1500

FIVE Confronting Alignment The best that can be said for alignment is that it adjusts the capabilities of the Internet to existing rules and authorities. Instead of new and untested rules and messy new transnational combinations of decision makers, it draws on seemingly familiar, well-established institutions and processes. In principle, it allows each country to go its own way, ensuring an overall diversity in policy, and autonomy for various polities. Working against these comforting advantages is the fundamental clash between alignment and the capabilities of digital technology. Alignment undermines the offering of global services, sacrifices technical and economic efficiencies, and limits innovation and new entry into markets. It empowers repressive governments and protectionist interests by insulating people from access to information environments outside their own country. And while doing this it does not avoid conflicts over jurisdiction. Alignment is the digital equivalent of building customs checkpoints, tariffs, and road blocks into the network – with the proviso that in the digital environment the effects of any locality’s barriers will be felt globally as well as locally. Longer term, it could lead to the erosion of open and uniform technical standards, by encouraging each nation-state to mistrust or restrict access to capabilities that have not been approved, certified, or developed nationally.

Unfortunately there has been little progress in dealing with alignment. One reason is that the problem has not been correctly identified. Alignment is misunderstood as “fragmentation,” and policy discussions of “fragmentation” include phenomena as diverse as the digital divide, language differences among people, putting commercial applications behind paywalls, migration from IPv4 to IPv6, data localization laws, and temporary Border Gateway Protocol (BGP) routing configuration errors.16 With such an incoherent understanding of the problem, it is unlikely that there will be a fruitful discussion of solutions.

In this chapter I briefly examine some of the ideas that have been floated about responses to the so-called fragmentation problem. Due to space limitations, the comparisons cannot be comprehensive, evidencebased evaluations; rather, they are quick critiques that indicate why this book pursues other options.

Enhanced international legal cooperation One of the most common proposed solutions to the problems of fragmented jurisdiction is to advocate legal harmonization. This refers to reformed, more efficient methods of legal cooperation among states. At its most ambitious this involves international treaties; more modest efforts in this direction include the reform of Mutual Legal Assistance Treaties (MLATs) (Swire and Hemmings, 2015) or attempts to standardize state-based requests to take down content emanating from platform providers in other jurisdictions (De La Chapelle and

.

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

Fehlinger, 2016).

The fundamental problem with solutions of this type is that they are conservative; that is, they build on existing institutions and try to preserve as much of the Westphalian status quo as possible. They do not ask why we need to preserve that status quo. The previous chapter described the “jurisdictional paradox” created by the attempt to reconcile global cyberspace and territorial sovereignty. Restated, it means that strict adherence to jurisdictional boundaries in the application of states’ Internet policies limits the effectiveness of those policies because of the way global Internet connectivity provides ways around them. The attempt to make the policies more effective leads them inexorably towards assertions of extraterritorial jurisdiction. Thus, the harder states try to assert territorial jurisdiction over the Internet, the more they transgress and undermine it.

Proposals for transnational due process or MLAT reform do not avoid this paradox. They must confront the question of whose law applies. Proposals for “transnational due process” tend to ignore the differences in substantive law and legal rights among countries. What happens if a censorious regime in, say, Turkey asks an American social media platform to censor a tweet or take down a website that would be legal in America. If the US applies foreign law to the data/enforcement requests, it is in effect giving these states extraterritorial jurisdiction – not just over the data of the service provider but, more importantly, over their users. On the other hand, if US companies continue to offer the most popular online platforms and the US government applies its own law to foreign requests for data or enforcement actions to those platforms, then it is globalizing US laws and procedures. Many states, especially the authoritarian ones, may not get what they want out of an MLAT request as a result, and thus are likely to continue to rely on data localization, protectionism, and other alignment mechanisms to enhance their control. Even assuming that the MLAT process can be made faster and more efficient, it is hard to see how it would overcome the jurisdictional paradox.

MLAT reform may provide some mild improvements among like-minded western states. But we are unlikely to ever get a comprehensive, binding multilateral treaty around Internet governance as a whole. It is the power imbalances among states and their inability to agree on Internet policy and law, after all, that produces the pressures toward alignment to begin with. Even if all state parties agreed to negotiate in good faith, governments have too many conflicting economic and technological interests, and their political values are too divergent, to arrive at a comprehensive multilateral legal solution.

Jurisdiction-based blocking and filtering of content, for example, is a response to the fact that we do not have global institutional capabilities to control Internet content, and no consensus among states about what should be controlled. A World Economic Forum paper characterizes efforts to legitimate and globalize censorship via international agreements as a form of “fragmentation,”17 but this is a confusing use of the term. Insofar as national filtering constitutes fragmentation, it already exists in dozens of countries, and is widely viewed as a sovereign right of each country. Thus, legitimizing it via international agreements does not increase the level of fragmentation. Indeed, if governments can agree on what to block and succeed in globalizing censorship, it might actually reduce fragmentation, if fragmentation means different

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

views of the Internet from different jurisdictions. The real policy conflict is about censorship, not fragmentation, and if one wants to respond to and resist such controls, the difference cannot be camouflaged as a debate about fragmentation. Countries that fear free expression and openness to the world are not going to change their approach simply because their critics invoke the F-word.

Another example concerns cyber-attacks and national security. The US and some of its militarily strong allies want to retain the right to respond militarily to severe cyber-attacks, whereas states with weaker militaries want to legally separate cybersecurity from traditional military self-defense and security concerns. The best the international system can do to regulate cyber-attacks is to hold recurring nonbinding talks like the Global Conferences on CyberSpace or the UN Group of Governmental Experts on cyber norms.18 Bilateral discussions among states may also bear fruit. These inter-state efforts, seasoned with some additional multistakeholder participation, may be able to take some of the rough edges off of inter-state cyber-conflict. But by relying on negotiations among states they do not alter, and may reinforce, the root causes of alignment.

Giving up: Embracing national interest over global Internet Another approach to the problem opts to give up – to accept and yield to alignment. Driven by national security concerns, this view more or less abandons a vision aimed at optimizing the value of digital communications for humanity as a whole, and shifts the policy conversation towards promoting the national interest of whatever state the analyst happens to come from. In essence, these voices see the future as a permanent, low-level cyber conflict among states, in which the great powers compete for influence and supremacy in the new cyber domain just as they have done in land, air, and sea. Alignment, in this view, is inevitable and should be embraced insofar as it serves the national interest.

In the United States, this approach is best exemplified by Council on Foreign Relations scholar Adam Segal (Segal, 2016). In his book The Hacked World Order, Segal expresses support for an open global Internet, but claims that it is only realistic to expect competing major powers to deviate from that objective. So the US, he says, must respond by adopting measures and policies that strengthen its own cyber-military capabilities. Some journalist reports indicate that this is exactly what the US has been doing for the past ten years (Harris, 2014). Segal calls for policies that push technical innovation forward, but his support for renewed research and development recalls the Eisenhower-era formation of a military–industrial complex. The US, he claims, should “harness commercial technology and turn it into military capabilities much more powerful than anyone else.” Note that the goal is not to protect or preserve the global and open character of the Internet, but to make one state – the US – “more powerful” than the others.

But how will superior US cyber-military power induce other states to make their Internet more open and free? The answer is simple: it won’t. On the contrary, it would set in motion a global

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

competition over cyber arms, surveillance, and information warfare capabilities, and it would stimulate industrial policy initiatives designed to make domestic technology industries rival those in the US. States who felt threatened by US initiatives would accelerate the territorialization and alignment processes we see now. The closing line of Segal’s Hacked World Order is revealing: “While the US will continue to strive for an open, secure, and global cyberspace, it must also prepare for the more likely future of a fractured Internet.”19

The “multistakeholder model” Others see multistakeholder models as the basis for a solution. To their credit, advocates of multistakeholder governance models recognize (partially) that the problem of governing the Internet induces innovation in political institutions, and they embrace that development. But their understanding of the nature of this institutional change is often dangerously incomplete.

Mainstream advocates of multistakeholder models tend to view it as nothing more than cooperation and dialogue among people from different “stakeholder” groups and geographic regions. Many proposals for multistakeholder governance speak vaguely about bringing people together to solve problems. They usually fail to specify who is represented, what procedures are used to make decisions, and who has the power to do what in their governance regime. Instead, they simply say “all affected [should] have a voice and method for influencing the process and providing input,”20 as if involving millions of people in collective action over contentious issues did not pose difficult problems of institutional design. Worse, they tend to view nation-states as “just another stakeholder,” an interest group equivalent to businesses, public interest advocacy groups, engineers, and developers. Vague calls for multistakeholder participation generally ignore the issue of how institutional arrangements distribute power among stakeholders. They tend to overlook the fact that states can coercively legislate outside the confines of the multistakeholder institution, and that businesses may be able to forum shop or ignore a multistakeholder venue altogether.

“How will superior US cyber-military power induce other states to make their Internet more open and free? The answer is simple: it won’t.”

These ameliorative views of multistakeholder models overlook the most critical feature of existing, functioning multistakeholder governance institutions: non-state actors are elevated to the same status as governments in the making of public policy. Indeed, they may even limit or exclude governments from certain roles. (ICANN’s bylaws, for example, do not allow government officials to be board members.) In this respect, multistakeholder institutions shift power from states to society. In their limited policy domain, they are a substitute for national governments and thus in some respects they pose a direct challenge to the state’s claim to supreme authority over public policy in communication and information. Few mainstream advocates of multistakeholder governance are willing to openly say this.

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

The battle over alignment, and over the scope and role of multistakeholder institutions, must be understood as contention for power. While states are legitimate stakeholders in Internet governance in their capacity as operators and users of both private and public networks, they also are the incumbent suppliers of governance capabilities. A state-centric approach to global governance cannot easily co-exist with a multistakeholder regime. Fundamentally they are in competition; one or the other must prevail in the domain of Internet governance.

Papers that indiscriminately tout the ability of multistakeholder institutions to solve governance problems typically draw on the successful examples of ICANN, the regional Internet address registries (RIRs) and the IETF. They imply that these organically developed Internet institutions can serve as the template for global Internet governance in other areas besides domain names, IP numbers, and standards. But attempts to use ICANN or the RIRs as a model for other forms of Internet governance must keep in mind three critical points.

First, ICANN and the RIRs can do effective governance because they have exclusive control of resources that are essential to the functioning of the Internet. As noted in Chapter 3, ICANN has a lock on the root of the domain name space, so domain name registries or registrars must come to it to obtain top level domain name assignments or certification as a registrar. In the case of the RIRs, network operators must obtain globally unique Internet protocol numbers to run their networks. In both cases, the network externality creates powerful incentives for all of the world’s network operators to converge on a common IP address and DNS governance platform so as to maintain compatibility with all other operators. In both cases, the multistakeholder institutions impose contractual terms on the recipients in exchange for the resources, and these contractual terms constitute the basis of governance.

No comparable form of centralized leverage exists for all the content, service, and application providers that are available on the Internet. Nor is it clear that we would want there to be such a centralized point of control. The whole point of the end-to-end architecture of the Internet is that it enables freedom of action and permissionless innovation for users and suppliers. This implies that there is no centralized point where all the world’s application developers, equipment manufacturers, network operators, and IT managers must go to get licenses or approvals ex ante; they can simply buy Internet access and go about their business.

A second critical point, often overlooked by those seeking to extend or build upon current MS models, is that ICANN, the IETF, and the RIRs have succeeded in avoiding alignment precisely because the resource domains they govern were largely sovereignty-free. The IETF standards that form the basis of the Internet were made without any political oversight and rely on voluntary adoption. The people who develop the standards act as individuals; their participation is not structured to represent jurisdictions or geographic regions. ICANN and the RIRs are private, nonprofit corporations with a transnational “jurisdiction” created through private contract. There is no national or international legal authority over the Internet’s name and number spaces per se. There is no sovereignty over the initial allocation or assignment of domain names and IP addresses.

The third point is that the scope of governance of the existing multistakeholder institutions is relatively narrow precisely because it is global. These global governance entities serve the

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

high-level mission of the Internet, which is to maintain its global compatibility and interoperability. In some cases they can be leveraged to address some common global problems, such as trademark-domain name conflicts or a few cybersecurity issues. But they do not – and almost certainly should not – be used to micro-govern Internet services, engage in extensive economic regulation, redistribute wealth, or respond to the divergent needs of nonglobal communities for more focused forms of governance. One of the fallacies of multistakeholder ideology is to assume that the global Internet governance institutions will become more responsive and better able to govern the Internet properly by including more and more people from a greater diversity of backgrounds. To multistakeholder evangelists, the answer to every problem is simple: more participation from more people! This is a dangerous mistake. Yes, more representation and participation is needed, but not always at the global level. Cramming more and more participants from increasingly diverse backgrounds and needs into a single, global collective governance entity will only overload the policy agenda of these institutions, and make them more politicized and less effective at handling their relatively narrow core mission.

Multistakeholder models are a feasible way out, then, but only if we view them as competitors to and substitutes for state power; only if they have some basis for contractual power; and only if their scope of governance is inversely correlated to the size of the community they engage.

The main contribution of this work has been to accurately identify the phenomenon that is underlying the “fragmentation/Balkanization” debate. The notion of alignment shows that the problem is deeply rooted in the political structure of the nation-state system, and thus there will be no easy solutions. But at least by clarifying what the problem is, the concept clears the path toward a better discussion of the way forward.

Notes 16. For recent examples of this incoherence, see the discussions of fragmentation at a May

2016 panel sponsored by Microsoft and the Washington DC Internet Society chapter, https://www.internetsociety.org/events/panel-internet-fragmenting and the plenary panel on fragmentation at the 2016 EuroDIG, http://eurodigwiki.org/wiki/PL_4:_Internet_fragmentation_and_digital_sovereignty:_implications_for_Europe

17. “. . . international policy frameworks that provide international political legitimacy for [censorship] also could be supportive sources of Internet fragmentation” (Drake, Cerf, and Kleinwachter, 2016, p. 39).

18. See the July 22, 2015 report of the UN GGE on Developments in the Field of Information and Telecommunications in the Context of International Security http://www.un.org/ga/search/view_doc.asp?symbol=A/70/174

19. I’ve focused on Segal because he is one of the more liberal analysts. He pays lip service, at least, to the open global Internet, and has a keen sense of what is being lost as we align nation-state security interests with cyberspace development. For a more militant and

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.

unapologetic version of giving in to alignment see Demchak and Dombrowski (2011).

20. This quote is taken from the Report of the so-called Global Commission on Internet Governance, “One Internet.” http://ourinternet.org/report. June 2016.

Mueller, M. (2017). Will the internet fragment? : Sovereignty, globalization and cyberspace. Polity Press. Created from usyd on 2023-05-09 06:52:17.

C op

yr ig

ht ©

2 01

7. P

ol ity

P re

ss . A

ll rig

ht s

re se

rv ed

.