Net Neutrality as Diplomacy
29 Yale L. & Pol’y Rev. Inter Alia 18
Popular imagination holds that the turf of a state’s foreign embassy is a little patch of its homeland. Enter the American Embassy in Beijing and you are in the United States. Indeed, in many contexts—such as resistance to search and seizure by a host country’s authorities—there is an inviolability to diplomatic outposts. These arrangements have been central to diplomacy for decades so that diplomats can perform their work without fear of harassment and coercion.
Complementing a state’s oasis on foreign territory is the ability to get there and back unharried. Diplomats are routinely granted immunity from detention as they travel, and la valise diplomatique—the diplomatic pouch—is a packet that cannot be seized, or in most cases even inspected, as it moves about. Each pouch is a link between a country and its outposts dispersed in alien territory around the world.
Citizens and their digital packets deserve much the same treatment as they traverse the global Internet. Just as states expect to conduct their official business on foreign soil without interference, so citizens should be able to lead digitally mediated—and increasingly distributed—lives without fear that their links to their online selves can be arbitrarily abridged or surveilled by their Internet Service Providers or any other party. Just as the sanctity of the embassy and la valise diplomatique is vital to the practice of international diplomacy, the ability of our personal bits to travel about the net unhindered is central to the lives we increasingly live online.
This frame differs from the usual criteria for debating the merits of net neutrality. It does not focus on what makes for more efficient provision of broadband services to end users. It is unaffected by what sorts of bundling of services by a local ISP might intrigue the ISP’s subscribers. It does not examine the costs and benefits of faraway content providers being asked to bargain for access to that local ISP’s customers. Instead, it recognizes that Internet users establish outposts far and wide, and that a new status quo of distributed selfhood is quickly taking hold. This status quo is enabled by neutral access among all points of presence on the Internet, and underscores the expectations that are settling as one point forges ongoing connections with others. These users have come to expect, and greatly benefit from the fact, that “far” is irrelevant—everything is in fact only a click away—and “wide” can become narrow on demand: data and activities from multiple sources can be seamlessly integrated into one’s personal page or portal.
This major shift occurred quietly and incrementally. Yesterday’s technology environment saw us safe in our own digital bunkers, occasionally sending an Internet Explorer or two out into the world: Our home PC or laptop anchored our papers, photos, and financial data, and we only connected to various faraway sites to browse news or download a batch of e-mail. As Internet connectivity has become saturating, more and more of our bits (storage) and flops (executed code) are naturally stowed far away from our personal home turf. This new reality is called the cloud and refers to anywhere other than where we are. For instance, our e-mail is now typically no longer stored under our own roofs. We use always-on, functionally interchangeable devices to get to it: a laptop, an Android phone, an iPad, or a library terminal can each be a window into our stuff, with the location of the box physically containing it irrelevant.
That is the promise of the cloud: We don’t need to have our bits here, with us, because they are always out there. And so long as we have a good signal, we need not confront the reality of the cloud that here and there are not the same place. In a world of undifferentiated Internet service—one where the network is neutral—a live connection is indeed all we need to feel that our digital stuff and activities are all safely in front of us. Apparent, not physical, unity is all that matters. But in a world without neutral service, the cloud’s promise becomes equivocal and contingent.
Broadband removes the speed bumps to communicating with a different site every time you want to paste a photo, a video, and an invoice into an e-mail, enabling the pieces of your life to spread out over a wide range of sites. And it allows a single Web page to be assembled from lots of different sites and sources, still appearing as a seamless whole—your browser fetches information from everywhere like a mad chef raiding a spice rack. Some photos are on Facebook; your email is on Gmail; favorite videos are on Vimeo; and bank account balances and portfolios are on Quicken—any of which are combinable with anything else. However, should your ISP choose to limit your access to any of these outposts, you are cut off. You are in one place, but your stuff is in another. Your software is stymied, too: A page might not be able to assemble itself because a single ingredient is inaccessible.
The cloud built over a neutral net has quietly transformed the trip your data used to make only between your hard drive and your screen into one that spans cities and countries. This provides so many more opportunities for generative transformations—especially as you might choose to share your data with others through a social network—but it comes with many more opportunities for meddling in what would otherwise be entirely your business. It only makes sense to embrace this transformation if we can rest assured that the pathways to our data will not be arbitrarily blocked or taken hostage. “Net neutrality” should mean at least this.
Many people, particularly ISPs wanting maximum flexibility to run their businesses, might argue that these concerns can be addressed without legislation. They might say that the only limits you will find in your access are ones to which you agree. If access to Facebook is important to you and an ISP provides poor (or no) connectivity to Facebook, you can fire your ISP. That is how markets work. The fact that you can walk away not only takes the teeth out of an ISP’s decision to block you from the activities you want, but likely prevents the ISP from making that decision to begin with. Why would a profit-motivated ISP block something its customers care about enough to leave over?
This is a powerful argument against legislating net neutrality. It depends, however, on a couple of assumptions for full effect. There have to be meaningful alternatives to an ISP that delivers less than what you want, and you have to know that you are getting less than what you want so you are motivated to switch. Both assumptions can turn out to be wrong. Your broadband contract may run long enough that, by the time you experience remorse in your choice, it is too late to do much about it. Switching may have other costs, such as waiting around for the cable guy to install a new modem once you have rejected the phone company’s Digital Subscriber Line (DSL), or lengthy downtime between carrier connections. For shared connections, various users’ preferences may not mesh—in which case switching from one non-neutral network to another that is non-neutral in a different way may not improve everyone’s lot. Most important, when there are only two effective choices in a market for high-speed connectivity, each may naturally end up bundling elements of a broadband deal that mix both good and bad. If there are no uniformly good third or fourth choices available, one can be metaphorically stuck with a choice between bad food and small portions.
Thus, in a strongly competitive market where providers are easily interchangeable, a rule requiring strong net neutrality could be bad: Maybe there are some bundles that consumers will like more than straight commoditized Internet, and the only way to know is to let the market experiment with various offers and see who goes for them. But less competition means less chance for the market to iron out problems, and it tends to privilege incumbent content and service providers in negotiations with ISPs for special access to their respective subscribers.
Then there is the issue of a buyer knowing whether he or she is getting a good deal. No access to Facebook is as bright a signal flare as one can imagine on today’s Internet—and thus nearly inconceivable for an ISP to attempt even without full-throated competition. But network connectivity issues can be difficult to diagnose, and a user separated from his or her data may not easily know whom, if anyone, to blame. Finding one site slower than another might be noticeable but not attributed to the ISP—instead you might just think that the site itself does not have its act together and shift that part of your digital life to a different site instead of changing ISPs. The idea that the ISP itself could be responsible for a poor site experience—because a competing site entered into a business relationship with the ISP and the one you chose did not—is not an intuitive one, and such business arrangements might rely on (and thus reinforce) this ignorance to succeed. Moreover, our digital lives will not just be found on marquee sites like Facebook—nor should they, if we are to embrace competition at the Internet’s application layer. Our activities will be multilaterally distributed across lots of different sites for lots of different purposes, where checking up on connectivity quality can become prohibitively costly.
In a market that is very far from perfect, these arguments about what works or does not work for consumers edge towards some kind of net neutrality regime, with lots of room for argument about where to draw the lines. Hence, there is a corresponding rich and diverse scholarly literature about what net neutrality regulation, if any, should look like, with some implementations far more restrictive than others. But these arguments miss a crucial frame that the diplomatic analogy suggests. The idea behind the diplomatic pouch’s bright line inviolability is that the pouch is an atomic unit of statecraft. Sovereigns are all better off if they can establish outposts in one another’s territories with certain activities beyond second-guessing or reproach.
Translated to the digital realm, the diplomatic example captures the notion that citizens have the right to communicate both with one another and, in a cloud environment, with their own remote selves, full stop. No party, public or private, should have the unchecked ability to abridge an individual’s lines of communication over our generic global Internet. If the government ran the Internet the way it maintains the highways, we would see this in the United States as a First Amendment right, but because private parties offer Internet access, we do not view it that way, at least not doctrinally.
Freedom of communication is no less important, however, simply because people get their access from private parties. Public accommodation doctrine developed in an age where physical travel was how people and ideas spread—and thus innkeepers were required to offer an available room to anyone ready to pay for it. Common carriage created a regime where private trains took all customers—and owed them the utmost duty of care to get them to their destinations unmolested. These doctrines mattered precisely in those circumstances where the market, left to its own devices, could not achieve the same result—at least not for everyone.
The rights of diplomacy arose out of states’ mutual self-interest: Each sovereign appreciated the value of having protection for its own outposts and the links to and from them, and reciprocity provided the classic framework by which to curtail one’s own activities with respect to others’ embassies in exchange for the corresponding benefits on foreign ground. On the Internet, such reciprocity could fuel parallel rights of passage for data on peer-to-peer ad hoc mesh networks—that is, networks created by users linking their devices together wirelessly without the explicit involvement of a commercial network provider. For “regular” Internet networks, maintained by commercial providers, there is no such reciprocity on which to build net neutrality. ISPs might claim that they unambiguously give something up when they are limited in how they can shape services for their own customers, even as they might appreciate a level playing field when attempting to peer with other ISPs—something they do not enjoy today. But that is not exactly true. Much Internet service is a shared resource, even within the “last mile” to the home. My cable modem’s connection bandwidth is shared with that of others. So too with wireless connectivity. A subscriber creating traffic for the purpose of slowing down others’ connectivity is engaging in a form of denial-of-service attack, and is fair game for sanctions by an ISP. Why should not a reciprocal rule apply to ISPs who purposefully slow down connectivity in a discriminatory fashion?
We enjoy access to massive archives of our digital trail in the form of emails, chats, comments, and other bits of personal ephemera, all stored conveniently out in the cloud, ready to be called up or shared in a moment, from wherever we happen to be, on whatever device we choose. The services stowing that data owe a commitment of privacy defined by a specific policy—one that we can review before we commit. Yet if any of the cloud services we use restrict our ability to extract our data, it can become stuck—and we can become locked into those services. The solution there is for such services to offer data portability policies to complement their privacy policies before we begin to patronize them, to help preserve our freedom to choose services over the long term. By dismissing the principle of net neutrality, however, we endanger that ability not just by one cloud service provider but across the board: ISPs can perform deep packet inspection to glean whatever they can about us as we correspond with different sites across the Internet, and our data can become stranded in places as the shifting sands of our ISPs’ access policies constrict access to places they disfavor. Just as international diplomacy depends on the principles of the inviolable embassy, la valise diplomatique, and mutual reciprocity to operate in the ultimate best interests of all involved, so does net neutrality depend on maintaining an online environment that preserves those aspects that made it such a valuable and central part of modern life in the first place.
A framing of Internet use that focuses on the persistent, undifferentiated right to get from here to there need not entail an absolutist position on net neutrality. Diplomatic pouches can be inspected or refused passage under exigent circumstances; innkeepers do not have to rent out rooms they do not have. The U.S. embassy in Beijing is in reality more Chinese territory than American. So, too, can viruses be rightfully detected and blocked by ISPs, and packet congestion dealt with, under the aegis of reasonable network management. But to reject net neutrality and privilege specialized ISP content-service bundles over the abilities of consumers to reach—prospectively and retroactively—the activities and data of their choice is to embrace a narrow and increasingly unrealistic view of the Internet, one that sees it as just another product with features to be optimized across clumps of customers. The Internet is that, but it is more: It is the paramount way we communicate with one another, and the means by which we establish our own digital selves. To allow anyone to deny us access to any one of those repositories would deny the ways in which the Internet has become so foundational to our very identities.
See Vienna Convention on Diplomatic Relations art. 27, Apr. 18, 1961, 3 U.S.T. 3227, 500 U.N.T.S. 95, available at http://untreaty.un.org/ilc/texts/instruments/english/conventions/9_1_196....
For more information about the rise of cloud computing, see John B. Horrigan, Use of Cloud Computing Applications and Services (2008), http://www.pewinternet.org/~/media/Files/Reports/2008/PIP_Cloud.Memo.pdf.... See also Jonathan Zittrain, The Future of the Internet: And How To Stop It 123-25 (2008) (discussing the concerns associated with diffuse computing practices).
For a free market oriented discussion of net neutrality, see Timothy B. Lee, The Durable Internet: Preserving Network Neutrality Without Regulation, Pol’y Analysis, Nov. 12, 2008, available at http://www.cato.org/pubs/pas/pa-626.pdf. See also Christopher S. Yoo, Beyond Network Neutrality, 19 Harv. J.L. & Tech. 1 (2005) (criticizing the doctrine of net neutrality).
The two options are cable and telephone company broadband. See Susan Crawford, The Looming Cable Monopoly, 29 Yale L. & Pol’y Rev. Inter Alia 34, 36 (2010).
See Zittrain, supra note 2, at 301 n.8 (surveying literature on the net neutrality debate). The topic continues to generate substantial academic debate. See, e.g., Frank Pasquale, Beyond Innovation and Competition: The Need for Qualified Transparency in Internet Intermediaries, 104 Nw. L. Rev. 105 (2010) (arguing that increased transparency is important for effective regulation); Philip J. Weiser, The Future of Internet Regulation, 43 U.C. Davis L. Rev. 529 (2009) (advocating for the creation of a self-regulatory body subject to oversight by a public agency); Kevin Werbach, Off the Hook, 95 Cornell L. Rev. 535 (2010) (arguing that the FCC could ensure net neutrality by exercising jurisdiction under the Communications Act’s provisions regarding interconnection).
For an analysis of potential First Amendment concerns that may arise in the context of municipal Wi-Fi and public networks, see Timothy Zick, Clouds, Cameras and Computers: The First Amendment and Networked Public Places, 59 Fla. L. Rev. 1 (2007). For an argument about net neutrality regulation’s potential First Amendment implications, see Tim Andrews, Does “Net Neutrality” Violate the First Amendment? (Nov. 11, 2009, 5:16 PM), http://www.atr.org/net-neutrality-violate-first-amendment-a4189#.
Heart of Atlanta Motel v. United States, 379 U.S. 241, 260 (1964).
Kline v. Santa Barbara Consol. Ry. Co., 90 P. 125, 127 (Cal. 1907).
See Zittrain, supra note 2, at 176-78 (discussing the importance of maintaining data portability).
For an argument that the increasing importance of the Internet expands the constitutional contours of one’s “home,” see Marc Jonathan Blitz, Stanley in Cyberspace: Why the Privacy Protection of the First Amendment Should Be More Like That of the Fourth, 62 Hastings L.J. (forthcoming Dec. 2010) (on file with author).