archive.today/8LJqQ.↩︎
Id.↩︎
Siddharth Varadarajan, Dawn Website Blocked as VSNL Plays Big Brother, Times of India (July 3, 1999), available at http://svaradarajan.com/1999/07/03/dawn-website-blocked-as-vsnl-plays-big-brother/.↩︎
An erstwhile employee of Satyam Infoway, India’s first private ISP, told me that the Dawn incident was merely one that was highly visible and hence reported in the press. He told me that Satyam Infoway would receive numerous requests — mostly unofficial and unrecorded — that would come from the Department of Telecommunication in those days, leading to websites being blocked without the press finding out.↩︎
Seema Kazi, the VSNL subscriber who brought this to light, noted that she, a Muslim, was told by a VSNL manager that this step was taken because “[m]uslims have links with Pakistan and because of reasons of security”. Seema Kazi, Letter to the Editor, Covert Censorship, Hindu (Nov. 11, 2000), http://www.thehindu.com/2000/11/11/stories/05111305.htm.↩︎
See VSNL Further Aff. ¶6, in Arun Mehta v. Videsh Sanchar Nigam Ltd., available at https://docs.google.com/View?docid=dc72g763_15d4hj95↩︎
The clause covering this in various licence agreements and “guidelines” covering different actors is different, and has also varied across time. I know of no comprehensive analysis of these licences as they pertain to freedom of speech and surveillance. The licence for ‘Internet service (Including Internet Telephony)’ as on April 19, 2002 included these clauses:
1.12.09. The [licensee] shall ensure that objectionable, obscene, unauthorised or any other content, messages or communications infringing copyright, [i]ntellectual property right and international & domestic cyber laws, in any form or inconsistent with the laws of India, are not carried in his network, the ISP should take all necessary measures to prevent it. In particular, [the licensee] is obliged to provide, without delay, all the tracing facilities of the nuisance[-causing] or malicious messages or communications transported through [its] equipment and network, to authorised officers of [the] Government of India/State Government, when such information is required for investigations of crimes or in the interest of national security. The licence shall be governed by the provisions of the Information Technology (IT) Act 2000, as modified from time to time. Any damages arising out of default on the part of licensee in this respect shall be sole responsibility of the licensee.
1.12.10. The use of the network for anti-national activities would be construed as an offence punishable under the Indian Penal Code or other applicable law. The networks cannot be used in such a manner as to endanger or make vulnerable [] networked infrastructure. Acts such as break-ins or attempted break-ins of Indian networks shall be regarded as an anti-national act and shall be dealt with in accordance with the Indian Penal Code. ISPs must ensure that their services are not used for such purposes.
Licence Agreement for Provision of Internet Service (Including Internet Telephony), Dep’t of Telecomm. (April 19, 2002), http://dot.gov.in/sites/default/files/internet_telephony_lce.doc. Nearly identical clauses are found in the 2007 licence as well. Licence Agreement for Provision of Internet Services, Dep’t of Telecomm. (Oct. 16, 2007), http://dot.gov.in/sites/default/files/internet-licence-dated%2016-10-2007_0.pdf [hereinafter 2007 ISP Licence].
Clause 27 of the 2007 Internet Service Guideline document, which formed the basis for the 2007 ISP licence, clarifies:
Flow of obscene, objectionable, unauthorised or any other content infringing copy-rights, intellectual property right and international & domestic [c]yber laws in any form over the ISP’s network is not permitted and the ISP is supposed to take such measures as to prevent it. Any damages/claim arising out of default on the part of the licensee in this respect shall be the sole responsibility of the licensee.
This pinning of liability in the licence terms is in direct opposition with the exemption from liability contained in section 79 of the IT Act. This is just one of the numerous instances of lack of coherence — and outright contradictions — in Indian information and telecommunications policy and law.↩︎
Clause 33.3, 2007 ISP Licence, supra note 9, states:
The LICENSEE shall take necessary measures to prevent objectionable, obscene, unauthorized or any other content, messages or communications infringing copyright, intellectual property etc., in any form, from being carried on his network, consistent with the established laws of the country. Once specific instances of such infringement are reported to the [licensee] by the enforcement agencies, the [licensee] shall ensure that the carriage of such material on [its] network is prevented immediately.↩︎
Information Technology Act, 2000, Section 67.↩︎
This contrasts with the way section 95 of India’s Code of Criminal Procedure provides for the seizure of books declared to be punishable under sections 124A, 153A, 153B, 292, 293, and 295A of the Indian Penal Code.↩︎
Computer Emergency Response Teams are groups that handle computer security breaches, and the Indian CERT [hereinafter CERT-In] describes itself as the “national nodal agency for responding to computer security incidents as and when they occur.” Indian Computer Emergency Response Team, http://www.cert-in.org.in/ (last visited Dec. 28, 2014).↩︎
The Gazette of India Extraordinary Part II – Section 3(i), Notification no. GSR. 181(E), Ministry of Communications and Information Technology (Department of Information Technology) – Government of India, Feb. 27, 2003, available at http://deity.gov.in/content/it-act-notification-no-181 (last visited Jan. 14, 2014). This was rescinded in May 2010 by another notification, after provisions on website blocking were introduced into the statute. The Gazette of India Extraordinary Part II – Section 3(i), Notification no. G.S.R. 410(E), Ministry of Communications and Information Technology (Department of Information Technology) – Government of India, May 17, 2010, available at http://www.egazette.nic.in/WriteReadData/2010/E_257_2011_010.pdf.↩︎
The Gazette of India Extraordinary Part II – Section 3(i), Notification no. G.S.R. 529(E), Ministry of Communications and Information Technology (Department of Information Technology) – Government of India, July 7, 2003.↩︎
See, e.g., Bennett Coleman v. Union of India, (1972) 2 SCC 788 (holding that restrictions on newsprint constituted a restriction on freedom of expression, and that the right to receive information is part of the right to freedom of speech and expression). Compare Sec’y, Min. of Info. & Broadcasting v. Cricket Ass’n, (1995) 2 SCC 161 (holding that scarcity of spectrum does means the government has to act as a ‘custodian’ of the airwaves and must act in the public interest). It is instructive to note the contrast between the Supreme Court’s decisions that scarcity of foreign exchange and newsprint cannot lead to greater government regulation of speech via newspapers, while also ruling that scarcity of spectrum may legitimately lead to increased government regulation of the airwaves.↩︎
Article 19(2) of the Constitution provides for the exceptions to the right to freedom of speech and expression enshrined in Article 19(1)(a). Post two amendments in 1951 and 1963, Article 19(2) states:
Article 19(2) — Nothing in sub clause (a) of clause ( 1 ) shall affect the operation of any existing law, or prevent the State from making any law, in so far as such law imposes reasonable restrictions on the exercise of the right conferred by the said sub clause in the interests of the sovereignty and integrity of India, the security of the State, friendly relations with foreign States, public order, decency or morality or in relation to contempt of court, defamation or incitement to an offence.
The Supreme Court has sometimes been very inconsistent in its application of Article 19(2): sometimes being very strict in its interpretation and sometimes loose. This is perhaps inevitable given the fact that the Indian Supreme Court hardly ever sits en banc, and this has caused many problems. See T.R. Andhyarujina, Restoring the Supreme Court’s Exclusivity, Hindu (Aug. 31, 2013), http://www.thehindu.com/todays-paper/tp-opinion/restoring-the-supreme-courts-exclusivity/article5077644.ece (“With the increasing load of appeals from High Court decisions the number of judges have had to be increased periodically from eight judges in 1950 when the Constitution came into force to 31 in 2008. Presently, the Supreme Court is composed of one bench of the Chief Justice’s Court of three judges and 13 or 14 benches of two judges in 13 or 14 courtrooms sitting regularly day after day. In no Supreme Court of other jurisdictions are there benches of 13 to 14 courts of two judges each as the Indian Supreme Court now has.”); see also T.R. Andhyarujina, Studying the U.S. Supreme Court's Working, (1994) 4 S.C.C. J. 1, available at http://www.ebc-india.com/lawyer/articles/94v4a1.htm. Traditionally, the Supreme Court has been seen as the bulwark of protection against governmental encroachment into fundamental rights, while the lower courts, including sometimes the High Courts (which are also constitutional courts, and have the power of judicial review of legislation), have not always enjoyed the same reputation.
Notification GSR181(E), supra note 15.↩︎
See Ministry of Communications and Information Technology, Blocking of Website, Press Info. Bureau (Sept. 22, 2003), http://pib.nic.in/archieve/lreleng/lyr2003/rsep2003/22092003/r2209200314.html; see also Yahoo! Groups Blocked in India, Sun. Morning Herald (Sep. 26, 2003), http://www.smh.com.au/articles/2003/09/26/1064083178553.html.↩︎
Suresh Ramasubramanian, Re: Dishnet Blocking Yahoogroups – More, India-GII Mailing List (Sept. 20, 2003, 04:39), http://permalink.gmane.org/gmane.org.telecom.india-gii/2863.↩︎
Chima, supra note 1, at 54.↩︎
Dean A. Murphy, Two Unlikely Allies Come Together in Fight Against Muslims, N.Y. Times (June 2, 2001), http://www.nytimes.com/2001/06/02/nyregion/two-unlikely-allies-come-together-in-fight-against-muslims.html.↩︎
Directions to Block Internet Websites, Dep’t of Telecomm. (July 13, 2006), available at https://www.flickr.com/photos/22315040@N05/15091167867/.↩︎
http://princesskimberly.blogspot.com, http://mypetjawa.mu.nu, http://pajamaeditors.blogspot.com, http://exposingtheleft.blogspot.com, http://www.thepiratescove.us, http://www.bamapachyderm.com, http://merrimusings.typepad.com, and http://mackers-world.com.↩︎
Vij, supra note 1.↩︎
Ministry of Communications & Information Technology, DoT Orders Internet Service Providers to Block Only the Specified Webpages/Websites, Press Info. Bureau (July 20, 2006), http://pib.nic.in/newsite/erelease.aspx?relid=18954.↩︎
Venkatesan Vembu, Save Our Savitha Bhabhi, DNA (July 3, 2009), http://www.dnaindia.com/analysis/column-save-our-savita-bhabhi-1270664.↩︎
Pranesh Prakash, DIT's Response to RTI on Website Blocking, Centre for Internet and Society (Apr. 07, 2011), http://cis-india.org/internet-governance/blog/rti-response-dit-blocking.↩︎
Isac, List of URLs Blocked by Reliance Infocomm, Anonymous (May 25, 2012), http://pastehtml.com/view/bywiha3f9.txt, archived at https://archive.today/Is7Sn.↩︎
I tried three ISPs: BSNL, Tata Indicom, and ACT Broadband.↩︎
Smitha Krishna Prasad, DoT Blocks Domain Sites — But Reasons and Authority Unclear, Centre for Internet and Society (Nov 21, 2012), http://cis-india.org/internet-governance/blog/dot-blocks-domain-sites.↩︎
Nikhil Pahwa, List of 104 Music Sites That The Indian Music Industry Wants Blocked, Medianama (Mar. 15, 2012), http://www.medianama.com/2012/03/223-list-of-104-music-sites-that-the-indian-music-industry-wants-blocked/↩︎
See Ananth Padmanabhan, Can Judges Order ISPs to Block Websites for Copyright Infringement? (Part 1), Centre for Internet and Society (Jan. 30, 2014), http://cis-india.org/a2k/blog/john-doe-orders-isp-blocking-websites-copyright-1.↩︎
Sites like Pastebin.com (which only hosts text content, not audio), and sites which have content that goes far beyond the limited copyright infringing material they may have, like Vimeo.com (a general video-hosting website) and Chakpak.com (a general entertianment website), have also been blocked.↩︎
The term “intermediary” is very broadly defined in s.2(w) of the IT Act: “‘intermediary’ with respect to any particular electronic records, means any person who on behalf of another person receives, stores or transmits that record or provides any service with respect to that record and includes telecommunications service providers, network service providers, internet service providers, web-hosting service providers, search engines, online payment sites, online-auction sites, online-market places and cyber cafes”.↩︎
Pranesh Prakash, Short Note on IT Amendment Act, 2008, Centre for Internet and Society (Feb. 2009), http://cis-india.org/internet-governance/publications/it-act/short-note-on-amendment-act-2008.↩︎
Pranesh Prakash, CIS Para-wise Comments on Intermediary Due Diligence Rules, 2011, Centre for Internet and Society (Feb 25, 2011, 04:45), http://cis-india.org/internet-governance/blog/intermediary-due-diligence.↩︎
Prashant Iyengar, CIS Para-wise Comments on Cyber Café Rules, 2011, Centre for Internet and Society (Feb 25, 2011, 03:30), http://cis-india.org/internet-governance/blog/cyber-cafe-rules.↩︎
The rules do not define the term.↩︎
Rishabh Dara, Intermediary Liability in India: Chilling Effects on Free Expression on the Internet, Centre for Internet and Society (Nov. 16, 2013), http://cis-india.org/internet-governance/chilling-effects-on-free-expression-on-internet/intermediary-liability-in-india.pdf (explaining the methodology of the experiment and its results).↩︎
This brings to mind the way Klement Gottwald, the Czech communist leader, had Vladimír Clementis, a fellow senior member of the Communist Party of Czechoslovakia, erased from a photograph of the two of them standing next to each other, after Clementis was indicted in the Slánský show trial. In his novel The Book of Laughter and Forgetting, Milan Kundera describes this episode, and then has a character state the book’s most famous line: “The struggle of man against power is the struggle of memory against forgetting.” Milan Kundera, The Book of Laughter and Forgetting 4 (Aaron Asher trans., HarperPerennial 1996) (1978).↩︎
Dara, supra note 44.↩︎
This section includes material I’ve previously published on the Centre for Internet and Society’s Internet Governance blog.↩︎
State of Madras v. V.G. Row, 1952 S.C.R. 597, 598.↩︎
In another chapter of this book, Andrews Rens considers the South African law (which is similar) and the principles of natural justice in some detail.↩︎
See Pranesh Prkash, E-Books Are Easier to Ban Than Books, Outlook Mag. (Jan. 27, 2012), http://www.outlookindia.com/article/Ebooks-Are-Easier-To-Ban-Than-Books-/279712.↩︎
See About Us, Chilling Effects Clearinghouse, https://www.chillingeffects.org/about (last visited May 15, 2014).↩︎
See Pranesh Prakash & Rishabh Dara, Counter-proposal by the Centre for Internet and Society: Draft Information Technology (Intermediary Due Diligence and Information Removal) Rules, 2012, Centre for Internet and Society, http://cis-india.org/internet-governance/counter-proposal-by-cis-draft-it-intermediary-due-diligence-and-information-removal-rules-2012.pdf↩︎
See Venkateshwara Theatre v. State of Andhra Pradesh and Ors., (1993) 3 S.C.R. 616. (“Just a difference in treatment of persons similarly situate leads of discrimination, so also discrimination can arise if persons who are unequals, i.e. differently placed, are treated similarly . . . . A law providing for equal treatment of unequal objects, transactions, or persons would be condemned as discriminatory if there is absence of rational relation to the object intended to be achieved by the law.”)↩︎
“Actual knowledge” is a requirement of Section 79(3)(b) of the IT Act, which states: “upon receiving actual knowledge, or on being notified by the appropriate Government or its agency that any information, data or communication link residing in or connected to a computer resource controlled by the intermediary is being used to commit the unlawful act, the intermediary fails to expeditiously remove or disable access to that material on that resource without vitiating the evidence in any manner.”↩︎
Nirmaljit Singh Narula v. Indijobs at Hubpages.com, CS (OS) No.871/2012 (Delhi H.C., Mar. 30, 2012), 187234253 Indian Kanoon ¶ 15, http://indiankanoon.org/doc/187234253/.↩︎
The Delhi High Court’s reading of the law seems to be contradictory to the ‘clarification’ that the Department of Electronics and Information Technology offered in March 2013 through a statement on its website: “It is clarified that the intended meaning of the said words is that the intermediary shall respond or acknowledge to the complainant within thirty six hours of receiving the complaint/grievances about any such information as mentioned in sub-rule (2) of Rule 3 and initiate appropriate action as per law. Further, the Grievance Officer of the intermediary shall redress such complaints promptly but in any case within one month from the date of receipt of complaint.” Clarification on the Information Technology (Intermediary Guidelines) Rules, 2011 under Section 79 of the Information Technology Act, 2000, Dep’t of Electronics & Info. Tech., http://deity.gov.in/sites/upload_files/dit/files/Clarification%2079rules%281%29.pdf. Interestingly, in May 2011 the Department of Electronics and Information Technology had stated, “In case any issue arises concerning the interpretation of the terms used by the Intermediary, which is not agreed to by the user or affected person, the same can only be adjudicated by a Court of Law. The Government or any of its agencies have no power to intervene or even interpret.” But it proceeded to do exactly that in its “Clarification” of March 2013, essentially disagreeing both the with court’s interpretation as well as its own previous statement.↩︎
Id.↩︎
Ministry of Communications & Information Technology, Exemption from Liability for Hosting Third Party Information: Diligence to be Observed under Intermediary Guidelines Rules, Press Info. Bureau (May 11, 2011, 16:36), http://pib.nic.in/newsite/erelease.aspx?relid=72066.↩︎
Nicholas Bramble explores this idea in depth in his forthcoming paper tentatively titled, “Speech and Safety Laboratories”, which he presented at the Freedom of Expression Scholars Conference 2014 held at Yale University in May 3, 2014.↩︎
The mainstream media also completely ignored the Cyber Cafe Rules, though those rules not only barred anonymous usage of cybercafes, but also required cybercafe operators to record the web-browsing histories of all their customers.↩︎
See Heather Timmons, India Asks Google, Facebook to Screen User Content, N.Y. Times: India Ink (Dec. 5, 2011, 06:33), http://india.blogs.nytimes.com/2011/12/05/india-asks-google-facebook-others-to-screen-user-content/.↩︎
See Pranesh Prakash, Press Coverage of Online Censorship Row, Centre for Internet and Society (Dec. 8, 2011), http://cis-india.org/internet-governance/blog/press-coverage-online-censorship↩︎
Judge Kumar notes in his summons order:
It seems that instead of regulating the undesirable and offensive content they have promoted the same for increasing the profits and promoting their business. They have closed their eyes and promoted obscene[,] derogatory[,] defamatory[,] and inflammatory material continuously on their network. It appears from a bare perusal of the documents that prima facie the accused in connivance with each other and other unknown persons are selling, publicly exhibiting[,] and have put into circulation obscene, lascivious content which also appeals to the prurient interests and tends to deprave and corrupt the persons who are likely to read, see or hear the same.
Vinay Rai v. Facebook India and Ors., Summons Order, Dec. 23, 2011, available at http://cis-india.org/internet-governance/resources/vinay-rai-v-facebook-summons-order-2011-12-23.
See, e.g., Danish Raza, Sibal Not a Lone Crusader for Internet Censorship: Meet the Others, Firstpost (Dec. 26, 2011), http://www.firstpost.com/india/sibal-not-a-lone-crusader-for-internet-censorship-meet-the-others-166052.html; Amol Sharma, Is India Ignoring its own Internet Protections?, Wall St. J.: India Real Time (Jan. 16, 2012), http://blogs.wsj.com/indiarealtime/2012/01/16/is-india-ignoring-its-own-internet-protections/; Aparna Viswanathan, Op-ed, The Curious Case of Vinay Rai , Hindu (Feb. 15, 2012), http://www.thehindu.com/todays-paper/tp-opinion/the-curious-case-of-vinay-rai/article2894391.ece; and Danish Raza, Vinay Rai vs Facebook: Govt Uses Courts to Censor the Internet, Firstpost (Jan. 13, 2012), http://www.firstpost.com/india/vinay-rai-vs-facebook-govt-uses-courts-to-censor-the-internet-181603.html.↩︎
There are various linkages between the violence in Assam and in Myanmar, and the resultant censorship. For a comparison of the similarities and differences in two situations, see Subir Bhaumik, Assam Violence Reverberates Across India, Al Jazeera (Aug. 16, 2012), http://www.aljazeera.com/indepth/features/2012/08/201281572950685537.html. Compare B. Raman, Op-ed, India: Fissures in Assam: Sons of Soil vs Bangladesh Intruders, Eurasia Rev. (July 29, 2012), http://www.eurasiareview.com/29072012-india-fissures-in-assam-sons-of-soil-vs-bangladesh-intruders-oped/ (equating the Muslims in Rakhine and in Assam as illegal immigrants from Bangladesh). There were also instances of photos from an earthquake in Tibet, and other such images being falsely circulated in Pakistan, India, and elsewhere as evidence of the mass murder of Rohingyas in Myanmar, while, as an example, gruesome photos of two rape-murder victims in El Salvador were circulated as being photos of Hindus decapitated and dismembered by Muslims in Assam. See Yousuf Saeed, How to Start a Riot out of Facebook, Kafila (Aug. 13, 2012), http://kafila.org/2012/08/13/how-to-start-a-riot-out-of-facebook-yousuf-saeed/; see also Faraz Ahmed, Social Media Is Lying to Your About Burma’s Muslim ‘Cleansing’, Express Tribune: Media Watchdog (July 19, 2012), http://blogs.tribune.com.pk/story/12867/social-media-is-lying-to-you-about-burmas-muslim-cleansing/; and Pranesh Prakash, Pranesh Prakash on Twitter: Gruesome & graphic example of hate speech & incitement to violence using lies abt Hindu women being raped & decapitated http://goo.gl/TdGnA, Twitter (Aug 22, 2012, 16:27), https://twitter.com/pranesh_prakash/status/238417175965212672.↩︎
See, e.g., Harichandan Arakali, Thousands Flee Bangalore over Assam Violence, Reuters (Aug. 16, 2012), http://in.reuters.com/article/2012/08/16/bangalore-assam-north-east-bodo-idINDEE87F0BU20120816.↩︎
Even things that couldn’t be blocked by ISPs, like HTML tags and Twitter user handles were requested to be blocked by the government’s orders. See Prakash, infra note 68.↩︎
See Pranesh Prakash, Analysing the Latest List of Blocked Sites (Communalism & Rioting Edition), Centre for Internet and Society (Aug. 22, 2012), http://cis-india.org/internet-governance/blog/analysing-blocked-sites-riots-communalism.↩︎
See Pranesh Prakash, Arbitrary Arrests for Comments on Bal Thackeray’s Death, Centre for Internet and Society (Nov. 19, 2012), http://cis-india.org/internet-governance/blog/bal-thackeray-comment-arbitrary-arrest-295A-66A.↩︎
See Pranesh Prakash, Breaking Down Section 66A of the IT Act, Centre for Internet and Society (Nov. 25, 2012), http://cis-india.org/internet-governance/blog/breaking-down-section-66-a-of-the-it-act.↩︎
Reuse
Citation
@incollection{prakashIndiaVisible2014,
author = {Prakash, Pranesh},
editor = {Prakash, Pranesh and Rizk, Nagla and Affonso Souza, Carlos},
title = {India: Visible and Invisible Censorship},
booktitle = {Global censorship: shifting modes, persisting paradigms},
pages = {107-130},
date = {2014},
url = {https://cis-india.org/a2k/a2kga-global-censorship#page=121},
langid = {en}
}