The Anatomy of Internet Shutdowns – II (Gujarat & Constitutional Questions)

Written by Nakul Nayak

In the last post I discussed the panoply of laws surrounding internet shutdowns in India and concluded that though there might be indirect regulatory connections to justify such shutdowns, they appear to be flimsy at best. In this post, I shall discuss the current internet shutdown in Gujarat in particular and whether these executive actions would pass constitutional muster.

As news reports would tell you, the Patidar reservation agitation began sometime in July, 2015. The community’s sole demand is to receive reservation benefits under India’s complex affirmative action formula. As momentum for the agitation picked up, a major demonstration dubbed Kranti Rally was organised in Ahmedabad, Gujarat’s largest city, with over half a million reportedly attending. Hardik Patel, the face of the movement, was arrested for not obtaining permission to stay on the ground after the rally and was later released. This coupled with instances of police violence heightened tensions within the state, manifesting itself in violence and vandalism of public property. That very night, Hardik Patel sent a WhatsApp message to his followers, urging them to maintain calm and simultaneously calling for a bundh the next day. Soon after, Whatsapp along with mobile internet in many parts of Gujarat shut down. They have been ever-since in major cities.

The current mobile internet disruptions are a blot on free speech in Gujarat. In the latter section of this post, I would go on to explain how this disruption is constitutionally overbroad and may constitute illegitimate prior restraint. However, taking off from where I left in the last post, my first quibble would be against the non-conformity with procedural propriety (assuming there is one) in directing the present blockage. Sec. 69A of the IT Act, which appears to be textually closest to affording kill switch powers, may only be initiated and implemented by the Central Government. On the other hand, sec. 5(2) of the Telegraph Act, which grants powers to detain or stop the transmission of messages, may be exercised by both the central and state governments. Contractually speaking, it appears from a liberal view of the Licence Agreement between the telecos and the Central Government, one may similarly conclude that both the central and state governments can direct actions of service disruptions.

However, when the Indian Express asked Dhananjay Dwivedi, Secretary of Science and Technology Department, when data services would be restored, he is reported to have said that “the decision was not taken by the state government but by local administration and police”. If this is true, then the directions to the telecos would smack of procedural lapses. Local authorities (as opposed to the state government) have nowhere been recognized under either of the statutes (IT Act or the Telegraph Act) or the Licence Agreement. They have assumed powers of ordering network disruptions without actually possessing them. In fairness, there are other reports that the District Collectors of Ahmdeabad, Vadodara etc. alongwith the police have taken this step. Technically, they would fall under the ambit of the ‘state government’. However, their websites have remained unavailable for access of notifications or circulars in this respect.

On substance, the current internet shutdown in Gujarat may fall foul to constitutional free speech standards on two counts:

a) overbreadth and

b) prior restraint.

The Gujarat authorities have directed telecos to disable access to 2G and 3G mobile data in the entire state. In addition, according to some reports, certain social media websites such as WhatsApp, Facebook, Twitter and Instagram have been specifically blocked. Thus, netizens may not even access these websites through broadband. While the wisdom of the move to disable social media is debateable, it is undeniable that the absolute constraint on data access is inimical to the central notion of Article 19(2). As the Supreme Court reiterated in Shreya Singhal, “restrictions on the freedom of speech must be couched in the narrowest possible terms”. Consider the celebrated SC decision in Kameshwar Prasad, where a rule, formulated in the interest of public order, forbidding participation in any demonstrations by public servants was challenged. After holding that demonstrations constitute expression, the Court went on to characterise the various kinds of demonstrations, finding that some may be peaceful, some passive and some capable of public disorder. Finally, the Court found the blanket-ban unconstitutional, concluding

The vice of the rule, in our opinion, consists in this that it lays a ban on every type of demonstration – be the same however innocent and however incapable of causing a breach of public tranquility and does not confine itself to those forms of demonstrations which might lead to that result.[1]

Now substitute “demonstrations” with “internet communications”. Such communications quite clearly stand as a form of speech under Art. 19(1)(a). Not all kinds of internet communications cause incitement or are capable of disturbing the public tranquillity. By directing a blanket ban on all mobile internet communications – innocent or otherwise – the Gujarat authorities’ actions are quite clearly overbroad. Its effects are not just felt by the protesters and rioters that it was meant for, but also hinders community life by hampering communication lines between ordinary citizens and debilitates economic life by demoralising businesses employing said ordinary citizens.

Further, the shutdown has been grounded on the public order/incitement restrictions of Art. 19(2) i.e. to prevent public disorder and the incitement to offences of fellow citizens. However, any public order speech restriction must be narrowly tailored and have “proximate and direct nexus with the expression”.[2] The Shreya Singhal judgment arrived at the Clear and Present Danger test in public order restrictions, expounded first by Justice Holmes in Schenck v US.[3] It may be relevant to note here that the US itself has abandoned this test for the more speech-liberal Brandenburg[4] test of imminent lawless action. The Indian SC recognised this as much in Arup Bhuyan. According to this standard, three conditions to be proven for a speech to constitute incitement;

Intent: That the speech must have the object of promoting violence

Imminence: That the speech must lead to imminent lawless action, and

Likelihood: That the speech was likely to create such lawless action.

The dichotomy in standard-setting aside, it is evident that not all communications over mobile internet have any connection with the Patidar reservation. In fact, most communication in any public order situation revolve around safety and emergency, with small pockets of incitement. Conversely, the blanket ban on all mobile texting conversation and blockage of social media in particular would fail the test of free speech.

But even beyond constitutional considerations, one must take note of the utilitarian benefits social media websites play during emergency situations. They may be used to mass-communicate first-hand information and alert authorities and first responders. In 2011, when an earthquake struck the US, more than 40,000 earthquake-related tweets were up on the web within a minute of the first shock. This was highly appreciated by Emergency Management professions. Closer home, the Nepal earthquake highlights the important role played by open data and social media in obviating confusion and consternation in the aftermath of the disaster. Facebook Safety Check was also instrumental in this regard.[5]

Now might be an appropriate time to traverse to our next argument on prior restraint. These may be defined as state restrictions on speech before its very publication. Prior restraints are generally regarded as unconstitutional under Indian law. In two celebrated decisions – Brij Bhushan and Romesh Thappar – the Supreme Court found prior restraints on print media to be generally unconstitutional. In Auto Shankar’s case, the Court went even further and held that defamatory materials of state officials may never be subject to prior restraints. The remedy of the officials would lie in post-publication prosecutions. However, in KA Abbas, the Supreme Court took a step back and struck a different balance between cinema and free speech. The Court decided to treat “motion pictures” differently from other forms of art and expression. The Court’s rationale arose

… from the instant appeal of the motion picture, its versatility, realism (often surrealism), and its coordination of the visual and aural senses. The art of the cameraman, with trick photography, vistavision and three dimensional representation thrown in, has made the cinema picture more true to life than even the theatre or indeed any other form of representative art. The motion picture is able to stir up emotions more deeply than any other product of art. … It is also for this reason that motion picture must be regarded differently from other forms of speech and expression. A person reading a book or other writing or hearing a speech or viewing a painting or sculpture is not so deeply stirred as by seeing a motion picture.

The Gujarat internet disruption portends important questions of prior restraint in an hitherto jurisprudentially unconventional medium; the internet. By disallowing access to mobile internet and blocking certain social media websites in particular, the authorities have no doubt imposed a wide array of prior restraint; foreclosing communication, business transactions, file sharing etc. Would mobile internet communication fall within the Brij BhushanRomesh Thappar print media standard of prior restraint, which lays an extremely strict scrutiny of speech restrictions? Or would file sharing, being a common feature of communication apps like WhatsApp/Instagram/Facebook/Youtube, attract the KA Abbas motion picture standard of prior restraint? Or would mobile internet communications warrant an all-together different prior restraint standard?

There appears to be no clear-cut answer yet. [Edit: Moreover, Nariman J., in Shreya Singhal accepted an intelligible differentia argument under Art. 14,[6] but limited it only to the creation of technology-specific offences.][7] The Court very clearly held that the threshold for curbing content on the internet cannot all-together be new. Yet the reasonableness of prior restraint standards over the internet still seems to be in limbo. To take an example, there are apps like Whatsapp that are *more* electronic in a sense than ordinary internet chatting apps that don’t allow file (photo/video) transfers. Would the latter merit the Brij Bhushan standard or the KA Abbas standard?

Concluding our two-part discussion on internet shutdowns, I observe the following:

  1. Internet shutdowns in India fall in a sort of regulatory no-man’s land. Accordingly, central, state, and local authorities are able to order them without thinking twice about accountability.
  2. The current internet disruptions in Gujarat do not distinguish between innocuous and incendiary speech, restricting all of them in a blanket-ban that is resonant of policies of dictatorial regimes such as Syria and Egypt. Devoid of being narrowly tailored, the shutdown is fatally overbroad.
  3. The prior restraint standards that may be applicable over internet communications must be addressed by the Courts/Parliament urgently. Sandwiched between conventional media and motion pictures, internet texting apps suffer from shades of grey that desperately need colouring.

Nakul Nayak was a Fellow at the Centre for Communication Governance from 2015-16.

[1] Kameshwar Prasad, page 384.

[2] S. Rangarajan v. P. Jagjivan & Ors., (1989) 2 SCC 574, para. 45.

[3] 63 L. Ed. 470, at 473-74.

[4] Brandenburg v. Ohio, 23 L. Ed. 2d 430 (1969).

[5] My thanks to Joshita Pai for pointing this to me.

[6] Shreya Singhal, paras. 97-98.

[7] I should thank Gautam Bhatia for correcting me on this.

(Nakul is a Research Fellow at the Centre)