23 May 2024

Internet Fragmentation and Human Rights: How to counter threats to an open, interoperable and global Internet

By Michaela Nakayama Shapiro and Sheetal Kumar

In a range of digital governance forums, including the Internet Corporation for Assigned Names and Numbers (ICANN), the Internet Engineering Task Force (IETF) , and the  Internet Governance Forum (IGF), the term ‘Internet fragmentation’ is being used with increasing frequency. It is the subject of session proposals and new research papers, which frame the threat it poses and suggest ways forward. But what does it mean, and how as human rights defenders should we approach it?

‘Internet fragmentation’ is a contested term. Depending on the source, it has also been referred to as ‘Internet balkanisation’ and the ‘splinternet’. In 2016, the World Economic Forum published a paper that defined Internet fragmentation “as any efforts or actions taken to impair seamless flow of data between Internet devices, whether by intervening in the common standards and protocols which are essential for exchange of data, or by limiting the types of data that an end user can receive or the ways in which they can receive it”. 

More recently, the term has received increasing attention due to the growing frequency of Internet shutdowns, as well as new regulation impacting data flows. These developments are fuelling a growing concern that the Internet is changing fundamentally—from an open network of networks through which data flows seamlessly and end-users play a pivotal role in its governance to one that is defined or mirrored by territorial boundaries, and less defined by its users—instead, controlled in a top-down manner by government or corporate actors.

Certain large-scale events have also shaped the Internet fragmentation narrative. For example, in 2022, following the invasion of Ukraine by Russia and the subsequent request by Ukraine that Russian ‘.ru’ domain names be disconnected, the possibility of the Internet being fragmented—in this case because users within certain jurisdictions would be disconnected en masse—received renewed attention. Although the European Internet registry did not honour the request, the request itself highlighted the fragility of the open Internet.

 

What is Internet fragmentation? 

As noted above, there is no one definition of Internet fragmentation. However, different definitions can be placed on a spectrum, ranging from narrower to broader conceptualisations. It is GPD’s view that employing an informed, broader conceptualisation is necessary to ensure a comprehensive and timely response to Internet fragmentation. However, it is important that this approach makes clear the links between different aspects of fragmentation, to avoid confusion and to support focused and effective action.

At the narrow end of the spectrum, the focus is on the extent to which the Internet enables connection and facilitates data transit across networks, e.g whether connection points within the Internet’s infrastructure and networks remain intact and interoperable, as well as the continued functioning of protocols and standards that allow for data to flow across networks. However, this perspective risks not sufficiently accounting for incremental changes at different layers that could threaten the technical underpinnings of the internet. 

A broader understanding of Internet fragmentation that considers how the increase in certain actions that limit user access—such as Internet shutdowns and regulation that interrupts data flows, such as data localisation laws—are, or can be, fragmentary is important. Proponents of this approach consider not only the impact of actions by governments and other actors on the Internet’s infrastructure, but also the impact of these actions on end users of the Internet and on their ability to connect, as well as their impact on the broader system of Internet governance. The concern is that such actions and measures by governments and other actors (like companies) could, collectively and over time, impact the Internet’s technical layer in a way that impairs its openness, interoperability and global nature. 

This spectrum for understanding internet fragmentation is captured in a framework developed by the IGF’s Policy Network on Internet Fragmentation (PNIF), which categorises internet fragmentation as: fragmentation of the user experience, technical fragmentation, and governance fragmentation. As defined by the PNIF, these elements refer respectively to the international human rights framework and the need to maintain a free flow of data; the interoperability of the global Internet infrastructure; and the coordination and multistakeholder governance of the Internet. 

 

Where are we seeing Internet fragmentation?

As noted above, there is strong disagreement as to whether (and if so, to what extent and where) there is currently fragmentation at the technical level. Although the impact on the technical layer is not immediate, there are concerning trends that collectively, over time, can threaten the technical layer and the critical properties of the internet. Increased centralisation and top-down control of the Internet—including but not limited to its technical infrastructure—can undermine the original vision of the Internet as a decentralised and open technology. 

There are five critical properties of the Internet, all enabling its openness and interoperability: a common protocol; a general purpose network; a common global identifier system; decentralised management and distributed routing; and a layered architecture of interoperable building blocks.

One example of how these are threatened can be seen in the proposed changes to the European Union’s eIDAS (Electronic Identification And trust Services) regulation. eIDAS aims to establish requirements and standards for qualified certificates for website authentication. However, proposed changes to article 45 of the regulation would both duplicate existing standards for the issuance and revocation of digital certificates and would shift the power of certification from browsers to governments by allowing member states to introduce their own root certificate authorities and requiring browsers to add them to their root trust stores for European users. This risk to both the governance model of the Internet and its technical underpinnings can also be seen in EU proposals to enforce ‘sender-pay’ agreements, which impact the decentralised management of the internet (one of the critical properties) by altering the nexus of control over the technical architecture of the Internet by providing telecommunication providers with the ability to control traffic flows. 

However, it is not only economic trends and government regulation that are driving internet fragmentation: since 2018, representatives of a number of  Chinese companies have put forward a set of proposals at the International Telecommunications Union (ITU) and the IETF to establish an ‘alternative’ network architecture and set of protocols from the current system. Known as ‘new IP,’ this alternative architecture would also alter the network layer to allow it to carry greater information about the content and identity of users, which would pose a risk to privacy and impact the decentralised management and an open architecture of the Internet. Increasing efforts to restrict data flows or control what data individuals can access online can also have an impact on the technical layer of the Internet. Network shutdowns, DNS-based blocking (among other censorship tools), firewalls, and walled gardens all represent efforts by governments to prevent users’ access to certain information, with potential impacts on the technical layer of the internet if more widely and consistently adopted.

 

How is Internet fragmentation relevant to human rights? 

An open, interoperable and global Internet is a key enabler of human rights in the digital age. This is because an open, interoperable and global internet has become a prerequisite for accessing, creating and sharing information in the digital age. If the Internet’s technical layer is no longer open or interoperable, then this will fundamentally impact individuals’ right to seek, receive and impart information (Article 19(2) of the International Covenant on Civil and Political Rights (ICCPR). 

Being unable to access accurate information can further impact individuals’ rights to health (Article 12, International Covenant on Economic, Social and Cultural Rights), and to free and fair elections, (Article 25, ICCPR); and being unable to express oneself and be heard online can impact individuals’ rights to freedom of association (Article 22, ICCPR) and to peaceful assembly (Article 21, ICCPR). Some drivers of Internet fragmentation are also linked to attempts to control access to information within territorial borders and extend state control and visibility of Internet access. This also impacts the right to privacy (Article 17(1), ICCPR). Privacy-violating measures whose increased/normalised use could lead to Internet fragmentation include the use of deep-packet inspection firewalls, and insistences that companies store consumer data or information on local servers (data localisation).

While individually these examples do not pose a direct threat to the technical architecture of the Internet and its governance model, collectively and over time they can undermine the Internet’s openness, interoperability and global nature. These actions could accelerate Internet fragmentation by normalising the control of access of information within borders and by impairing the rollout of protocols and standards that support interconnectivity and interoperability. Employing only the narrowest definitions of Internet fragmentation could undermine much needed actions to both preserve and strengthen the norms and standards that are needed to protect the open Internet, and ensure it evolves in a way that is rights-respecting and inclusive. In this way, countering Internet fragmentation trends and ensuring the free exercise of human rights are mutually reinforcing efforts: through safeguarding one, the other is also further strengthened. 

A comprehensive, rights-respecting approach to countering Internet fragmentation—which seeks to anchor both technical standards and regulatory frameworks and policy in human rights—is therefore needed. Such an approach would include commitments to align technical standards with regulatory frameworks and policy and would thus protect human rights in the digital age. For this reason, GPD is a strong advocate for a holistic approach to protecting an open, interoperable and global internet that addresses all three categories of fragmentation as outlined in the PNIF framework. This requires protecting (1) the technical layer of the internet; (2) user rights; and (3) the multistakeholder approach to internet governance. 

Yet, as outlined above, coming to a consensus on a single definition of Internet fragmentation might be a challenge. However, this is not necessary: for the purposes of protecting and preserving an open, interoperable and global Internet, having a set of key features of the Internet we want to preserve and for which stakeholders can jointly advocate is more important.

What can stakeholders do to counter Internet fragmentation and protect an open, interoperable and global Internet?

Building on the Internet Society’s five “critical properties” of the Internet and the PNIF framework, stakeholders can work to find common ground around shared advocacy objectives. As part of this task, stakeholders should work together to foster a greater shared understanding of both current and emerging threats to an open Internet, as well as a shared commitment among stakeholders to rights-respecting norms in relation to Internet governance. The Internet governance community could also agree upon a set of clear definitions of the elements of the Internet that require protection, which would support more focused action on threats to an open Internet. 

Below we outline a set of five key objectives for Internet governance for the community to rally behind that both centre human rights and help to counter fragmentation.

  1. Commit to preserving and strengthening the multistakeholder model, with a focus on ensuring policymaking and standard-setting processes are open, inclusive, consensus-driven and transparent.
  2. Promote efforts to expand meaningful connectivity to the Internet: Access must be coupled with meaningful use of the Internet, which is enabled by digital literacy, trust, and safe, secure online environments
  3. Protect and promote the global free flow of information, ensuring that the economic and social benefits of the Internet and related digital technologies continue to flourish and support the UN Sustainable Development Goals.
  4. Ensure policy and regulation promote user control and interoperable data flows: this means protecting the Internet’s key characteristics and fundamental properties while avoiding implementing any unnecessary barriers to the global free flow of information.
  5. Protect existing open Internet technical standards and commit to open technical standards for the Internet itself and digital technologies: technical standards should be general-purpose, scalable, interoperable, have multiple implementations and be developed in an inclusive, multistakeholder process.

Achieving these objectives will require cooperation across stakeholder groups. 

In the coming years, there is a high likelihood of increased standardisation of digital technologies as well as increased regulation. There is also an increased likelihood of corporate centralisation and consolidation, and attempts at addressing this through regulation which could have unintended negative effects on an open Internet. For this reason, stakeholders must be meaningfully engaged in standards-developing processes and regulatory efforts with the view to identifying threats to an open Internet— and should realise these threats might be inadvertent. As part of this, it is critical to avoid the risk of duplication in standards development and to support interoperable technical standards.

Stakeholders must also work together to overcome silos in the Internet governance landscape. With the upcoming World Summit on the Information Society (WSIS) +20 Review process and the Summit for the Future, there is a real opportunity to reassert and recommit to the multistakeholder governance model and recognise the need for greater coordination within the Internet governance ecosystem. These junctures offer opportunities to strengthen Internet governance norms and to push for continued support for an open, interoperable and global Internet that is underpinned by human rights and multistakeholder governance.