What the Law of the Sea Teaches Us About the Regulation of the Information Ocean

On Friday I gave a short talk at a conference organized to honor my colleague Bernard Oxman, who is taking up one of our very rare chairs here at UM law in this, his thirtieth year as a UM professor. (Unlike most law schools, we don't have a tradition of having chaired professorships. That may slowly be changing, fundraising willing.)

Every panelist was asked to respond to an essay Bernie wrote for the centennial volume of the American Journal of International Law. Unfortunately, Bernie's essay was about the Law of the Sea, a subject in which he is a (the?) leading expert, but about which my ignorance is vast and deep.

Thus, the title of this essay, “What the Law of the Sea Teaches Us About the Regulation of the Information Ocean.”

The audience was polite, even kind, about my remarks, so I'm posting the text (without footnotes) here. I'd sort of like to publish the footnoted version somewhere, as it tickles me to have written, however tangentially, about the law of the sea, but I have no idea where to send this.

What the Law of the Sea Teaches Us About the Regulation of the Information Ocean

By A. Michael Froomkin
University of Miami School of Law

When it comes to the Law of the Sea, I am a fish out of water: I swim in different, virtual, oceans. Thus, the question I will explore today at this event honoring Professor Bernard Oxman is one where I have the least chance of drowning: I want to consider to what extent Professor Oxman's eloquent paper charting the hazards to navigation in the future development of the Law of the Sea provides a useful intellectual cartography for those of us who sail the Information Ocean.

The Territorial Temptation: A Siren Song At Sea, 100 Am. J. Int'l L. 830 (2006), contrasts the land-based regime of post-Westphalian international law, a system that from the start was organized around principles of territoriality — or should one say, territorial-based sovereignties? — with the less imperfect law of the sea. Where on land we find many examples of multilateralism yoked to serve naked self-interest — poorly understood? — at sea, for a long period, we find instead the 'progressive triumph', a regime of mare liberum, one that at least until recently was dominated by what Prof. Oxman terms “enlightened self-interest: a rational conclusion that the interests of states in unrestricted access to the rest of the world outweighed their interests in restricting the access of others at sea.”

Prof. Oxman's article is no Jeremiad, and he no Ancient Mariner, but it is nonetheless a cautionary tale: he notes that rules motivated by a surrender to the same 'territorial temptation' that has for so long ruled on land now increasingly extend into the other 71% of the Earth, running from the sea-bed to its surface. Prof. Oxman warns us that this expansion, plus a focus on short-term goals at the expense of long-run systemic values, threatens to drain the virtues of the law of the sea and leave it looking like … land.

Thus, Prof. Oxman's concluding warning:

If experience teaches us the difficulties of overcoming states' resistance to restraints on the discretion that accompanies territorial sovereignty, and the power of emotional appeals to territorial sovereignty by those who would resist international restraints, why allow the territorial temptation to expand its reach in the sea? The need for common ground rules and cooperation by users in an area open to all is self-evident; it is an indispensable concomitant of a regime of freedom of action itself, as demonstrated by the basic principle that high seas freedoms “shall be exercised by all States with due regard for the interests of other States in their exercise of the freedom of the high seas.” Even though yielding to the territorial temptation with respect to a particular environmental problem may promise some short-term or tactical benefit, doing so may augment the difficulties of achieving a desired level of international regulation of environmental problems in that area and elsewhere.


There is a great deal in that warning and in the history leading up to it, that resonates with the current state of regulation of the Internet. Like the mare liberum, the Internet in its early days was wide-open and in practice largely unregulated; like the contemporary Law of the Sea, today's Internet faces a series of national and supra-national regulatory initiatives that seek in various ways, for reasons both fair and foul, to re-impose the control of territorial sovereignties over something that they first did not well control, and now see how to control only too well.

In what may soon be seen as a departed Golden Age, packets were free to roam from port to port; users could exchange information at will. But, just like mariners in the age of sail contemplating ocean journeys, would-be internauts faced substantial barriers to entry: only 30 years ago, internet access still required access to a mainframe somewhere and thus tended to be limited to scientists and academics, and the data moved at a mere trickle as machines mostly communicated via modems with baud speeds we sneer at today. The introduction of the (relatively) affordable personal computer, increases in processing and modem speed, and the invention of a graphical interface for unmediated one-to-many communication, what we now call the World Wide Web, all combined to bring the wonders of internet access and communication to the masses. The rest is surfing, and will some day be history.

Discussions of Internet regulation are too frequently captive of terrestrial metaphors. For example, the early adopters wrote of the “information super-highway” and of “long pipes” (but now mock the late-comers such as Senator Stevens who talk “about a series of tubes”). In this vision, packets, the basic unit of internet data transport, are analogized to cars negotiating the road system; like a traveler facing a long journey, there are in theory a large number of possible routes to one's destination. And like the roads facing that traveler, there are some routes that are both quicker and more traveled than others.

Hypnotized by the world wide web, we talk now of “cyberspace” and exchange articles as to whether the very idea of “place” is a handy simplification, an obfuscation, or a reification. For example, some have argued that the Internet is, or is usefully understood as, or might profitably be mythologized as, its own jurisdiction, with its own set of laws, rules created by the users in a truly bottom-up democratic manner. The essentialist reaction spurns these utopian visions as both fantastic and mistaken. One ought not, they say, to be able to opt out of the rules that bind our neighbors — territorial rules — because one uses a computer, any more than because one used a telephone, a car, or a spatula. And states, spurned as territorial dinosaurs by the digerati, who cast themselves as the small nimble mammals of a changed ecology, are in this counter-utopian vision, repositories and guarantors of traditional community values.

To the intellectual and especially practical challenge posed by modem-wielding libertarians and dreamers, the defenders of the terrestrial state's authority and legitimacy cast it as the guardian of our traditional values, and the defender of morality, and of the weak. So, for example, when France seeks to use its courts to prohibit (somewhat) US-based Yahoo! from selling Nazi memorabilia online in French, to French residents, the defenders of the state see this as a return to business as usual, and on the whole a salutary moment.

Whether or not triumphalism is justified, the positive story told by Joel Reidenberg, and by Jack Goldsmith & Tim Wu is surely correct: The empire has begun to strike back. States are asserting themselves to regulate the Internet as never before, both individually and in concert. The movement in the direction of business-as-ususal means that Nazi items are sold in English for dollars rather than in French for Euros — a primarily symbolic victory but one whose symbolism may actually matter. Unfortunately, it also means that in China a Google search for many terms disfavored by the authorities will result in a much less rich result than it would if conducted here, or even an error message that gives no hint of the real reasons for the search's failure. And the movement towards business as usual also means a movement towards business, as governments, working through international organizations, seek first to build an infrastructure of surveillance and enforcement to police existing intellectual property rights (and not-so-incidentally make whole new realms of data available to law enforcement) and second, to propertize and assign new categories of rights to content providers or intermediaries at the expense of consumers.

Yes, the same tide of 'territorial temptation' which covers the land now and threatens the Law of the Sea also menaces cyberspace.

The root causes of these two dangers have much in common: just as the Internet is one of the most exciting and even defining technological developments of our time, so too the sailing ship was a crowning technological achievement of its day. A tall ship is a highly complex machine that requires enormous organization and technical expertise to run properly. In broad terms the same is true of a complex network. Both depend on an extensive external infrastructure, be they boatwrights and ship's chandlers or fab labs and electrical and telephone networks. (A critical difference, however, is that the tall ship required a well-drilled team to work properly. In a good network the work tends to be more distributed and in a really good network it may be more fault-tolerant as well.)

Whatever the differences, note this critical similarity: government attempts to regulate in both spaces tracked technological advances. Before the three-mile rule it was a commonplace that the territorial limits of sovereignty were akin to the range of cannon shot — a technological constraint. As first shipping, then air, then sub-surface, transport improved, the state's ambitions grew with them. So too in virtual spaces. It has not taken long for the State — and for states in concert — to seek to (re)extend their reach to new activities enabled by new technologies. The are, in substantial part, regulating now because now is when they can.

So let us go with the flow and see how well the aquatic metaphor guides our navigation around the shoals of the territorial temptation's encroachment on the Information Ocean.

I. Similarities

The mare liberum was characterized by freedom of navigation. Freedom to go where one would did not necessarily mean freedom to trade, much less freedom to propagandize or proselytize, as territorial law took hold again once one reached port. But the freedom of the seas meant trade — and not just commerce, but trade in ideas as well.

The analogy to the Internet is appealing. Anyone familiar with the World Wide Web will see the surface similarity to a web surfer's freedom to navigate to any site, anywhere. True, an average web surfer in the US seeking what we call high-speed access is somewhat limited in that she has at best two choices, monopoly DLS and monopoly cable, but in the day there were probably only so many ships in port. And if the modern 'net surfer is likely to pass certain chokepoints on the way to foreign content, well the seas are well known for their straights and narrows also.

And that is only with yesterday's technology. Today, and even more tomorrow, Internet access will become wireless and untethered, whether via cell phones, Blackberries, or laptops connected to wireless clouds. What is more, plans are afoot to extend the reach of open access points by converting every participating wireless laptop into a repeater. By replicating the packet-switching architecture at a higher level we will reduce the effectiveness of endpoint and last-mile chokepoints. Wireless access may yet become nearly ubiquitous and too cheap to meter (and with voice-over-IP (VOIP), may take wireless telephony with it). Meanwhile enthusiasts of electronic democracy and popular empowerment offer a vision sharply at odds with that of Cassandras of globalization for whom the Internet provides yet another occasion for decision-making authority to seep away towards relatively undemocratic trans-national bodies such as ICANN, or perhaps the ITU or even, horror of horrors, the United Nations.

One need not adopt a Utopian vision in order to accept that the freedom of navigation online also means an enormous traffic in ideas both domestically and across borders. It has called forward innovations in user-to-user and one-to-many technologies ranging from blogs to YouTube, from Kazaa to MySpace. The ease of trading and selling digital content has had other consequences. It has fostered new marketing practices such as shareware and the endlessly upgraded annual version. And this flourishing of creativity has had its parallel in legal innovation, including the Copyleft or Creative Commons movement — an attempt to standardize license terms in a way that encourages sharing and attribution.

Equally innovative, albeit in a different way, are the attempts to redefine the purchase of digital property as a mere license — and to define the terms of that license in ways that favor the content-provider. EULAs attempt to undo the first sale doctrine, to prevent the purchaser from sharing the content, sometimes even from using it to criticize the manufacturer. And of course the EULAs attempt to impose a wholly different regime of rights and remedies from those consumers ordinarily expect when buying stuff. (Here, I fear, the analogy — if there is one — would be to employers' encounters with admiralty law, not to the Law of the Sea.)

Freedom of navigation online also means a substantial traffic in commerce — e-commerce as we now call it — although estimates of the actual dollar value (and percentage of consumer spending in general) vary widely. Whatever the actual amount today, everyone agrees it will be greater next year. Some even predict the growth of a new economy in which intermediaries are replaced by huge open networks of buyers and sellers trading with e-cash on anonymous electronic exchanges, engaging in regulatory arbitrage — and evading their taxes.

Along with all these developments come new evils, or old evils in new packages, notably the trade in illicit content. In the United States that means pornography, gambling, and malware — the modern equivalents perhaps of the bacteria and viruses spread by sea-borne travelers, sometimes with such disastrous results for indigenous populations. We've seen nothing quite that bad online yet, although the huge volumes of spam delivered by thousands of zombie computers harnessed to form botnets now threaten the ecology of email. Perhaps the best oceanic analogy for the modern spam, however, is cross-border pollution, and as we shall see, the analogy extends to the methods being advocated to control it as well.

And, of course, there's “Piracy”. On the high seas, piracy was the classic offense against the laws of nations. On the Internet, the term “piracy” has been, well, pirated, to mean making unlicensed copies of copyrighted works. And the content industries have made an enormous effort on multiple fronts to preserve their business models. Where once the Navy patrolled the seas, now the copyright police patrol third-world bazaars, online markets, and file sharing networks. They file lawsuits against file-sharing software creators, teenaged codebreakers who find ways to decrypt DVDs, middlemen providing matching services for file sharers, and even thousands of end-users. The content industries have also leaned heavily on the designers of hardware and the definers of standards to build “trusted computers” and “digital rights management” systems designed to limit what users can do with their computers, to put roadblocks on the unsanctioned exchange of information.

I would be remiss if I did not note some places where my extended analogy founders.

  • The Internet standards process requires even more standardization than sea did/does. Ships could have different designs, so long as they floated. Internet enabling hardware takes many forms, but if your devices don't adhere to some basic standards, they are not going to be able to communicate with much.

  • Packets are not like fish. They are artificial, and in the normal case do not reproduce. More to the point, if you are on the outside of the Great Firewall of China — and, for now, even on the inside if you are skilled — you can route around blockages much more successfully than schools of fish and pods of dolphins can evade driftnets. Unlike driftnets, the effects of censorship need not be felt abroad, except as a loss of readership.

  • And packets are not like ships. The freedom of navigation was used to go places. With the Internet, much as we like to talk about “going” to a web site, the inescapable fact is that you don't actually move. That said, your data can move, and indeed can in some cases reside entirely offshore, making datahavens something comparable to the modern flag of convenience.

So, to recap, there are many similarities between the sea's effect on trade and commerce, what Chief Justice Marshall's era called intercourse, and the modern Information Ocean's effects on commerce and the exchange of ideas. And in both cases a relaxed, standardized, but not anarchic, international regime enabled (or at least enhanced) these benefits. And, in both cases, some of the benefits are threatened by the possible imposition of a more nationalistic and more controlling legal regime, a regime that shows every sign of giving into not just the territorial temptation but temptation in general.

II. Internet Regulation Dangers

The move to regulate the Internet threatens not only to subject it to the same rules that apply to other, meatspace, activities, but in some cases to subject it to more stringent rules; these in turn are often designed not only to regulate the Internet but to control various offline practices ranging from pedophillia and money laundering to parody and dissidence.

As with the law of the sea, the regulatory impulse proceeds on multiple fronts. There is of course national regulation, rapidly extending its reach to e-commerce, to online libel, and especially to digital property. The digital property issues appear in multiple forms and with every type of intellectual property. This is not the place to list them all; they include the regulation of online content — in trademark everyitng from passing off, to old fashioned infringement, and on to tarnishment. Only a few of these — new and exotic forms of conduct prohibited as 'initial interest confusion' and the regulation of domain names — have been particularly controversial. The story regarding copyright is more complex still, with the primary controversies relating to data exchanged online as opposed to online display, although even here the overenthusiastic use of the DMCA's takedown provisions increasingly prove troublesome, as does an epidemic of craven self-policing in excess of anything the law requires.

The urge to regulate has many sources. In the US, it is in part a reaction to the use of the Internet as a tool of regulatory arbitrage in which Americans evade local rules banning pornography, gambling, or the importation of cheap Canadian medicines. Another part of the urge to regulate, not limited to the US, reacts to the danger that computer-mediated communications, whether anonymous, encrypted, or hidden in plain sight, may enable terrorism or otherwise undermine national security. And, of course, many nations the urge to control and regulate is part of a larger program of control and even repression. The recent contretemps about Google acceding to China's demands that 'sensitive' searches be sanitized is only the most publicized of a series of attempts in China, Singapore, Saudi Arabia and many other countries to clamp down on the subversive effect of such a great freedom to communicate.

For those of us with the good fortune to live under less repressive governments, the greatest dangers, however, are not in national legislation, but at the international level. First, the US has entered into a series of bilateral agreements, mostly with smaller economies, in which the US has begun to impose IP regulatory regimes on its trading partners that are more stringent than those it imposes on itself. Even when treaties do not change existing rules, they lock them in and thus place weighty obstacles in the way of democratic re-examination of their provisions. Some have even suggested that the current US administration is embarking on a strategy to lock in our current practices in this manner in order to prevent their repeal by its successors.

Multilateral processes are also being harnessed to this goal. Negotiations on The Hague Convention on Jurisdiction, for example, almost foundered on its applicability to e-commerce, to other online activities, and on the protection that would be given to various idiosyncratic sui generis IP rights.

In addition to the traditional multilateral processes, Internet regulation enjoys, if that's the word, a hybrid semi-private regulator called ICANN, which controls the domain name system. ICANN's technical control makes its decisions almost instantly enforceable. So far that power has been relatively sparingly, primarily to create new rights for incumbent trademark holders in the domain name system, and to prevent the creation of interesting new top-level domains that might become a source of heartburn for incumbent service providers or for large trademark interests who fear an increase in both infringement and especially monitoring costs. This is not the place for the complex details, but it is relevant to remark that the latest proposals on domain name policy to emanate from ICANN give each participating national governments unlimited rights to veto new TLD names whose semantic content (e.g. “Tibet”) they find troubling for any reason.

Currently, content providers are mounting a strong effort to add to what critics call 'the copyright grab' or the 'digital enclosure' movement at the ongoing negotiations over the WIPO broadcast treaty. Due to the heavy-handed efforts by the chair to define the agenda, the conference has been focused on creating new rights for broadcasters, who it is proposed would be given the right to prevent the retransmission of otherwise public domain content that they transmitted. The effect on the Internet — which is where most of that retransmission happens, often after important transformation and the creation of exciting if oft satirical derivative works — would be significant.

Most serious of all are the multi-lateral efforts to require internet intermediaries to record most or all of what their customers do online and to keep the records available for months or years in case police or intelligence agencies should later wish to inspect them. Governments have come to understand that technologies of online surveillance and monitoring require wide dispersion to be effective or their targets will move their operations offshore.

The initial move towards a surveillance regime, widely believed to have been orchestrated by the FBI (and/or the NSA) was through the Council of Europe's Cybercrime Convention. Initial proposals to create a duty to monitor at the request of a foreign power were heavily watered down after an outcry by civil liberties organizations in the US and Europe. The final draft created mechanisms for increased cooperation between law enforcement agencies but did not require one government to, for example, cooperate in the monitoring of foreign dissidents visiting or sheltering on its territory.

That victory was short-lived. The European Union recently adopted a far more comprehensive rule, one justified as necessary to help fight terrorism and organized crime, that requires ISPs to keep records of what their clients do online. This directive will take effect this August, although member states may postpone compliance until March 2009. Under the terms of the so-called Data Retention Directive, Internet Service Providers must retain for up to two years the date, destination, and duration of their customers' communications — and made these available to law enforcement authorities. US Attorney General Alberto Gonzales recently proposed that the US adopt a similar rule, although the bill seems to have little chance of passage in the House — this year.

It can be seen from this gale-speed summary that like the Law of the Sea, there are proposals afoot for the regulation of the Internet that threaten not merely to tame it, but to change the overall character of both the regulatory environment and the thing itself. That said, there does appear to be one important difference from the Oxman conception of the recent evolution of the law of the sea. By and large the private actors, and especially the NGOs, are the heros of the internet story; one gathers that Prof. Oxman perhaps sees them playing a more ambiguous role at sea. In the Internet context, the truly dangerous parties clearly are the governments, which are reclaiming the intellectual agenda as a prelude to colluding, in the name of security, against the privacy and in some cases freedom of their citizens.


Thus, at last, we are in position to reconsider Professor Oxman's envoi:

The need for common ground rules and cooperation by users in an area open to all is self-evident;

True online, at least as much as at sea.

it is an indispensable concomitant of a regime of freedom of action itself,

Again, true.

as demonstrated by the basic principle that high seas freedoms “shall be exercised by all States with due regard for the interests of other States in their exercise of the freedom of the high seas.”

Here, I confess, the analogy breaks down a bit: the online spillover effects of some online activity is much greater than most sea-based activity. If the US has a liberal information regime, it becomes a means by which citizens elsewhere can route around local content control. The same is true in reverse: restrictive US policies on pornography, gambling, or the sharing of digital intellectual property are undermined by foreign rules or the failure of foreign domestic enforcement. The cures require the imposition of a regime of control and enumeration that would, I fear, be worse than any of the diseases so far identified.

Even though yielding to the territorial temptation with respect to a particular environmental problem may promise some short-term or tactical benefit, doing so may augment the difficulties of achieving a desired level of international regulation of environmental problems in that area and elsewhere.

In contrast, this last point is even truer online.

It seems, therefore, that the Law of the Sea has something to teach those of us who immerse themselves in cyberspace — but there is a real danger that the lessons will not be learned in either arena, and we will lose sight of “enlightened self-interest: a rational conclusion that the interests of states in unrestricted access to the rest of the world outweighed their interests in restricting the access of others” … whether that access is to the sea or to the Internet.

This entry was posted in Talks & Conferences. Bookmark the permalink.

One Response to What the Law of the Sea Teaches Us About the Regulation of the Information Ocean

  1. Kaleberg says:

    This is not the first time I’ve seen this analogy. The Story of Ping, a Captain Kangaroo favorite children’s book set on the Yangse River, was apparently not about ducks and boats on the river, but about internet packet and host management. Check out:


    Look for customer comments, particularly the one from Upper Volta (ostensibly).

Comments are closed.