Internet Governance/Background and Key Concepts

What is “governance”?
The term governance often gives rise to confusion because it is (erroneously) assumed that it must refer solely to acts or duties of the government. Of course, governments do play an important role in many kinds of governance. However, in fact, the concept is far broader, and extends beyond merely the State. For example, we have seen increasing reference recently to the notion of “corporate governance”, a process that involves oversight both by the State and by a host of non-State bodies, including corporations themselves.

Don McLean points out that the word governance derives from the Latin word “gubernare”, which refers to the action of steering a ship. This etymology suggests a broader definition for governance. One important implication of this broader view is that governance includes multiple tools and mechanisms. Traditional law and policy are certainly among those mechanisms. However, as we shall see throughout this primer, governance can take place through many other channels. Technology, social norms, decision-making procedures, and institutional design: all of these are as equally important in governance as law or policy.

What is “Internet governance”?
This broader view of governance is particularly important when it comes to discussions of Internet governance. In recent years, the notion of Internet governance has become hotly contested, a subject of political and ideological debate. Many divisions can be identified:


 * Technical or holistic? Some people feel that Internet governance is a purely technical matter, best left to programmers and engineers; others argue for a more holistic approach that would take account of the social, legal and economic consequences of technical decisions.


 * What is the place of governments? Another division is between those who would give greater (or even sole) authority to national governments, and those who would either include a wider range of actors (including civil society and private sector) or eliminate government altogether. The traditional role of national governments is also challenged by the global nature of the Internet and the resulting importance of supra-national entities. In recent years, questions regarding participation by governments in these entities have become more frequent, with some arguing that the relevance of governments has diminished, and others suggesting a need for greater government participation.


 *  Evolutionary or revolutionary? A further split can be identified between those who believe existing institutions and laws can be modified to manage the Internet (the “evolutionary” approach), and those who believe that an altogether new system is required (the “revolutionary” approach).

Leaving aside the merits of these various positions, the disagreement itself suggests a certain conceptual confusion. It shows that despite over a decade of debate and discussion, Internet governance remains a work in progress, a concept in search of definition. Still, this primer begins from the premise that certain principles can be established to lay down a working definition of Internet governance.

It was precisely in search of such a working definition that, as part of the United Nations-initiated WSIS, a Working Group was established in 2004 and asked to develop a definition of Internet governance. In its report, issued in June 2005, WGIG proposed the following definition:


 * Internet governance is the development and application by Governments, the private sector and civil society, in their respective roles, of shared principles, norms, rules, decision-making procedures, and programmes that shape the evolution and use of the Internet.

For the purposes of this primer, it is the broad and holistic view indicated by this definition that will be used to discuss Internet governance. Two points in particular stand out and will recur in what follows. First, that Internet governance includes a wider variety of actors than just the government; actors from the private sector and civil society are also important stakeholders—and second, that Internet governance refers to more than just Internet domain name and address management or technical decision-making. Indeed, the report of the WGIG goes on to make it clear that Internet governance “also includes other significant public policy issues, such as critical Internet resources, the security and safety of the Internet, and developmental aspects and issues.” These and a variety of other issues will be considered throughout this primer.

What are “layers” and how are they relevant to Internet governance?
One way to conceptualize this more holistic approach is with reference to “layers” of governance. This method was originally proposed by law professor Yochai Benkler, who argued that modern communications networks should be understood as a series of “layers” rather than as an assorted bouquet of different technologies. Benkler lists three such layers: a “physical infrastructure” layer, through which information travels; a “code” or “logical” layer that controls the infrastructure; and a “content” layer, which contains the information that runs through the network.

Today, it has become fairly common to conceptualize the Internet in this fashion. Some would change the names of the layers, and others would include additional layers. The important point, however, is not which specific layers we choose, but the more general point that the Internet can be broken up into discrete analytical categories; and that, consequently, Internet governance itself takes place on multiple levels (or “layers”). In taking a holistic approach to governance, therefore, it is critical that we consider multiple layers.

In this primer, we focus on the three original layers mentioned by Benkler: infrastructure, logical, and content. Each of these layers, displayed in Figure 1, is discussed in greater detail in Section II. That section will also discuss the range of actors involved in governance at each layer (see also Appendix 1).

Figure 1: Internet Governance Issues by Layer

What is ICT and what is its relevance to the Internet?
ICT is an acronym that, unpacked, means “Information and Communications Technology”. As defined on Wikipedia,


 * ICT is the technology required for information processing. In particular the use of electronic computers and computer software to convert, store, protect, process, transmit, and retrieve information from anywhere, anytime.

The use of ICT as a general term is sometimes criticized for a lack of precision, but it is becoming increasingly common given the growing reality of digital convergence. Convergence refers to the phenomenon by which various different forms of digital content (voice, data, rich media like movies and music) can now not only travel along the same physical infrastructure but also be managed and manipulated together by the same systems. It will be covered more fully in Section IV, but what is important to understand here is that, in discussing the Internet, we are in fact discussing several underlying technologies and means of access. This is particularly relevant at the infrastructure layer, where a variety of technologies are deployed. Indeed, the Internet is accessed through a range of infrastructures including traditional telecommunications, cable, satellite, and various wireless methods. Internet governance will therefore have an impact on all these underlying technologies.

Why is the Internet difficult to govern?
The conceptual confusion over the meaning of Internet governance also stems from the fact that there exists no single central authority or mechanism – no traditional form of “governance” – with responsibility for all aspects of the network. This lack of a single authority is, in part, due to historical reasons but also due to the network’s technical architecture, which makes it very difficult to exert control. Unlike traditional networks (a telephone network or an early office LAN, for example), the Internet is not reliant upon a central server. Instead, the Internet is a network of autonomous networks, and control rests with the various distributed facilities that, together, make up the collaborative resource referred to as the Internet. For this reason, the Internet is said to be empowered at the edges (i.e., at the individual facilities); it is also sometimes defined as an end-to-end (e2e) network.

The e2e nature of the network is largely a result of its technical design, and particularly of its packet-based data transfer. Using the TCP/IP (Transmission Control Protocol/Internet Protocol) suite, messages on the Internet are broken up into individual packets of data; the network is generally neutral with regard to these packets, and simply routes them using the most efficient path available, without regard to content or origin. This means that intelligence on the Internet rests at the edges: the power to innovate, to create new applications and services or types of content, rests with individual users. The network is also said to be “dumb” with regard to the content that it carries; as long as data packets fit into the basic TCP/IP protocol, the network simply routes them along, without discrimination or control.

The Internet’s open standards and e2e model are at the root of its tremendous success and power to drive innovation. However, they are also why the Internet is so hard to manage. On a neutral network, there exists no gatekeeper or central authority to verify the contents of a packet. Viruses, spam, pornography, voice packets (from phone calls), and innocuous email messages: all of these are treated equally. In addition, the fact that multiple pathways exist to route packets from one source to another makes it very difficult for any party to block information; the packets will simply find another route.

This dilemma – the same technical architecture that allows the Internet to flourish also permits a number of harmful activities – is at the heart of many current debates over Internet governance. While the need for some form of control to limit harmful content is widely recognized, there is also widespread agreement that governance mechanisms should facilitate and not compromise the Internet’s core technical architecture. In particular, solutions must be found that maintain the principle of e2e and the underlying open standards upon which it is based.

What is the history of governance on the Internet?
As noted, there exists no central authority on the Internet. Instead, there exists a multitude of actors, institutions and bodies, exerting control or authority in a variety of ways, and at multiple levels. This does not necessarily imply anarchy, as some may suggest; these participants in general have formal and well-defined roles, and they address specific tasks or responsibilities. Section II contains a more detailed discussion of some of these actors and the issues they address. Here, we identify certain key milestones in Internet governance.

Somewhat surprisingly, given the Internet’s eventual distance from the state, the network actually began as a government project. In the 1960s, the US Defense Department sponsored the development of the ARPANET by the Defense Advanced Research Projects Agency (DARPA). ARPANET was a distributed network that was designed to foster communication between research centres. Soon, this network, which remained under the control of the US government, was being used by a wider set of users, particularly in the academic community. In the 1970s, DARPA also developed a mobile packet radio network and a packet satellite network. These were conceptually integrated into an Internet in 1973. In 1983, the operational Internet was launched. The National Science Foundation Network (NSFNET) joined the Internet in 1986, spreading access to the Internet to an international community of users.

The Internet Activities Board was formed in 1984 and was made up of a number of task forces. One of these became the Internet Engineering Task Force (IETF) in 1986. It was created to manage the development of technical standards for the Internet. It represented early seeds of “governance” on the Internet, albeit of a unique kind. IETF governed through a consultative, open and co-operative approach. Decisions were made by consensus, and involved a wide variety of individuals and institutions. This freewheeling and decentralized decision-making process remains in many ways the hallmark of Internet governance; it accounts for a significant amount of resistance to any attempt by national governments to exert control over the network.

The Domain Name System (DNS) was developed in the mid-1980s. It was managed by the Internet Assigned Numbers Authority (IANA) at the University of Southern California’s (USC’s) Information Sciences Institute under US Government contract for many years. The root servers of the DNS were operated voluntarily by 13 different organizations. The root zone file, defining the top level domains of the DNS was distributed by IANA to the root server operators. In 1994, the US Government outsourced the management of the DNS to a private entity, Network Solutions. This led to a considerable degree of dissatisfaction among other members of the Internet community, who feared that the network would become too commercialized. A long process of conflict and deliberation led ultimately to the creation, in 1998, of ICANN, a non-profit corporation charged with managing the Internet’s DNS. ICANN inherited the responsibilities of IANA in the course of its creation. Please refer to Appendix 2 for more information regarding the roles and functions of ICANN and IANA.

The creation of ICANN, in many ways, initiated the latest round of debates over Internet governance. Although created with primarily a technical mandate (i.e., managing the DNS, the allocation of Internet address space, and the recordation of parameters unique to the Internet protocol suite), ICANN quickly became embroiled in a host of controversies. Among other things, critics charge the organization with a lack of democracy and transparency, with being too close to the US Government, and with a perceived exclusion of voices from the developing world.

Recently, controversy over the need for and nature of Internet governance has combined with changes in the underlying nature of the network itself to suggest that a new model of governance may be required. These changes include the tremendous growth of the network, and the increasing reliance of our social, economic and even political lives on what WGIG calls a “global facility.” Together, all these events have called into question the relatively informal, consensual and trust-based models of governance that currently exist, and have led to exploration of new or modified governance models.

In 2002, the UN General Assembly took an important step towards such a new model when it called for a global conference, WSIS that would consider alternatives and ways to increase participation by developing countries. The first WSIS meeting was held in 2003, in Geneva, where representatives adopted a Declaration of Principles and a Plan of Action. A follow-up meeting is scheduled for November 2005 in Tunis.

For the moment, the WSIS process remains incomplete, and its final outcome is not yet clear. Nonetheless, certain issues have risen to the top of the agenda, and are likely to feature prominently in future discussions. These include debates over the respective roles and authority of ICANN, the International Telecommunications Union (ITU) and, especially, national governments. Many developing countries, in particular, would like to see the role of States extended, while others remain wary of granting governments too much control. In addition, questions regarding participation by developing countries and the Internet’s role in fostering human and social development are also on the agenda. We discuss these issues more fully later, in Section III.

Should there be governance on the Internet?
The Internet’s architectural and ideological foundations – open standards, e2e, empowered users, absence of control – have from the start bred a certain libertarian streak that rejects any attempt to exert control, particularly by the government. Since the very success of the network stems from an absence of the control, the argument goes, government intervention would only stifle the network.

The most famous expression of this extreme form of libertarianism remains John Perry Barlow’s “Declaration of the Independence of Cyberspace,” quoted in the Introduction. At the start of that manifesto, Barlow declares that:


 * Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather.

Today, it is apparent that his vision was somewhat utopian. Nonetheless, many less radical observers continue to believe that the Internet’s success depends on keeping governance (by the State, or by any other authority) to a minimum. Some observers have invoked traditions of the commons, or the public forum, to envision a virtual space where ideas are exchanged freely, without rules, without regulation, and without controls.

These ideas are important to acknowledge because the Internet’s success does, to a significant extent, depend on its free and open culture. Nonetheless, it is also clear that the absence of rules can be as detrimental to this commons as the existence of bad rules; anarchy is as harmful as stifling regulation. The right question is, therefore, not whether there should be governance but rather what constitutes good governance. The goal for any governance mechanism should be to balance rules and freedom, control and anarchy, process and innovation.