In our discussion of surveillance on this blog, we have argued that although bulk metadata surveillance does not track and store the content of communication, it may nonetheless violate the freedom of expression and freedom of association by having a chilling effect. Briefly, the chilling effect – a doctrine that is used regularly in the United States, Canada, South Africa and the ECHR, to name just four jurisdictions – applies to cases where governmental laws and governmental (or private) activities are of a nature that – while not directly censoring free speech – nonetheless have the impact of self-censorship. The classic example is that of excessively vaguely worded libel laws – in order to keep on the right side of the law, citizens will end up refraining from engaging in completely legal and legitimate forms of speech. In the United States, we observed how disclosure of the member lists of organizations advocating (at the time) unpopular causes such as the NAACP was held to violate the First Amendment rights of association and expression because there was a probable effect that if individuals knew that a hostile government had access to their membership of such organizations, they would think twice before joining them. We noticed similar judgments in the context of revealing trade union membership, and so on. And we argued that S. 66A of the IT Act, which criminalized, inter alia, speech that is “annoying” is a classic case of a law that would exercise a deeply chilling effect on free expression.
In this context, it is surprising to note that the doctrine of the chilling effect has virtually no foothold in India. The phrase is mentioned in seventeen Supreme Court opinions, primarily in the context of how rent control legislation has a chilling effect on housing (this mischief seems to have been caused by a 1984 opinion, and Krishna Iyer J.’s perhaps somewhat overzealous penchant for evocative prose)! A few opinions refer to the chilling effect on government functionaries if all their action is subjected to public scrutiny – which, as well, is a different context. The Bombay High Court has two opinions, both of which are on rent control. The Calcutta High Court has four, two of which relate – again – to public officials, and two of which seem to have taken the phrase a little too literally, using it in the context of noise pollution caused by firecrackers! It is in one judgment of the Madras High Court that we find it used in a free speech context – that is, accompanying a citation of New York Times v Sullivan, the classic US case on free press and libel laws (the other Madras High Court judgment is on rent control). The Allahabad, Punjab & Haryana and Andhra Pradesh High Courts have a combined six judgments on rent control. In a solitary Karnataka High Court judgment, the petitioners make a 19(1)(a) argument in the context of over-zealous contempt of court laws, although the Court does not refer to it in its judgment (it does seem to accept the argument – or a variant of it – in substance) (the other two Karnataka High Court cases are about the conduct of public functionaries). As for the rest of the High Courts, there is a deafening silence.
The one exception is the Delhi High Court. In four judgments, all within the last ten years, the Court has directly recognized the impact of the chilling effect on the freedom of expression. Indeed, in Ram Jethmalani v Subramaniam Swamy, the Court not only cited the Sullivan doctrine, but read the Supreme Court case of R. Rajgopal to have incorporated it into Indian law! The Delhi High Court observed: “if a person is under fear of being sued, he may not express himself freely on public issues, and this would chill the public debate.” M.K. Sharma v Sangeeta Gupta is a case with a complex factual matrix, but in essence, the Court held that statements made to an investigating committee set up under the aegis of the Vishaka guidelines on sexual harassment cannot be subject to a defamation suit, since this would “inevitably have a chilling effect thus deterring an aggrieved person from availing of a remedy provided by the Supreme Court against sexual harassment.” In 2009, in Petronet v Indian Petro Group, the Court again cited Sullivan and Rajgopal, observing that the problem that both the US Supreme Court and the UK House of Lords were trying to address what could be “damaging self-censorship by the media to the impoverishment of political discourse.” Similar were the observations in Nirmaljit Singh v Yashwant Singh.
Thus, all the Delhi High Court cases refer to libel laws, two of them being about the media, and two of them about individuals. While the Delhi High Court cases represent a start, it must be noted that the chilling effect does not only operate where there is a direct threat of legal prosecution – if that was so, NAACP and the trade union cases would have been decided the other way. The doctrine is predicated upon the understanding that coercion can take place in various extra-legal ways: through reprisals in the workplace for putatively unrelated reasons, for instance. The basic point at issue is that if one is aware that the government (or other institutions that possess authority or power over one, or even society at large) know the details of the relationships one is entering into, one will think twice before associating with entities/persons that are deeply unpopular (and yet, perfectly legal).
Is there any discernible reason why this doctrine has not taken root in Indian free speech and free association law? Even our Southern neighbor, Sri Lanka, has recognized this: in Channa Pieris v Attorney General, responding to a case of governmental arrest and detention, of members of a movement called Ratawesi Perumana, the Court observed – in words that are directly on point, and which deserve to be stated in full:
“There was in the matters before us no direct call to desist from expressive activities… The fact that the respondents took no direct action to restrict the right of the petitioners and members of the Ratawesi Peramuna to associate freely in orderly group activity however, does not end the matter… Freedoms such as these are protected not only against obvious and heavy handed frontal attack, but also from being smothered or stifled or chilled by more subtle interference. We need to consider the probable deterrent effect of the arrest and detention even though such effect may have been unintended.
The Court then quotes Justice Harlan’s opinion in NAACP that we have discussed before, and goes on to hold:
“In the matters before us, the arrest and detention of the petitioners must be regarded as entailing the likelihood of a restraint upon the exercise by them of their right to freedom of association in much the same way as manifestations of hostility to their activities by the armed persons who stole their posters at Matara. The arrest and detention was in my opinion likely to adversely affect the ability of the petitioners to pursue their collective effort to foster beliefs which they admittedly have the right to advocate, in that it may induce some of them to withdraw from the Peramuna and dissuade others from joining it because of fear of police action. It must also have certainly had a chilling effect on the expressive and associational activities of those who had the temerity to continue to be members of the Ratawesi Peramuna.”
And this, surely, must be correct. A doctrine that finds firm constitutional ground in countries having constitutional cultures of free speech and association as diverse as the US and Canada, the ECHR and South Africa and Sri Lanka, is surely near-univeral in its scope. If it hasn’t yet found footing in India, perhaps that is simply because it hasn’t yet been litigated. The Delhi High Court cases, and the scattered observations from Madras and Karnataka do indicate, however, that the argument is there to be made; and what better context to make it in than that of bulk surveillance?