India Plans Chinese Style Internet Censorship
A unique form of hate speech had begun developing in India when a video circulated on WhatsApp along with inflammatory messages led to the death of around 60 people in the 2013 Muzzafarnagar riots. Throughout this article, this form of hate speech will be called WhatsApp hate. WhatsApp hate includes, but is not limited to, messages in the form of texts or visual representations which would promote disharmony, or ill-will amongst communities based on caste, religion language, place of birth or residence or any other ground. It is a form of speech which seeks to polarize opinion and may be articulated along a provocative spectrum. It may take the form of spreading unverified or false information or offensive messaging that transgresses the conventional lines of political discourse.
Much like the way anti-Semitic and white supremacist sentiment on Facebook is unique to the United States, WhatsApp hate – generally a form which promotes Hindu-Muslim divisions, is unique to India. For this reason, in this article, the focus shall be on the intermediary liability regime around WhatsApp. This article is divided into four sections. In the first section, the specific features which make WhatsApp the preferred driver of hate speech in India will be determined in order to understand what needs to be tackled. The second section shall briefly summarize the existing regime of intermediary liability. Then it would proceed to look at the proposed new Intermediary Guidelines Intermediaries Guidelines (Amendment) Rules 2018 and the impact they would have on WhatsApp particularly. The third section shall analyse the Rules and argue how they are problematic in terms of undermining fundamental rights and causing a chilling effect. The fourth section shall attempt to provide some possible alternatives to the proposed rules which would lead to the creation of a more principally acceptable regime of intermediary liability.
The Preference for WhatsApp
WhatsApp, in India, has turned into the primary platform of sharing information including what would constitute hate speech for a few reasons.
First, for people looking to target individuals more personally than through a post on Facebook, WhatsApp allows sending messages to groups of up to 256 people at once. In fact, when it had launched, the same message could be forwarded 20 times. WhatsApp has reduced the forwarding limit to 5 times in wake of the use of the platform by political blocs to spread messages that would divide communities. But even with this reduction, it is possible to spread the message to about 1300 people at once.
Second, WhatsApp is preferred because of the ‘end-to-end encryption’ which it uses which ensures that not even the systems of the platform are able to read what is sent over WhatsApp. This feature, combined with the handicapped Section 79 after Shreya Singhal v. Union of India, meant that there was virtually no check on what could be sent over WhatsApp.
Third, a social factor which perpetuates the sharing of information in India, even by those with no vested interests, has been found to be that sharing news etc. on WhatsApp groups promotes a sense of homophily amongst users and leads to validation of their beliefs and identities. All of these reasons combined make WhatsApp the primary driver of hate speech that is unique and hence, worthy of being termed WhatsApp hate.
The Intermediary Liability Regime in India
In India as well, a similar form of intermediary immunity - a fancy term for not placing the legal burden on platforms like WhatsApp and Facebook for content posted by their users - is envisaged in Section 79 of the Information Technology Act (‘IT Act’). As of now, the interpretation in Shreya Singhal v. Union of India is the law of the land with respect to interpreting this immunity. In this case, the Supreme Court read down the provision in Section 79 so that the intermediaries are shielded from taking down any content until there was a court order directing the intermediary to do so and they would only be liable in case they fail to follow such order.
This is problematic because it is unreasonable to expect citizens to file lawsuits in order to make intermediaries take content down. And the consequence of this is that, effectively, the only means to take down unlawful content is using tools within these platforms and then leave it to the discretion of the platforms. Further, this interpretation completely ignored the Parliament’s inclusion of some degree of responsibility on intermediaries under the act and took away the ‘due diligence’ and ‘actual knowledge’ factors from Section 79.
In order to correct the present situation, the Ministry of Electronics and Information Technology (‘MeitY’) brought proposed the Intermediaries Guidelines (Amendment) Rules 2018 (‘proposed rules’). The proposed rules have four changes which are discussed below along with their harmful consequences:
Draft Rule 3(4): This rule inserts Rule 3(4), inserts a monthly requirement (at the least) to inform users about the legal requirements such as the terms and conditions and privacy.
Draft Rule 3(5): The rule states a traceability requirement such that the platform, like WhatsApp, must make available any information or assistance demanded by Government authorities. The platform should also enable traceability of sources of information when asked for.
Draft Rule 3(8): increases the data retention period from 90 to 180 days and allows for further discretionary retention when required by “government agencies”.
Draft Rule 3(9): The rule states that the intermediary shall proactively deploy automated tools to take down or disable access to lawful content.
A primary consequence of these proposed rules is that it they force WhatsApp to break its end-to-end encryption feature. This is because, in order to provide Government agencies with desired information or trace of the user, it is necessary move beyond a system which disallows WhatsApp systems from seeing what is shared over the platform. If WhatsApp does not provide agencies with the information they require, it would go against the rules and lose its immunity under Section 79. Hence, it is left with no choice but to comply and decrypt data.
The result of this is that a host of Government agencies, which is an undefined term, can now gather identities of individuals and the information they share over WhatsApp and force WhatsApp to store this information. Needless to say, this constitutes a massive breach of the citizen’s right to privacy without following any form of due process. The Government agencies have discretionary power with respect to viewing information sent and received over WhatsApp with no constraints whatsoever. There is no requirement of transparency or informing the traced individual about an ongoing investigation. Hence, effectively, the provision creates a situation similar to China and enables Government agencies to surveil citizens through their interactions on WhatsApp.
Another consequence of the proposed rules is the creation of a chilling effect on free speech. This arises from the powers of Government agencies, the responsibility on platforms to proactively monitor and remove content and the ‘nanny requirement’ of sending notices to users each month. The first manifestation of the chilling effect will be by WhatsApp itself. Similar to the manner in which a comparable law in Germany has led to Facebook ‘deleting in doubt’ posts from its platform. The reason for doing this is to avoid liability which it may face. In India, any automated mechanisms WhatsApp will put in place will suffer from a similar problem of over deletion due to abundant caution along with lacunae due to culturally unique reactions to information which such mechanisms would not envisage. The second manifestation of the chilling effect will be amongst users who will now live under the fear of being secretly investigated by Government agencies and moderate speech. The monthly reminders will make users feel like they are no longer in a public park but in a guarded school yard and hence, must tread carefully. This would prevent full exercise of the right under Article 19(1)(a).
The third consequence of the formulation of these rules is that they, to a limited extent, reinstate Section 66A which was struck down by the Supreme Court. This is because, after the breaking of encryption, and allowing Government agencies to demand removal of content, these Government agencies have turned into arbiters on what constitutes unlawful speech. With access to information which flows on WhatsApp, it is in the hands of these agencies to classify speech as lawful or unlawful and take actions against the identities they trace using this information.
Hence, even though the Rules attempt to solve the problem of end-to-end encryption as stated earlier, they involve provisions which are extreme in nature and do not even address the problem of the specific nature of group chats.
The proposed rules suffer from several major flaws and hence, it is important to lay down possible alternatives to, what is certainly, a clampdown on WhatsApp. It has briefly been mentioned how the German law, which imposes intermediary responsibility and fines on the failure to comply, is one of the strictest in their world, and has still failed to achieve its goal of regulating Facebook. The German experience shows that there is a need for a more cooperative conceptualization of intermediary liability. Some such measures are suggested below:
Customize in Automated Tools to include Constitutional Policy by Embedding Government Regulators - France had ‘embedded’ hate speech regulators within the Facebook system for a period of 6 months to examine how it combats hate speech. This example can be taken forward in relation to the proposed guidelines. Instead of asking the platform to use automated mechanisms of regulation which will inevitably harm the right to free speech, it is beneficial to allow an authority with the knowledge of the existing constitutional and public policy considerations, cultural uniqueness, applicable laws etc. to verify the mechanisms used by the platform as being suitable for the country. Such an embedded authority can also assist a platform like WhatsApp in editing its existing mechanisms such that it leads to the least possible damage to free speech and clamps down on hate speech more effectively. This will allow the regulation of hate speech without violating the citizens’ right to privacy, and also let the platform itself keep a semblance of control over its functioning while adapting itself to the unique needs of India as mentioned in the first section.
Creation of an Independent Authority to monitor the effectiveness of these mechanisms - Such an authority would act as a bridge between the Government and the platform. This authority would require access to a wide range of information held by the platforms, the ability to use different identities and access to algorithms to verify if the platforms are publishing content correctly. This can be done through the use of Application Program Interfaces (APIs) which can provide proportional access to network systems in real time.  After this check has been conducted, the authority will report back to the Government with its findings. Based on these findings, the Government may choose its course of action with respect to regulation of hate speech on that platform. This could prevent the breaking of the end-to-end encryption permanently and allow citizens to exercise their right to free speech and thus, reduce the chilling effect. It would also allow the platform to maintain its integrity while still allowing regulation in a cooperative fashion. It would also make sure that the decryption provision is retained in S.69(4) of the act and does not automatically follow from the rules.
This article attempted to conceptualize the reasons behind the existence of the unique brand of hate speech called WhatsApp-hate in India. It has been shown that a hard, invasive approach, as adopted by the proposed guidelines, to counter the WhatsApp hate in India causes violations of Fundamental rights of citizens and a chilling effect on the exercise of Freedom of speech. The guidelines which are proposed force WhatsApp to give up its end-to-end encryption and yet may not be as effective in countering hate speech. Thus, it is concluded that a cooperative, less invasive approach based on regulation using customization according to context of automated tools of regulation, and independent authorities to verify and report functioning and progress of regulation of hate speech on WhatsApp is a better alternative to the proposed guidelines.
Dipankar Ghosh, ‘Muzaffarnagar rioters used WhatsApp to fan flames, find police’ (The Indian Express, 12 September 2013) <http://archive.indianexpress.com/news/muzaffarnagar-rioters-used-WhatsApp-to-fan-flames-find-police/1168072/1> accessed 21 May 2019
 Matti Pohjonen and Sahana Udupa, ‘Extreme Speech Online: An Anthropological Critique of Hate Speech Debates’ (2017) 11 International Journal of Communication <https://ijoc.org/index.php/ijoc/article/viewFile/5843/1965> accessed 21 May 2019
 Soma Basu, ‘Manufacturing Islamophobia on WhatsApp in India’ (The Diplomat, 10 May 2019) <https://thediplomat.com/2019/05/manufacturing-islamophobia-on-WhatsApp-in-india/> accessed 21 May 2019
 Jacob Kastrenakes, ‘WhatsApp limits message forwarding in fight against misinformation’ (The Verge, 21 January 2019) <https://www.theverge.com/2019/1/21/18191455/WhatsApp-forwarding-limit-five-messages-misinformation-battle>
 In Shreya Singhal v. Union of India WRIT PETITION (CRIMINAL) NO.167 OF 2012
Sec.79 of India’s Information technology act was read down to hold intermediaries liable only when they failed to comply with a court order demanding removal of information. Court orders on such matters are extremely hard to obtain in India.
 Santanu Chakraborty et al, ‘Duty, Identity and Credibility’ (BBC News) <http://downloads.bbc.co.uk/mediacentre/duty-identity-credibility.pdf>
 Shreya Singhal v. Union of India, WRIT PETITION (CRIMINAL) NO.167 OF 2012.
 Prashant Reddy, ‘Liability, not Encryption, is what India’s new Intermediary Regulations are trying to fix’ (The Wire, 28 December 2018) <https://thewire.in/government/liability-not-encryption-is-what-indias-new-intermediary-regulations-are-trying-to-fix> accessed 21 May 2019
 Abhijit Ahaskar, ‘What the government’s draft IT guidelines say’ (Livemint, 12 February 2019) <https://www.livemint.com/technology/tech-news/what-the-government-s-draft-it-intermediary-guidelines-say-1549959448471.html> accessed 21 May 2019.
 Divij Joshi, ‘Draft Intermediary Guideline Rules will Undermine Fair Dealing and Access to Knowledge Online’ (SpicyIP, 31 January 2019) <https://spicyip.com/2019/01/draft-intermediary-guidelines-rules-will-undermine-fair-dealing-and-access-to-knowledge-online.html> accessed 21 May 2019
 Linda Kinstler, ‘Germany’s Attempt to Fix Facebook is Backfiring’ (The Atlantic, 18 May 2018) <https://www.theatlantic.com/international/archive/2018/05/germany-facebook-afd/560435/> accessed 22 May 2019
 Mathieu Rosemain, ‘France to ‘embed’ regulators at Facebook to combat hate speech’ (Reuters, 12 November 2018) <https://www.reuters.com/article/us-france-facebook-macron/france-to-embed-regulators-at-facebook-to-combat-hate-speech-idUSKCN1NH1UK> accessed 22 May 2019
 Paul Schiff Berman, ‘Cyberspace and the State Action Debate: The Cultural Value of Applying Constitutional Norms to ‘Private’ Regulation’  71 U. Colo. L. Rev.
 Andre Oboler, ‘How technology can be used to combat online hate speech’ (World Economic Forum, 13 March 2018) <https://www.weforum.org/agenda/2018/03/technology-and-regulation-must-work-in-concert-to-combat-hate-speech-online> accessed 21 May 2019.
...We have a small favor to ask. Polemics and Pedantics is a non-profit educational venture whose writers work only because of their penchant for the art. If you like our work, please support us by sharing it on social media and helping us reach more people. Remember to subscribe and never miss an update by providing your email on the Contact Page. We don't sell ads, and won't spam you or share your details with anyone. Comments and suggestions are welcome at email@example.com.
About the Author
Pallavi Khatri is a student of Law at the National Law School of India University, Bangalore. She is also a typical geek who can weave Game of Thrones and Star Wars into any conversation. Pallavi has a keen interest in ADR and has participated in multiple competitions on the subject. An avid reader of crime fiction, Pallavi loves criminal law and knows the IPC by heart. She also has an interest in legal policy; expect a lot of articles on problems relating to courts, national security, and gender equality