This is the last post in our series of posts studying the access to the digital economy in BRICS countries. In previous posts, we have studied the access to the internet, the access to decent work, the access to do digital business, the access to consumer rights, and the access to welfare. In keeping with our attempt to broaden the understanding of access to include access for people in all their roles, in this post, we examine the access to the freedom of information and expression in the digital economy in BRICS countries.
Free access to information and expression are both independently important for a healthy digital economy. In addition, both these types of access are interdependent. Without free access to information, the right to freedom of expression is ineffectual. Article 19 of the UN Declaration of Human Rights states,
“Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.”
This right contains the right to seek and receive information regardless of frontiers as part of the right to freedom of expression. Freedom of access to information is impeded by both state and non-state actors, including through restrictive intellectual property regimes, vigilante-led intimidation, the prevalence of illiteracy, and overt state censorship. Thus, we examine some impediments to free access to information along with expression in BRICS countries.
The notion of intellectual property (IP) has its roots in censorship. Scott (2001) has shown that the early examples of patents include patents granted to printing presses that would print only approved books. Copyright law, along with the custom to have the printer’s and author’s name included in a book, was developed to assign accountability per laws against heresy, sedition and treason.
Mainstream economic analysis too recognises that IP rights often create inefficiency. Boldrin and Levine’s (2002) seminal work shows how current IP regimes confuse the property rights over objects embedded with ideas, with the property rights over ideas themselves. That is, current IP regimes allow for control over the use of ideas after the sale of an object, creating an information monopoly ostensibly to reward innovation. This strand of enquiry recognises that IP rights are a government-enforced monopoly on the use and spread of ideas.
IP regimes in their implementation exhibit a clear Global North and Global South divide. Global South countries, particularly BRICS countries, have been at the forefront of defending public healthcare against IP rights. The most recent example of such action has been the demand by India, South Africa and other southern countries for a temporary waiver on patents for Covid-19 vaccines, so as to increase access. There was a clean divide between North and South countries at the WTO over this issue, with northern countries arguing for the continuation of patents. The same divide was evident in the debate on IP rights on genetic resources, with corporations and research institutions from northern countries patenting biological and traditional knowledge from southern countries on the basis of the claim that they discovered this knowledge. In general, empirical work demonstrates that the adoption of IPRs in developing countries is due to external pressure rather than domestic needs, or that developing countries are policy takers in the domain of IPR.
Another significant impediment to access to digital information is the lack of accessibility for persons with disabilities. Further analysis on this impediment can be found in Vidhi’s other research, including research on making cities accessible by design and an upcoming project on digital accessibility for persons with disabilities.
The right to freedom of expression can also be restricted by itself, including through liability on intermediaries like social media companies. Intermediary liability describes the extent to which intermediaries are liable for content created by their users. An unreasonably high level of intermediary liability incentivises intermediaries to delete or disallow content that could potentially be illegal, hampering free speech. An unreasonably low level of intermediary liability can allow harassment and hate speech to flourish, chilling free speech. We analyse the different levels of intermediary liability chosen by BRICS countries, differentiating by the type of intermediary. We also briefly analyse content moderation laws and rules in BRICS countries and how they work to either stifle speech or protect people from online harms.
Part I of our post broadly examines the access to freedom of information in the digital economy in BRICS countries in the context of digital information, through the lens of copyright laws. Part II of our post broadly examines the access to freedom of expression in the digital economy in BRICS countries.
Part I: Access to freedom of information
The Law on Copyright and Neighbouring Rights, 1998 (Brazil Copyright Law or BCL) regulates copyright law in Brazil. Brazil’s laws on copyright infringement create a very restrictive framework for the digital sharing of information. Under the BCL, a copyrighted work may not be reproduced without the express consent of the author of the work, save for a few limited exceptions.
The BCL protects the author’s economic rights first and foremost. Under Section 29, any form of reproduction, translation, distribution on the internet, indirect use in broadcasting or storage on a computer system must have the express consent of the author. These provisions set out a high standard of copyright protection for various use cases. Additionally, reproducing a work with an intent to explain or comment on it without taking permission from the author, generally permitted under the doctrine of fair dealing, is expressly prohibited under Section 33 of the BCL. Under the BCL, any annotations or commentaries must be published separately. This distinguishes the BCL from other liberal copyright jurisdictions such as India which permits reviews, comments or explanations as part of fair dealing of copyrighted works for non-commercial purposes.
Brazil’s laws provide safe harbour provisions and a clearly defined intermediary liability framework. Article 18 of the Marco Civil da Internet, 2014 (the Civil Rights Framework or CRF) offers blanket protections to any intermediary for any damages resulting from actions of third parties. The intermediary is only liable when a court order directing the intermediary to block access to the infringing content has not been carried out. However, Article 19(2) of the CRF does note that intermediary liability and takedown of content relating to copyright offences must be balanced against the freedom of expression provided in Brazil’s constitution. In 2015, Brazil’s Superior Court of Justice held that Orkut could not be held liable for its users sharing links that violate copyright laws. Applying the doctrine of contributory negligence, the court ruled that Orkut did not provide any infrastructure to download content or share files. Therefore, it merely acted as a platform and did not play a part in its user’s actions. This robust approach towards offering safe harbour protection to intermediaries is part of a series of verdicts protecting platforms from liability, balancing the need to protect the right to expression against the right to privacy. In a landmark decision in 2018, the Court held that the CRF requires intermediaries to take down content specified in a court order. It required Facebook to provide a URL locator to link the specific content to be taken down per the court order, failing which, Facebook could be personally held liable for continuing to host the content.
Russia follows continental European law on copyrights, as opposed to the common law form of copyright law followed in England and other common law countries. As can be seen from its provisions, digital sharing of information is permitted through Russia’s jurisprudence. Digital sharing of information in accordance with copyright law is moderately restrictive, with a textual approach to copyright law as opposed to a doctrinal approach which would allow greater flexibility. Recent introductions of intermediary liability set out stringent penalties for enabling digital sharing of information violative of Russian copyright law.
Under Article 15.2 of the Federal Law on Information, Information Technologies and the Protection of Information, 2006 (Russia IT Law), a copyright holder may approach the Roskomnadzor (the Russian Federal Service for Supervision of Communications, Information Technology and Mass Media) notifying the copyright infringement by any service provider on the Internet illegally offering access to the copyrighted work. Within 3 days of providing this notice, the Roskomnadzor shall send a notice to the entity hosting the copyrighted work, informing them of the infringement and directing them to block access to the infringing content. If the hosting service provider or the owner of the site refuses to block the content within the given timelines, the internet service provider is obliged to do the same within 24 hours of receiving a notice from the Roskomnadzor. The copyright holder may also directly approach the owner of the website seeking takedown of the copyrighted material who is obliged to take down the infringing content within 24 hours, under Article 15.7 of the Russia IT Law.
Part IV of the Civil Code of the Russian Federation (CCRF) provides for intellectual property regulations. This code defines copyrights and the rights of authors, while simultaneously providing for limited cases of fair dealing. Section 1274 of the CCRF sets out the conditions under which copyrighted information can be distributed freely for informational, scientific, educational or cultural purposes. This provision also covers the adaptation of copyrighted works to allow persons with disabilities to access the work. Significantly, this allows the fair dealing of information for distribution in terms of parodies or caricatures, with court decisions allowing parody or satire as exempted from copyright. Fair dealing under Russian copyright laws are specified in the legal text, and as such, are limited in their scope, unlike the doctrine of fair dealing followed in some other countries.
Section 1299 of the CCRF protects the concept of digital rights management (DRM), wherein the copyright holder may use technological measures to prevent or restrict the sharing or storage of copyright works without the permission of the owner. While violations of DRM are not permitted, this provision accommodates the idea of fair dealing to an extent. Users seeking to remove the DRM from a copyrighted work for uses covered under fair dealing clauses may require the author to lift the DRM. This is not mandatory in all cases, and the author is only required to lift the DRM for fair dealing cases if it is technologically possible and does not require significant costs.
Under the Russian Anti-Piracy Law 2013, courts are authorised to pass preliminary interim orders protecting the rights of copyright holders of films and television works from being circulated on the internet or other telecommunication networks. These provisions are similar to the Digital Millennium Copyright Act, 1998 applicable in the US. Article 15 allows copyright holders to seek take-down orders from courts if copyrighted works are distributed on the Internet or other telecommunication networks in violation of the exclusive rights held by the copyright holders. Additionally, this law sets out the liability of intermediaries involved in carrying such works, and provides safe harbour provisions based on their level of involvement and response to take-down requests from the copyright holders. However, the Anti-Piracy Law has prescribed stringent penalties for intermediaries that violate copyright laws by carrying or distributing such content illegally, including the power to order the relevant hosting service to permanently block the intermediary’s site on the internet.
The Copyright Act, 1957 (Copyright Act India or CAI) regulates the protection of copyrights in India. The CAI has been amended six times since it has been notified, with the latest major amendments in 2012. Section 52 of the CAI provides for a wide range of fair dealing exceptions to copyright infringement, including review, commentary, educational and research uses, increasing accessibility, and making three-dimensional objects from a two-dimensional artwork, allowing reverse engineering of technical devices. These instances allow various levels of access to copyrighted works for research, reporting or criticism. In 1995, the Delhi HC interpreted Section 52 as a tool that protects the freedom of expression guaranteed under the Indian Constitution. In Section 65A, the CAI prohibits circumvention of technological protection measures on copyrighted works. However, the provision excludes any circumvention for the purposes covered under fair dealing under the CAI.
In 2016, the Supreme Court delivered a landmark verdict regarding the law of fair dealing in The Chancellor, Masters & Scholars of the University of Oxford v. Rameshwari Photocopy Services. The appellants were a consortium of major international publishing houses seeking a permanent injunction on the defendant’s photocopy services that reproduced, printed, sold and distributed the appellants’ publications as course packs in the University of Delhi. Under Section 52(1)(i) of the CAI, the reproduction of any copyrighted work by a pupil in the course of instruction was permitted as fair dealing. Keeping this in mind, a single-judge bench ruled in favour of the defendants, and refused to grant a permanent injunction. While a division bench has set aside this ruling to allow the parties to appeal, the parties in this case have since chosen to withdraw and not to pursue this case anymore.
Section 52(1)(c) permits intermediaries to host or store copies of copyrighted works, provided that the intermediary is not aware that it is an infringing copy. However, this awareness must be actual knowledge regarding the specific infringing content. In 2016, the Delhi High Court upheld the safe harbour provisions enjoyed by intermediaries under the Information Technology Act, 2000, in the case MySpace v. Super Cassettes Industries Ltd. The court held that unless intermediaries have actual knowledge, they cannot be held liable for merely providing a social media website that was subsequently used by its users to share infringing content. Under the CAI, the owner of the copyright is required to provide this knowledge to the intermediaries through a written complaint. Upon receiving a written complaint on the infringement, the intermediary must block access for a maximum of twenty-one days while the owner may obtain a court order confirming the blocked access. If the intermediary does not obtain the court order, access to the content can be permitted by the intermediary.
The Copyright Law of the People’s Republic of China, 1990 (Copyright Law of China or CLC) sets out protections for copyrighted works. The CLC was first notified in 1990, with major amendments in 2001, 2010 and most recently in 2020. The latest amendments to the CLC, effective from June 2021, expand the scope of copyrightable works, reshape the scope of fair dealing, and drastically increase the penal consequences of copyright violations.
Article 3 of the CLC defines works as original expressions of intellectual achievements in certain fields such as written, musical, photographic or graphic works, computer software etc. Previously an exhaustive list, the definition of works has been expanded by the latest amendment to the CLC to include ‘other intellectual achievements conforming to the characteristics of works’. This expansion acknowledges that advances in technology allow for developing intellectual property works in myriad ways, and avoid hard-coding certain forms of works as ‘protected works’.
Article 24 of the CLC allows the use of copyrighted works without obtaining permission from the author in certain circumstances. Unlike Brazil’s copyright law, the CLC allows fair dealing of copyrighted works as long as the copyrighted work being commented on has been appropriately cited. Additionally, changes to the fair dealing exceptions allow dissemination of copyrighted works to persons with print disabilities such as visual impairments. The copyright holder’s rights have also been strengthened under the 2021 amendments, requiring that the exceptions for fair dealing must not in any way affect the normal use or legitimate rights of the copyright holder.
Like the Russian CCRF, the CLC recognises the concept of DRM as a legitimate means to protect copyrighted works by rights holders. Unlike the Russian copyright law which requires permission from the author for fair dealing exceptions to DRMs, the CLC expressly permits unilateral circumvention of DRMs in certain limited cases. This includes the use of published works for classroom teaching or scientific research, providing published works to print-disabled individuals without making profits, testing of security standards or encryptions in computers, and research on reverse engineering computer software.
The Regulations on Protecting Right to Dissemination via Information Networks, 2006 were notified in 2006 and were amended in 2013 to clarify the role of intermediaries in copyright protection. Under Article 22, an intermediary or a network service provider providing or allowing storage of works or audio-visual recordings would not be liable if it was not aware of any infringement of copyright, did not gain any economic benefit from it, and deleted the works or recordings upon receiving a notice from the rights holder. Article 23 reflects similar safe harbour provisions for any network service provider that allows searching or provides links to any URLs linking to any works being disseminated in a manner that infringes the holder’s rights. However, the intermediary would be jointly liable if it was aware of the infringement and continued permitting the infringing work to be hosted and available.
The Copyright Act, 1978 (South Africa Copyright Act or SACA) sets out the copyright law in South Africa. The SACA was amended in 2013 to introduce provisions that protect any form of works recognised as having indigenous origin by South Africa’s indigenous communities. The Amendment Act of 2013 creates a National Trust Fund for Indigenous Knowledge and a National Database for Indigenous Knowledge to further protect these works.
Section 12 of the SACA prescribes exceptions to the general protection of copyrighted works under the fair dealing doctrine. The practices permitted are materially similar to the provisions under Indian law, with a few distinctions. For example, while the use of a literary work for teaching purposes is permitted, the law is not as liberally worded as India’s provisions. In 2016, the Pretoria High Court decided a landmark case on the interpretation of fair dealing in news reporting under the SACA, in the case of Moneyweb v Media24. In this case, the appellants alleged that seven articles published by the defendants were copies of their original work. In the case, the Court ruled that only three of the seven articles published by the appellants were original, and warranted copyright protection. Of those three, the Court held that the portions used by the defendants in two articles were not substantial portions, and would therefore not be considered an infringement of copyright of the original articles. The last article was copied verbatim, requiring a consideration of fair dealing as a defence. The Court ruled that for the infringing article, the defendants created a substitute of the original article, published within seven hours of the original article being published, and was reproduced verbatim. This ruling marks a beginning of fair dealing jurisprudence in South Africa, and may be highly relevant for future cases in discussing the ‘fairness’ of an infringement.
The ‘fair dealing’ provision in the SACA stays silent on adaptations made for copyrighted works to be accessible for the differently abled. In 2021, this was challenged in the Pretoria High Court by Blind SA, an advocacy group in Blind SA v Ministry of Trade, Industry and Commerce. Blind SA argued that the SACA was unconstitutional in its current form as it violated the rights of those suffering from print or visual impairments by barring or severely limiting parties from adapting work into more accessible formats, including digital formats such as screen readers or audio descriptors for broadcasted works. It argued that currently less than 0.5% of all published works are available in accessible formats like Braille. In this case, the High Court held the provisions of the SACA that bar adaptations for the differently abled as unconstitutional. In September 2022, the Constitutional Court of South Africa upheld the High Court’s verdict, but suspended the verdict for 24 months to allow the Parliament to cure this defect. In the interim, the Constitutional Court read in certain definitions and a new provision allowing persons with print or visual impairments to immediately be able to make accessible adaptations without being prosecuted for copyright infringement. This landmark verdict provides an example of the urgency needed for approaching questions of access to information and education.
The South Africa Copyright Amendment Bill, 2017 (SACAB) attempts a major overhaul of the SACA, in line with the principles of equality and protection to the marginalised under the South African Bill of Rights. The SACAB introduces provisions permitting adaptations that allow increased accessibility for persons with print or visual impairments, paving the way for South Africa to sign and ratify the Marrakesh Treaty to Facilitate Access to Published Works for Persons who are Blind, Visually Impaired or Otherwise Print Disabled, 2013, and to formalise the Constitutional Court’s ruling in Blind SA. Additionally, the SACAB introduces exceptions to copyright infringement in terms of educational materials and resources, by permitting course packs to be reproduced in the course of instruction. This clause aims to provide widespread access to affordable education for marginalised groups in South Africa. Parallels can be drawn between this clause and the ruling in the DU photocopy case by the Delhi High Court in India, which cited similar reasons for its verdict holding that course packs published in the course of instruction did not amount to copyright infringement. In September 2022, the SACAB was passed by the National Assembly (the lower house of the SA Parliament), and has now been sent to the National Council of Provinces.
Part II. Access to freedom of expression
In Brazil, the CRF offers a regulatory framework for conduct on the internet, setting out rights of users and obligations of intermediaries. As discussed above, the CRF describes the parameters of intermediary liability and offers safe harbour clauses for takedown of content when the intermediary receives a court order. The CRF also protects victims of revenge pornography, or other unauthorised publication of personal images. Article 21 of the CRF states that the intermediary platform shall be held liable if it fails to block access or take down any images, videos or other material containing nudity or sexual acts, after it has so been notified by the person in question. In addition to this existing internet regulatory framework, Brazil has had several court decisions and key legislations proposed in recent years.
Brazil’s courts have taken an approach favouring freedom of speech and expression online, particularly from the vantage points of journalistic rights and public interest. In the case of Fernando Capez v. Juca Kfouri, the appellant, an elected representative, had approached the court stating that the defendant, a sports journalist, had written various blogs that harmed his reputation and honour. The appellant requested the court to order the defendant to abstain from writing any further offensive pieces. While this prayer was granted by the trial court, it was overturned by the São Paulo Court of Appeals which reasoned that ordering the defendant to abstain from future posts that may hypothetically be offensive would amount to censorship.
Recent court decisions have expanded the right to freedom of expression in response to issues raised by digital content. In the case of Manoel Conde Neto v. Folha de São Paulo, the appellant, a politician, had obtained a takedown order against a popular newspaper regarding an article published by the latter. The São Paulo Court of Appeals reversed the injunction, holding that the publication of true and socially relevant news stories could not be justified, and was akin to censorship. In 2021, the case of Nelson Curi et al v. Globo Comunicação e Participações brought forward the issue of the right to be forgotten against the public’s right to know. The appellants requested takedown of images of their relative’s murder in 1958, broadcast by the defendants in a television program. The court declined to issue a takedown order, stating that the relevant program was of public interest and could not be suppressed under the right to be forgotten or the broader right to privacy.
However, the recent elections in Brazil have brought forward interesting developments regarding content takedown and misinformation. In October 2022, Brazil’s Supreme Electoral Court approved a resolution allowing itself to unilaterally order tech companies to take down online posts, videos and attack advertisements that spread misinformation. Companies that failed to comply with these immediate takedown orders (issued where the content has already been declared as misinformation, but has been published through a new channel or account) within two hours risked suspension of their services across Brazil. These checks have been defended as attempts to reduce the damage caused by the virality of misinformation. However, critics have noted that this dramatic expansion of the electoral court’s powers sets a concerning precedent that may be used to censor legitimate criticisms.
Brazil has also introduced two legislative measures that may redefine this space. On September 9, 2021, the president of Brazil signed off on new provisions to the CRF.Under these provisions, intermediaries were restricted from acting against accounts that violated their terms of service, and were only authorised to act against any account or digital content in limited cases including violations of intellectual property rights or in response to a court order. However, these new provisions have been rejected by the National Congress of Brazil, and the changes introduced to the CRF under the measure have been nullified.
In Russia, the Russia IT Law regulates online content and its dissemination. Under Article 10.4.1, platforms offering encrypted messaging services must provide federal agencies the ability to decode these encrypted messages. Blocking or requiring deletion of online content is preceded by entry into an automated information system register, containing all network addresses of sites and pages that carry prohibited information.. Sites that distribute content relating to drugs, child pornography, suicide, gambling and lotteries, e-commerce online sale of restricted medicines or alcohol and home-made explosives are included in this registry. Under Article 15.1, the hosting service provider must inform the owner of the website to take down the page containing the prohibited content, failing which, the hosting service provider is obliged to block access to the site. If the hosting provider fails to act within the stated timelines, the internet service provider is required to block access to the site. Within 3 days of removing the prohibited content or receiving a court order cancelling the inclusion within the prohibited content registry, access to the website shall be unblocked. In 2019, similar takedown procedures have been implemented under Article 15.1-1 for information that shows ‘a clear disrespect for society, the state or the bodies exercising state power in the Russian Federation’. Such provisions, which are vaguely worded and present a danger of arbitrary use, introduce a chilling effect to the freedom of expression within Russia.
In recent years, Russia has moved to curtail freedom of expression through amendments in its criminal code pertaining to various current developments. In December 2020, Russia introduced amendments punishing libel committed in the context of spreading false information to undermine someone’s reputation, libel committed through the use of Internet, libel regarding accusations of someone having committed crimes against sexual freedom of a person and lastly, punishing slander regarding accusations of someone suffering from a disease that poses a danger to others. These newly criminalised forms of speech carry sentences ranging between two and five years of imprisonment, fines ranging between 500,000 to 5 million roubles, or forced labour. While it is not known whether the specific reference of libels regarding allegations of sexual misconduct, the use of the Internet or the effect on someone’s reputation are connected to the #MeToo movement, these amendments result in criminalising the very form and freedom of expression used throughout that movement. The timing of the amendment regarding criminalising slander that accuses another person of having a disease that may cause harm to others, may be noted in line with the COVID-19 pandemic. In the same period, Russia introduced amendments to its code on administrative offences, inserting fines for entities that refuse to comply with directions regarding deletion of web pages or blocking access to sites or pages carrying prohibited content. For legal entities, these fines range from 3 million roubles to 8 million roubles for the first offence, and between 10 percent to 20 percent of the revenues of the entity in the previous year for repeated violations. Pursuant to these amendments, the Roskomnadzor has notified TikTok and Russian social media network VKontakte in January 2021, asking it to take down content that involves minors in illegal actions, including calls to participate in ‘illegal’ or ‘unauthorised’ rallies. These instances clearly indicate the use of such amendments to censor calls for rallying in favour of Russian opposition figures, and evidence the use of civil and criminal penalties to censor and suppress free expression.
In March 2022, Russia introduced amendments in its criminal code, adding new punishments for spreading false information or discrediting the use of armed forces by Russia, along with publicly calling for restrictive measures by foreign states/entities against Russia. These provisions have ostensibly been included in the context of the Russia-Ukraine conflict, and negatively affect the freedom of expression and freedom of the press reporting from the frontlines in Russia. As recently as October 2022, reports indicate the use of speech laws to fine platforms such as TikTok, Twitch and the Wikimedia Foundation for offences such as ‘promoting LGBT propaganda’ and discrediting the Russian war effort.
In India, digital content is regulated by the Information Technology Act, 2000 (IT Act). The IT Act sets out penal provisions for certain forms of digital content. Section 66E of the IT Act protects violations of privacy, by punishing any sharing of intimate pictures of a person without their consent with fine or imprisonment up to three years. Similarly, Section 67 and 67A punish the sharing of any obscene, lascivious material or content involving sexually explicit acts with fines or imprisonment of up to five or seven years. Section 67B criminalises the creation or sharing of child pornography content with fines or imprisonment of up to five or seven years. Section 69A empowers government agencies to direct intermediaries to block access to contents on the grounds such as sovereignty, security of the state, and preserving public order. Section 79 of the IT Act offers safe harbour provisions for intermediaries, offering them refuge from liability if they acted as a mere conduit. However, a platform may be held liable if they have been notified by a government agency to delete or block access to content hosted by the intermediary. The provisions relating to blocking access to content or content takedown, in the context of intermediary liability, have been actively used by the Indian government in recent years. Information disclosed by Twitter for the period between July 2021 and December 2021 show India having been the world’s fifth largest requestor for removal of content. The figures released by Twitter also indicate the proactive use of the IT Act for taking down content posted by verified journalists and news outlets, with India leading the world with 114 such demands in that period, followed by Turkey (78) and Russia (55).
Section 66A of the IT Act sought to punish certain forms of digital content such as grossly offensive messages or messages sent to cause annoyance, inconvenience, danger, criminal intimidation, enmity, hatred or ill-will. However, in the landmark verdict of Shreya Singhal v. Union of India, the validity of this provision was appealed before the Supreme Court. The Court remarked on the chilling effect caused by such over-broad grounds as well as the subjective nature of offensive or annoying content. Stating that the provision criminalises several forms of speech that are legal under the right to free speech in the Indian constitution (Article 19(1)(a)), the Court ruled this provision as wholly unconstitutional and void.
In 2021, India issued the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules (IT Rules 2021). The IT Rules 2021 set out extensive obligations for digital platforms. Rule 3(1)(d) requires intermediaries to delete or disable access to unlawful content within thirty-six hours of receiving a court order or directions from a government agency under Section 79 of the IT Act. Rule 4 sets out additional requirements for large intermediaries, such as tracing the first originator of information based on a court order or directions issued under Section 69 of the IT Act, having a local office in India, appointing nodal contact persons for coordination with law enforcement agencies, and proactively monitoring digital content for information regarding rape or child abuse. Lastly, the IT Rules 2021 regulate dissemination of news by digital publication houses or news aggregators, and sets out a code of conduct and an oversight mechanism for news regulation by the central government. This latter provision was stayed by various high courts across India, with the Madras HC expressing concerns regarding the potential of government censorship over journalism.
In October 2022, the IT Rules were amended to include a grievance appellate committee (GAC), which is responsible for handling grievances against intermediaries and is chaired by three members appointed by the central government. The GAC offers a direct route for aggrieved persons to appeal decisions taken by the intermediary, as opposed to the alternative route of approaching the court. However, the completely executive nature of this committee overseeing all intermediary grievances, and the lack of natural justice principles such as fair hearings for both parties, raise concerns regarding the power of the central government to influence the conduct of intermediary platforms.
China takes a restrictive approach towards content moderation, freedom of expression and online anonymity. Under Article 24 of the Cybersecurity Law of the People’s Republic of of China, 2017 (‘China Cybersecurity Law’ or ‘CCL’), network operators that provide messaging services, social media services or website registration services are required to obtain the real identity of their users prior to signing the agreement to use these services.
This measure was followed by the following regulatory changes regarding internet posts, comments on internet posts, user accounts, and content moderation requirements in 2017. Article 8 of the Provisions on the Management of Internet Forum Community Services, 2017 similarly impose a ‘real name on file, user choice on profile’ approach, requiring service providers to register and authenticate users in order to make posts on their forum. In the same year, China issued the Provisions on the Administration of Internet Thread Commenting Services (China Comments Regulation) to regulate content being shared through comments on internet posts. Under Article 5(1), platforms must similarly register and verify the real identity of the commenter, who may then choose to use any name for their public-facing profiles. Users who do not submit their real identities shall not be permitted to comment on internet posts. Article 5(3) requires all websites, apps and other intermediary platforms that allow comments to be posted by users to pre-review comments on news-related posts. This is a marked departure from the ‘mere conduit’ approach taken towards several intermediary platforms across the world.
This requirement to review comments ex ante is currently limited to news information threads. In June 2022, the Cyberspace Administration of China issued draft guidelines seeking to amend the China Comments Regulation. In Article 5(4) of the draft guidelines, a pre-review of comments on all online posts must be conducted by the entity offering comment services. This obligation requires platforms to maintain teams that can conduct this editorial exercise on user-submitted comments, and report unlawful comments to the relevant authorities.
Additionally, Article 8 requires platforms to perform credit assessments of the users’ comments behaviour and to designate the scope of services and functionality of users based on this credit score. Article 8 further requires platforms to blacklist users with low credit scores and prevent them from making new accounts to post comments. Such measures are very susceptible to being misused in the context of internet censorship, particularly with the vagueness attached to the phrase ‘unlawful’ information for comments to not be published, as opposed to defining a clear criteria under which comments shall be rejected by the platform prior to publication.
In 2021, China introduced Provisions on the Administration of Public Account Information Services for Internet Users, aiming to regulate ‘public’ user accounts on social media platforms. Under these regulations, public accounts refer to any user account whose content is freely available to the public, and is not restricted to limited people (such as the ‘private’ mode on Instagram). In addition to the existing registration and authentication requirements for creating a public account, platforms that allow public accounts must verify the descriptions, names and avatars used by public accounts to ensure its accuracy. This move is designed to ensure that public accounts are not run by impersonators and do not falsely describe themselves or affiliate themselves with the government. Under Article 9, users creating public accounts to post content relating to economics, education, medical health, or justice must furnish proof of their professional background and qualifications to the platform. These requirements are meant to tackle the spread of misinformation and to ensure the traceability of users publishing content for public consumption.
Lastly, in August 2022, the Provisions on the Management of Internet User Account Information come into effect. This regulation seeks to create a regulatory framework for user accounts across all internet services. Reiterating existing real-name verification and registration requirements for creating user accounts, Article 12 of this regulation states that ISPs shall display the IP address of user accounts on their profile page, in order to ‘facilitate the public’s supervision in the public interest’.
These measures introduce a grave chilling effect to the freedom of expression and limits the ordinary citizen’s access to only state-approved information. With every aspect of user interaction (account creation, public accounts, posts and comments) within a social media platform structure being regulated, China has a robust, yet restrictive and limiting, set of provisions pertaining to access to information and freedom of expression.
In terms of protecting civil rights of internet users online, Article 36 of the Tort Liability Law of the People’s Republic of China, 2010 states that the users and intermediaries bear tort liability for infringing on the civil rights of other users. Under this law, the person affected may inform the intermediary to take measures such as content takedown, blocking access or suspension of services to the infringing material. The intermediary shall be held liable if they do not act on such information in a timely manner, or if they are aware of a user using their services to infringe civil rights of other users without taking the necessary measures. These provisions were further bolstered with respect to minors through the Law of the People’s Republic of China on Protection of Minors (China Minor Protection Law or CMPL), effective from June 2021. Article 69 of the CMPL requires facilities providing internet access to children to install child-friendly software and technological measures. Digital service providers must obtain consent from parents and guardians in the event of the child user being below the age of 14. Article 74 prohibits digital service providers from providing products and services meant to induce child users to indulge in the internet. Under this provision, video game providers, social media and audio/video content providers are required to build in mechanisms to curb usage by children, including through time limits, additional authority and spending limits. Children playing online games must be registered on a central electronic identity authentication system. Lastly, Article 77 protects children from cyber-bullying and harms to children by authorising affected minors or their guardians to require the game provider to stop the harm through blocking, disconnection of services or deletion of harmful content.
The process for content takedown requests in South Africa are highlighted in the Electronic Communications and Transactions Act, 2002 (ECTA). Under Section 73 of the ECTA, ISPs are exempted from liability regarding content shared on their platform, provided that the intermediary acted as a mere conduit and did not modify the data being shared. Similarly, an ISP is not liable for content hosted by a user, provided that it was not aware of such hosting having infringed the rights of a third party, and upon receiving a takedown notice from the victim has acted quickly to delete or block access to that content. These takedown notices can also be shared with representative bodies for ISPs such as the ISPA (the Internet Service Providers’ Association), which is the recognised representative body for South African ISPs, and limits liability of its members. Lastly, keeping in mind the notice-and-takedown structure of the ECTA, ISPs are specifically exempted from any general obligation to monitor data being transmitted or stored. This provision allows for greater trust regarding privacy and data protection in communications within South Africa, positively affecting the freedom of expression for South African internet users.
In 2016, South Africa released the Prevention and Combating of Hate Crimes and Hate Speech Bill for public comments. Clause 4 of this Bill defines hate speech as any communication that, for a reasonable person, can be understood to incite harm or promote hatred based on certain grounds including race, disability, gender identity, nationality or religion. The provision punishes the creation or spread of hate speech on the internet with fine and/or imprisonment up to 3 years for a first offence, and fine and/or imprisonment up to 5 years for subsequent offences. However, this Bill continues to be in legislative limbo, with parliamentarians stating, in September 2022, that they continue to work on the provisions to finalise a working draft.
However, recent legislative introductions raise concerns regarding their effect on the right to free expression. In March 2022, amendments to the Films and Publications Act, 1996 (SAFPA) came into effect, seeking to regulate cyber-harms perpetrated through sharing of videos and photos. Section 18F of the SAFPA prohibits sharing of any private sexual photos or films without the consent of the individuals involved. Section 18G prohibits distribution of films depicting sexual violence and violence against children while section 18H prohibits distribution of games, films or publications that incite imminent violence, advocate hate speech or serve as war propaganda. These prohibitions include distribution in any medium, including the internet as a whole. Section 15A of the SAFPA authorises compliance officers to pull access from any commercially distributed films and games that fails to comply with classification or legal requirements. Additionally, section 18E of the SAFPA allows any person to complain to the FPB regarding any unclassified content, prohibited content, or even for potentially prohibited content uploaded online, including by non-commercial online distributors, i.e., content creators sharing online content for personal purposes. If the FPB finds merit in the complaint – in terms of legal violations or a lack of classification – it may order the content to be taken down following the process under the ECTA. For this reason, critics have raised concerns regarding the wide-ranging requirements of content creators having to seek classification from the Films and Publications Board (FPB) for all video content, which may include content streamed or shared on popular social media platforms such as Twitch, Tiktok, Facebook or Instagram.
While more barriers exist to the right to the freedom of expression in BRICS countries, the right to the freedom of information also faces significant challenges. A Global North push in this direction seems to have made the BRICS intellectual property landscape much more restrictive. As for the freedom of expression, barriers seem to be growing but are sometimes met with resistance. Much of the law in these areas has been defined by courts, demonstrating that governments find it difficult to frame law for all scenarios in the digital economy.
As this post brings our series to a close, we have realised that such descriptive work relating to the digital economy in BRICS countries can be useful. As more countries look to join the BRICS alliance and it rises in prominence again, it is important for researchers to continue to study these countries comparatively. Analysis of India’s policy actions in comparison to countries that are placed somewhat similarly has also been fruitful in different ways in comparison to analysis that positions India alongside the US or the EU. We hope to continue this work in more detail, and we welcome feedback and suggestions.
(The authors would like to thank Prachi Mathur and Diksha Singh for their valuable research assistance towards this article.)