Tag Archives: net neutrality

Net Neutrality and Public Highways

(My thanks to Malavika Prasad for bringing this case to my attention)

With the recent release of the Telecom Regulatory Authority of India’s [“TRAI”] “Consultation Paper” on the regulatory framework for over-the-top [“OTT”] services, net neutrality is up for sustained debate in India. Previously, I had written about how net neutrality, in the context of the internet, should be understood as a core free speech issue, and it might be helpful to consider the controllers of the “gateways” to the internet (or, in other words, the owners of the infrastructure of speech on the internet) as having public obligations of non-discriminatory access (even though they might be private parties). The idea of public obligations inhering upon private parties because of their control of public infrastructure, or their performing of a public function, has been upheld by the American Supreme Court in Marsh vs Alabama and by the Indian Supreme Court in the concurring judgment of Mohan J. in Unnikrishnan.

In the net neutrality debate this time, another bit of imagery has been doing the rounds: that of a public highway. It is permissible to charge a toll for the use of a highway, the rate of which might be proportional to how much one uses – so goes the analogy – but it is impermissible to charge differential toll rates based upon the make of the car you drive, or depending upon whether you’re going to work, or to vacation.

This is a particularly interesting analogy, because, as it turns out, the Indian Supreme Court has ruled upon precisely this point: that of non-discriminatory access to public highways. Saghir Ahmed vs State of UP, decided in 1954 by a five-judge bench, involved a constitutional challenge to the UP Road Transport Act, which allowed the State government to take exclusive control of running and operating road transport services within the state, if it believed such a step to be necessary in the public interest. Or, in other words, it allowed the State government to create a transport monopoly by executive fiat – which it actually did, for a part of the road network.

In deciding upon the validity of the State government notification, as well as the constitutionality of the Act, the Court noted:

“According to English law, which has been applied all along in India, a highway has its origin, apart from statute, in dedication, either express or implied, by the owner of land of a right of passage over it to the public and the acceptance of that right by the public . In the large majority of cases this dedication is presumed from long and uninterrupted user of a way by the public, and the presumption in such cases is so strong as to dispense with all enquiry into the actual intention of the owner of the soil and it is not even material to enquire who the owner was.

In response to the Attorney-General’s argument that the rights of commercial passage over a highway were determined by the Motor Vehicles Act, the Court observed:

“But the right of the public to use motor vehicles on the public road cannot, in any sense, be regarded as a right created by the Motor Vehicles Act. The right exists anterior to any legislation on this subject as an incident of public rights over a highway. The State only controls and regulates it for the purpose of ensuring safety, peace, health and good morals of the public. Once the position is accepted that a member of the public is entitled to ply motor vehicles on the public road as an incident of his right of passage over a highway, the question is really immaterial whether he plies a vehicle for pleasure or pastime or for the purpose of trade and business. The nature of the right in respect to the highway is not in any way affected thereby and we cannot agree with the learned AdvocateGeneral that the user of a public road for purposes of trade is an extraordinary or special use of the highway which can be acquired only under special sanction from the State.”

A few things ought to be noted:

(1) Although Saghir Ahmed was a case about State action, and consequently, implicated the petitioners’ Article 19(1)(g) and 14 rights, the Court’s logic here is based upon the nature of the utility (public highway) rather than the nature of the owner. In the first excerpted paragraph, the Court makes it clear that the question of ownership is immaterial, since whatever property rights the owner has, he is deemed to have intended to give up his right to the extent that passage requires.

(2) The power of the State to control and regulate the public utility must be for the purposes of ensuring safety, peace, health and morals.

(3) The nature of the use of the public utility (i.e., in this case, whether commercial or for pleasure) does not affect the scope of the right of use in any way.

(4) The right stems from long and uninterrupted prior use, presumably stretching back to the beginning of road networks, and consequently, being somehow part of the very nature, or essential characteristic, of a “road”.

Note the striking similarities with the net neutrality debate, with respect to each of the elements. There is, of course, a danger with pushing any analogy too far, but the vocabulary of the TRAI Consultation Paper itself conjures up an image of the internet “highway”. In paragraph 2, it states:

“The term over-the-top (OTT) refers to applications and services which are accessible over the internet and ride on operators’ networks offering internet access services e.g. social networks, search engines, amateur video aggregation sites etc.”

And, in para 3:

Carriage is separated from content in internet networks, enabling OTT content and application service providers to deal directly with end users.”

A full elaboration, of course, would need significant unpacking. What, precisely, is the public highway here? Is it the spectrum? And is the argument then that a spectrum auction by the original owner (i.e., the government) does not carry with it complete rights of ownership, but rather, attendant obligations that act as limits upon those rights. One of those obligations being to provide non-discriminatory access to a public utility, whose public character remains unchanged despite the ownership being in private hands. Of course, such an argument would also need to establish the analogy between roads and the internet, both in terms of their public character (perhaps not so difficult), and the establishment of a right of non-discriminatory access through a long period of uninterrupted usage (perhaps harder in the case of the internet).



Filed under Article 12: Meaning of "State", Internet Freedom, Net Neutrality, Public goods

Net Neutrality, Free Speech and the Indian Constitution – III: Conceptions of Free Speech and Democracy

(This post first appeared on the CIS blog, here)

In the modern State, effective exercise of free speech rights is increasingly dependent upon an infrastructure that includes newspapers, television and the internet. Access to a significant part of this infrastructure is determined by money. Consequently, if what we value about free speech is the ability to communicate one’s message to a non-trivial audience, financial resources influence both who can speak and, consequently, what is spoken. The nature of the public discourse – what information and what ideas circulate in the public sphere – is contingent upon a distribution of resources that is arguably unjust and certainly unequal.

There are two opposing theories about how we should understand the right to free speech in this context. Call the first one of these the libertarian conception of free speech. The libertarian conception takes as given the existing distribution of income and resources, and consequently, the unequal speaking power that that engenders. It prohibits any intervention designed to remedy the situation. The most famous summary of this vision was provided by the American Supreme Court, when it first struck down campaign finance regulations, in Buckley v. Valeo“the concept that government may restrict the speech of some [in] order to enhance the relative voice of others is wholly foreign to the First Amendment.” This theory is part of the broader libertarian worldview, which would restrict government’s role in a polity to enforcing property and criminal law, and views any government-imposed restriction on what people can do within the existing structure of these laws as presumptively wrong.


We can tentatively label the second theory as the social-democratic theory of free speech. This theory focuses not so much on the individual speaker’s right not to be restricted in using their resources to speak as much as they want, but upon the collective interest in maintaining a public discourse that is open, inclusive and home to a multiplicity of diverse and antagonistic ideas and viewpoints. Often, in order to achieve this goal, governments regulate access to the infrastructure of speech so as to ensure that participation is not entirely skewed by inequality in resources. When this is done, it is often justified in the name of democracy: a functioning democracy, it is argued, requires a thriving public sphere that is not closed off to some or most persons.

Surprisingly, one of the most powerful judicial statements for this vision also comes from the United States. In Red Lion v. FCC, while upholding the “fairness doctrine”, which required broadcasting stations to cover “both sides” of a political issue, and provide a right of reply in case of personal attacks, the Supreme Court noted:

“[Free speech requires] preserv[ing] an uninhibited marketplace of ideas in which truth will ultimately prevail, rather than to countenance monopolization of that market, whether it be by the Government itself or a private licensee… it is the right of the public to receive suitable access to social, political, esthetic, moral, and other ideas and experiences which is crucial here.”

What of India? In the early days of the Supreme Court, it adopted something akin to the libertarian theory of free speech. In Sakal Papers v. Union of India, for example, it struck down certain newspaper regulations that the government was defending on grounds of opening up the market and allowing smaller players to compete, holding that Article 19(1)(a) – in language similar to what Buckley v. Valeo would hold, more than fifteen years later – did not permit the government to infringe the free speech rights of some in order to allow others to speak. The Court continued with this approach in its next major newspaper regulation case,Bennett Coleman v. Union of India, but this time, it had to contend with a strong dissent from Justice Mathew. After noting that “it is no use having a right to express your idea, unless you have got a medium for expressing it”, Justice Mathew went on to hold:

What is, therefore, required is an interpretation of Article 19(1)(a) which focuses on the idea that restraining the hand of the government is quite useless in assuring free speech, if a restraint on access is effectively secured by private groups. A Constitutional prohibition against governmental restriction on the expression is effective only if the Constitution ensures an adequate opportunity for discussion… Any scheme of distribution of newsprint which would make the freedom of speech a reality by making it possible the dissemination of ideas as news with as many different facets and colours as possible would not violate the fundamental right of the freedom of speech of the petitioners. In other words, a scheme for distribution of a commodity like newsprint which will subserve the purpose of free flow of ideas to the market from as many different sources as possible would be a step to advance and enrich that freedom. If the scheme of distribution is calculated to prevent even an oligopoly ruling the market and thus check the tendency to monopoly in the market, that will not be open to any objection on the ground that the scheme involves a regulation of the press which would amount to an abridgment of the freedom of speech.


In Justice Mathew’s view, therefore, freedom of speech is not only the speaker’s right (the libertarian view), but a complex balancing act between the listeners’ right to be exposed to a wide range of material, as well as the collective, societal right to have an open and inclusive public discourse, which can only be achieved by preventing the monopolization of the instruments, infrastructure and access-points of speech.

Over the years, the Court has moved away from the majority opinions in Sakal Papers and Bennett Coleman, and steadily come around to Justice Mathew’s view. This is particularly evident from two cases in the 1990s: in Union of India v. The Motion Picture Association, the Court upheld various provisions of the Cinematograph Act that imposed certain forms of compelled speech on moviemakers while exhibiting their movies, on the ground that “to earmark a small portion of time of this entertainment medium for the purpose of showing scientific, educational or documentary films, or for showing news films has to be looked at in this context of promoting dissemination of ideas, information and knowledge to the masses so that there may be an informed debate and decision making on public issues. Clearly, the impugned provisions are designed to further free speech and expression and not to curtail it.

LIC v. Manubhai D. Shah is even more on point. In that case, the Court upheld a right of reply in an in-house magazine, “because fairness demanded that both view points were placed before the readers, however limited be their number, to enable them to draw their own conclusions and unreasonable because there was no logic or proper justification for refusing publication… the respondent’s fundamental right of speech and expression clearly entitled him to insist that his views on the subject should reach those who read the magazine so that they have a complete picture before them and not a one sided or distorted one…” This goes even further than Justice Mathew’s dissent in Bennett Coleman, and the opinion of the Court in Motion Picture Association, in holding that not merely is it permitted to structure the public sphere in an equal and inclusive manner, but that it is a requirementof Article 19(1)(a).

We can now bring the threads of the separate arguments in the three posts together. In the first post, we found that public law and constitutional obligations can be imposed upon private parties when they discharge public functions. In the second post, it was argued that the internet has replaced the park, the street and the public square as the quintessential forum for the circulation of speech. ISPs, in their role as gatekeepers, now play the role that government once did in controlling and keeping open these avenues of expression. Consequently, they can be subjected to public law free speech obligations. And lastly, we discussed how the constitutional conception of free speech in India, that the Court has gradually evolved over many years, is a social-democratic one, that requires the keeping open of a free and inclusive public sphere. And if there is one thing that fast-lanes over the internet threaten, it is certainly a free and inclusive (digital) public sphere. A combination of these arguments provides us with an arguable case for imposing obligations of net neutrality upon ISPs, even in the absence of a statutory or regulatory obligations, grounded within the constitutional guarantee of the freedom of speech and expression.


Filed under Free Speech, Internet Freedom

Net Neutrality, Free Speech and the Indian Constitution – II: Private parties, public obligations

(This post first appeared on the CIS website, here)

To sum up the previous post: under Article 12 of the Constitution, fundamental rights can be enforced only against the State, or State-like entities that are under the functional, financial and administrative control of the State. In the context of net neutrality, it is clear that privately-owned ISPs do not meet the exacting standards of Article 12. Nonetheless, we also found that the Indian Supreme Court has held private entities, which do not fall within the contours of Article 12, to an effectively similar standard of obligations under Part III as State organizations in certain cases. Most prominent among these is the case of education: private educational institutions have been required to adhere to standards of equal treatment which are identical in content to Article 14, even though their source lies elsewhere. If, therefore, we are to impose obligations of net neutrality upon private ISPs, a similar argument must be found.

I will suggest that the best hope is by invoking the free speech guarantee of Article 19(1)(a). To understand how an obligation of free speech might operate in this case, let us turn to the case of Marsh v. Alabama, an American Supreme Court case from 1946.

Marsh v. Alabama involved a “company town”. The “town” of Chickasaw was owned by a private company, the Gulf Shipbuilding Corporation. In its structure it resembled a regular township: it had building, streets, a sewage system, and a “business block”, where stores and business places had been rented out to merchants and other service providers. The residents of the “town” used the business block as their shopping center, to get to which they used the company-owned pavement and street. Highway traffic regularly came in through the town, and its facilities were used by wayfarers. As the Court noted:

In short the town and its shopping district are accessible to and freely used by the public in general and there is nothing to distinguish them from any other town and shopping center except the fact that the title to the property belongs to a private corporation.”

Marsh, who was a Jehovah’s Witness, arrived in Chickasaw with the intention of distributing religious literature on the streets. She was asked to leave the sidewalk, and on declining, she was arrested by the police, and charged under an anti-trespassing statute. She argued that if the statute was applied to her, it would violate her free speech and freedom of religion rights under the American First Amendment. The lower Courts rejected her argument, holding that since the street was owned by a private corporation, she had no constitutional free speech rights, and the situation was analogous to being invited into a person’s  private house. The Supreme Court, however, reversed the lower Courts, and found for Marsh.

Four (connected) strands of reasoning run through the Supreme Court’s (brief) opinion. First, it found that streets, sidewalks and public places have historically been critically important sites for dissemination and reception of news, information and opinions, whether it is through distribution of literature, street-corner oratory, or whatever else. Secondly, it found that private ownership did not carry with it a right to exclusive dominion. Rather, “the owners of privately held bridges, ferries, turnpikes and railroads may not operate them as freely as a farmer does his farm. Since these facilities are built and operated primarily to benefit the public and since their operation is essentially a public function, it is subject to state regulation.” Thirdly, it noted that a large number of Americans throughout the United States lived in company towns, and acted just as other American citizens did, in their duties as residents of a community. It would therefore be perverse to deny them rights enjoyed by those who lived in State-municipality run towns. And fourthly, on balance, it held that the private rights of property-owners was subordinate to the right of the people to “enjoy freedom of press and religion.”

No one factor, then, but a combination of factors underlie the Court’s decision to impose constitutional obligations upon a private party. It mattered that, historically, there have been a number of spaces traditionally dedicated to public speech: parks, squares and streets – whose public character remained unchanged despite the nature of ownership. It mattered that individuals had no feasible exit option – that is, no other place they could go to in order to exercise their free speech rights. And it mattered that free speech occupied a significant enough place in the Constitutional scheme so as to override the exclusionary rights that normally tend to go with private property.

The case of the privately-owned street in the privately-owned town presents a striking analogy when we start thinking seriously about net neutrality. First of all, in the digital age, the traditional sites of public discourse – parks, town squares, streets – have been replaced by their digital equivalents. The lonely orator standing on the soap-box in the street corner now tweets his opinions and instagrams his photographs. The street-pamphleteer of yesteryear now updates his Facebook status to reflect his political opinions. Specialty and general-interest blogs constitute a multiplicity of town-squares where a speaker makes his point, and his hearers gather in the comments section to discuss and debate the issue. While these examples may seem frivolous at first blush, the basic point is a serious one: the role of opinion formation and transmission that once served by open, publicly accessible physical infrastructure, held – in a manner of speaking – in public trust by the government, is now served in the digital world, under the control of private gatekeepers. To that extent, it is a public function, undertaken in public interest, as the Court held in Marsh v. Alabama.

The absence of an exit option is equally important. The internet has become not only space of exchanging information, but it has become a primary – non-replaceable source – of the same. Like the citizens of Chickasaw lacked a feasible alternative space to exercise their public free speech rights (and we operate on the assumption that it would be unreasonably expensive and disruptive for them to move to a different town), there is now no feasible alternative space to the internet, as it exists today, where the main online spaces are owned by private parties, and access to those spaces is determined by gatekeepers – which are the ISPs.

The analogy is not perfect, of course, but there is a case to be made that in acting as the gatekeepers of the internet, privately-owned ISPs are in a position quite similar to the corporate owners of they public streets Company Town.

In the last post, we saw how it is possible – constitutionally – to impose public obligations upon private parties, although the Court has never made its jurisprudential foundation clear. Here, then, is a thought: public obligations ought to be imposed when the private entity is providing a public function and/or when the private entity is in effectively exclusive control of a public good. There is an argument that ISPs satisfy both conditions. Of course, we need to examine in detail how precisely the rights of free expression are implicated in the ISP context. That is the subject for the next post.


Filed under Free Speech, Internet Freedom

Net Neutrality, Free Speech and the Indian Constitution – I

(This piece first appeared on the CIS website, here)

Net neutrality is rapidly becoming one of the most important issues facing internet governance and internet freedom today, and it is quite likely that it will soon raise issues of law and legal policy in India. In this post (and the next), I will discuss net neutrality, free speech and the Indian Constitution.

I will not here go into the debates surrounding the multiple meanings of the concept of “net neutrality” but take, for the purposes of this post, the following definition:

The idea that all Internet traffic should be treated equally is known as network neutrality. In other words, no matter who uploads or downloads data, or what kind of data is involved, networks should treat all of those packets in the same manner.

In other words, put simply, net neutrality requires the extant gatekeepers of the internet – such as, for instance, broadband companies – to accord a form of equal and non-discriminatory treatment to all those who want to access the internet. Examples of possible discrimination – as the quote above illustrates – include, for instance, blocking content or providing differential internet speed (perhaps on the basis of a tiered system of payment for access).

Net neutrality has its proponents and opponents, and I do not have space here to address that dispute. This post – and the next – are premised on the assumption that net neutrality is both an important and a desirable goal (this brief article in the Times of India provides a decent, basic primer on the stakes involved).

An example of net neutrality in practice is the American Federal Communications Commission’s Open Internet Order of 2010, which was the subject of litigation in the recently concluded Verizon v. FCCThe Open Internet order imposed obligations of transparency, no blocking, and no unreasonable discrimination, upon internet service providers. The second and third requirements were vacated by a United States Court of Appeals. The rationale for the Court’s decision was that ISPs could not be equated, in law, to “common carriers”. A common carrier is an entity that offers to transport persons and/or goods in exchange for a fee (for example, shipping companies, or bus companies). A common carrier is licensed to be one, and often, one of the conditions for license is an obligation not to discriminate. That is, the common carrier cannot refuse to carry an individual who is willing and able to pay the requisite fees, in the absence of a compelling reason (for example, if the individual wishes the carrier to transport contraband). Proponents of net neutrality have long called for treating ISPs as common carriers, a proposition – as observed above – was rejected by the Court.

With this background, let us turn to India. In India, internet service providers are both state-owned (BSNL and MTNL), and privately-owned (Airtel, Spectranet, Reliance, Sify etc). Unlike many other countries, however, India has no network-neutrality laws. As this informative article observes:

The Telecom Regulatory Authority of India (TRAI), in its guidelines for issuing licences for providing Unified Access Service, promotes the principle of non-discrimination but does not enforce it… the Information Technology Act does not provide regulatory provisions relating to Internet access, and does not expressly prohibit an ISP from controlling the Internet to suit their business interests.”

In the absence of either legislation or regulation, there are two options. One, of course, is to invoke the rule of common carriers as a common law rule in court, should an ISP violate the principles of net neutrality. In this post (and the next), however, I would like to analyze net neutrality within a constitutional framework – in particular, within the framework of the constitutional guarantee of freedom of speech and expression.

In order to do so, two questions become important, and I shall address them in turn. First, given that most of the ISPs are privately owned, how does the Constitution even come into the picture? Our fundamental rights are enforceable vertically, that is, between individuals and the State, and not horizontally – that is, between two individuals, or two private parties. Where the Constitution intends to depart from this principle (for instance, Article 15(2)), it specifically and expressly states so. As far as Article 19 and the fundamental freedoms are concerned, however, it is clear that they do not admit of horizontal application.

Yet what, precisely, are we to understand by the term “State”? Consider Article 12: 

“In this part, unless the context otherwise requires, the State includes the Government and Parliament of India and the Government and the Legislature of each of the States and all local or other authorities within the territory of India or under the control of the Government of India.”

The key question is what, precisely, falls within the meaning of “other authorities”. The paradigmatic example – and this is something Ambedkar had in mind, as is evidenced by the Constituent Assembly Debates – is the statutory corporation – i.e., a company established under a statute. There are, however, more difficult cases, for instance, public-private partnerships of varying types. For the last fifty years, the Supreme Court has struggled with the issue of defining “other authorities” for the purposes of Part III of the Constitution, with the pendulum swinging wildly at times. In the case of Pradeep Kumar Biswas v. Indian Institute of Chemical Biology, a 2002 judgment by a Constitution bench, the Court settled upon the following definition:

“The question in each case would be whether in the light of the cumulative facts as established, the body is financially, functionally and administratively dominated by or under the control of the Government. Such control must be particular to the body in question and must be pervasive. If this is found then the body is a State within Article 12. On the other hand, when the control is merely regulatory whether under statute or otherwise, it would not serve to make the body a State.”

Very obviously, this dooms the ISP argument. There is no way to argue that ISPs are under the pervasive financial, functional and administrative domination or control of the State. If we step back for a moment, though, the Pradeep Kumar Biswas test seems to be radically under-inclusive. Consider the following hypothetical: tomorrow, the government decides to privatize the nation’s water supply to private company X. Company X is the sole distributor of water in the country. On gaining control, it decides to cut off the water supply to all households populated by members of a certain religion. There seems something deeply wrong in the argument that there is no remedy under discrimination law against the conduct of the company.

The argument could take two forms. One could argue that there is a certain minimum baseline of State functions (ensuring reasonable access to public utilities, overall maintenance of communications, defence and so on). The baseline may vary depending on your personal political philosophy (education? Health? Infrastructure?), but within the baseline, as established, if a private entity performs a State function, it is assimilated to the State. One could also argue, however, that even if Part III isn’tdirectly applicable, certain functions are of a public nature, and attract public law obligations that are identical in content to fundamental rights obligations under Part III, although their source is not Part III.

To unpack this idea, consider Justice Mohan’s concurring opinion in Unnikrishnan v. State of Andhra Pradesh, a case that involved the constitutionality of high capitation fees charged by private educational institutions. One of the arguments raised against the educational institutions turned upon the applicability of Article 14’s guarantee of equality. The bench avoided the issue of whether Article 14 directly applied to private educational institutions by framing the issue as a question of the constitutionality of the legislation that regulated capitation fees. Justice Mohan, however, observed:

What is the nature of functions discharged by these institutions? They discharge a public duty. If a student desires to acquire a degree, for example, in medicine, he will have to route through a medical college. These medical colleges are the instruments to attain the qualification. If, therefore, what is discharged by the educational institution, is a public duty that requires… [it to] act fairly. In such a case, it will be subject to Article 14.

In light of Pradeep Kumar Biswas, it is obviously difficult to hold the direct application of the Constitution to private entities. We can take Justice Mohan, however, to be making a slightly different point: performing what are quintessentially public duties attract certain obligations that circumscribe the otherwise free action of private entities. The nature of the obligation itself depends upon the nature of the public act. Education, it would seem, is an activity that is characterized by open and non-discriminatory access. Consequently, even private educational institutions are required to abide by the norms of fairness articulated by Article 14, even though they may not, as a matter of constitutional law, be held in violation of the Article 14 that is found in the constitutional text. Again, the content of the obligation is the same, but its source (the constitutional text, as opposed to norms of public law) is different.

We have therefore established that in certain cases, it is possible to subject private entities performing public functions to constitutional norms without bringing them under Article 12’s definition of the State, and without the need for an enacted statute, or a set of regulations. In the next post, we shall explore in greater detail what this means, and how it might be relevant to ISPs and net neutrality.


Filed under Free Speech, Internet Freedom