Code 2.0

The Sovereign of the Space: Rules

When you enter the world of MMOG Second Life as a new character, the rules of Second Life are explained to you. Some of these rules are the techniques you will need to get around in Second Life — how to move, or how to fly. Some of the rules are normative commands that tell you what you can and can’t do.

It is impossible when confronting this introduction not to notice that these constraints are constructed. God didn’t make Second Life. No one is confused about whether he or she did. Nor is it likely that one entering this space wouldn’t notice that one important dimension to that construction is construction through code. That you can fly is a choice of the coders. Where you can fly is a choice of the coders. That when you bump into someone, a warning box is displaced is a choice of the coders. That you can turn off IM conversations from people you don’t want to hear from is a choice of the coders. No one mistakes that there are choices made here. Everyone recognizes that a critical part of the cyberspace world is made through code. As Second Life’s CEO, Philip Rosedale, put it to me: “What is God in a virtual world? Your only God is the code[1]”.

Now, as I’ve said from the start, we should distinguish between richly controlling spaces and thinly controlling spaces. Spaces like Second Life richly control the life of people playing there. Indeed, the whole objective of playing there is create the impression that one is there. These, again, are the sorts of places I call cyberspace.

Cyberspace is very different from life on a bill-paying website, or on a site holding your e-mail. Code controls these, too. But the control, or sovereignty, of those sites is distinct from the control of Second Life. In Second Life, or in what I’ve defined to be cyberspace generally, the control is ubiquitous; on a bill-paying website, or on what I’ve called the Internet, the control is passing, transitory.

Interestingly, there is an important dynamic shift that we’ve already identified, more in thinly controlling spaces than thick. This is the preference for code controls where code controls are possible.

Think again about the bill-paying website. It is of course against the law to access someone’s bank account and transfer funds from that account without the authorization of the account owner. But no bank would ever simply rely upon the law to enforce that rule. Every bank adds a complex set of code to authenticate who you are when you enter a bill-paying website. Where a policy objective can be coded, then the only limit on that coding is the marginal cost of code versus the marginal benefit of the added control.

But in a thickly controlling environment such as Second Life, there’s a limit to the use of code to guide social behavior. Sometimes, in other words, better code can weaken community. As Second Life’s Rosedale put it,

A second way in which better code can weaken community is even more important. As Second Life is, it doesn’t enable people easily to segregate. As Rosedale described,

The code thus doesn’t simply make all problems go away. It doesn’t remove the need for neighbors to work stuff out. And in this way, the code helps build community. The practice of interaction builds bonds that would not be built if the code produced the same results, automatically. Optimal design leaves certain problems to the players to work out — not because the solution couldn’t be coded, but also because coding a solution would have collateral costs.

Nonetheless, it is still the sovereign in these virtual spaces that chooses one modality over another. The trade-off is complicated. Perfect efficiency of results is not always perfectly efficient. But still the choice of means remains.

The Sovereign of the Space: Choosing Rules

But how is that choice made? Or more directly, what about democracy? In real space, the rule is that sovereigns are legitimate only if democratic. We barely tolerate (most) nondemocratic regimes. The general norm for real space life is that ultimately, the people rule.

But the single most interesting nondevelopment in cyberspace is that, again, as Castronova puts it, “one does not find much democracy at all in synthetic worlds”[4]. The one real exception is a world called “A Tale in the Desert”[5]. Democracy has not broken out across cyberspace, or on the Internet. Instead, democracy is a rare exception to a fairly strong rule — that the “owner” of the space is the sovereign. And in Castronova’s view, the owner is not ordinarily a very good sovereign:

This isn’t to say that aggregated views don’t matter in cyberspace. Indeed, they are crucial to central aspects of the Internet as it is just now. A kind of voting — as manifested through links — guides search engines. Technorati, as I’ve already described, relies upon the same to rank blogs. And important sites, such as Slashdot, routinely use rankings or votes of editors to determine which comments will rise to the top.

These are all democracy-like. But they are not democracy. Democracy is the practice of the people choosing the rules that will govern a particular place. And with the exception of Wikipedia, and “A Tale in the Desert”, there are very few major Internet or cyberspace institutions that run by the rule of the people.

So what explains this democracy gap? And should we expect it to change?

Our history of self-government has a particular form, with two importantly contingent features. Before our founding, life was geographically based — a nation was a society located in a physical space, with a single sovereign allegiance. As we’ll consider more extensively in the chapter that follows, the conceptual revolution of the American Republic was that citizens could have two sovereigns — more precisely, that they (as the ultimate sovereign) could vest their sovereign power in two different delegates. Their state government was one delegate, the federal government was another; individuals living in a single geographic location could thus be citizens of both governments. That was the idea of the founding document, and the Fourteenth Amendment made it explicit: “All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside.”

Citizenship in this sense did not always mean a right to contribute to the self-government of whatever community you were a citizen of[7]. Even today there are citizens that have no right to vote — e.g., children. But for those recognized as members of civil and political society, citizenship is an entitlement: It is a right to participate in the governing of the political community of which they are members. As a citizen of the United States, I have the right to vote in U.S. elections; as a citizen of California, I have the right to vote in California elections. I have both rights at the same time.

At this level, the link between entitlement and geography makes sense. But as mobility has increased, the at-one-time obvious link between geography and citizenship has become less and less obvious. I live in San Francisco, but I work in Palo Alto. The rules give me full participation rights in San Francisco but none in Palo Alto. Why does this make sense?

Political theorists have noted this problem for some time[8]. Scholars such as Richard Ford and Lani Guinier have developed powerful alternative conceptions of self-government that would enable a kind of self-government not tied directly to geography. With one such alternative, voters choose (within limits) the community where their votes count. Thus if I felt participating in the future of Palo Alto was more important than participating in the future of San Francisco, I would have the right to vote in Palo Alto though I lived in San Francisco.

These complications are magnified when we consider the link between geography and cyberspace. Even if I should have the right to vote in the community where I work, should I have the right to vote in the community where I play? Why would real-space citizens need to have any control over cyber-places or their architectures? You might spend most of your life in a mall, but no one would say you have a right to control the mall’s architecture. Or you might like to visit Disney World every weekend, but it would be odd to claim that you therefore have a right to regulate Disney World. Why isn’t cyberspace more like a mall or a theme park than like the district in which you live and vote?

Your relationship to a mall, or to Disney World, is the relationship of consumer to merchant. If you don’t like two-all-beef-patties-special-sauce-lettuce-cheese-pickles-onions-on-a-sesa-me-s eed-bun, then you can go to Burger King; McDonald’s has no duty to let you vote on how it makes its hamburgers. If you don’t like the local mall, you can go to another. The power you have over these institutions is your ability to exit. They compete for your attention, your custom, and your loyalty; if they compete well, you will give them your custom; if they don’t, you will go somewhere else. That competition is crucial in disciplining these institutions. What makes them work well is this competition among these potential sources for your custom.

This merchant-sovereign part of our life is important. It is where we spend most of our time, and most people are more satisfied with this part of their lives than they are with the part within which they get to vote. In this sense, all these places are sovereigns; they all impose rules on us. But our recourse with respect to merchant-sovereigns is simply to take our business elsewhere.

But the merchant-sovereign part of our life is not exclusive. There are also citizen-sovereign parts of our life. There are no states that get to say to their citizens: “You have no right to vote here; if you don’t like it, leave.” Our role in relation to our governments is that of a stakeholder with a voice. We have a right — if the government is to be called democratic — to participate in its structuring.

And this is true not just with governments. It would be an odd university that gave its faculty no right to vote on issues central to the university (though it is an odd corporation that gives its employees a right to vote on issues related to employment). It would be an odd social club that did not give members some control over its functions — though again, there are such clubs, just as there are nondemocratic governments. Even the church allows its members to determine a great deal of how members are governed. In these institutions, we are members, not consumers — or, not just consumers. These institutions give consumers control over the rules that will govern them. In this sense, these institutions are citizen-sovereignties.

As a descriptive matter, then, cyberspace is not yet dominated (or even broadly populated) by citizen-sovereignties. The sovereignties we see so far are all merchant-sovereignties. And this is even more clearly true with the Internet. To the extent sites are sovereign, they are merchant-sovereigns. Our relationship to them is the same as our relationship to McDonald’s.

Some theorists have tried to collapse these two different models into one. Some have tried to carry the member model into every sphere of social life — the workplace, the mall, the local pub.[9] Others have tried to carry the consumer model into every sphere of social life — followers of Charles Tiebout, for example, have tried to explain competition among governments along the lines of the choices we make among toothpastes.[10] But even if we cannot articulate perfectly the justifications for treating these choices differently, it would be a mistake to collapse these different spheres into one. It would be hell to have to vote on the design of toothpaste, and tyranny if our only recourse against a government we didn’t like was to move to a different land.

But then is it a problem that cyberspace is comprised of just merchant-sovereignties? The first defense for merchant-sovereignties is developed in the writings of David Post and his sometime coauthor David Johnson.[11] Post’s article “Anarchy, State, and the Internet” best sets the stage. Communities in cyberspace, Post argues, are governed by “rule-sets.” We can understand these rule-sets to be the requirements, whether embedded in the architecture or promulgated in a set of rules, that constrain behavior in a particular place. The world of cyberspace, he argues, is comprised by these rule-sets. Individuals will choose to enter one rule-set or another. As rule-sets compete for our attention, the world of cyberspace will come to be defined by this competition of merchant-sovereigns for customers.

Post’s account again is descriptively accurate. It is also, Post argues, normatively recommended. Sovereigns should be understand as a firm’s market power is understood in antitrust law. By “market power” antitrust lawyers and economists mean a firm’s ability to raise prices profitably. In a perfectly competitive market, a firm with no market power is the one that cannot raise its prices because it would lose so much in sales as to make the increase not worth it.[12] The firm that does have market power can raise prices and see its profits increase. The firm with market power also has the ability to force consumers to accept a price for a good that is higher than the price in a competitive market.

We might imagine an analogous constraint operating on government. Sovereigns, like firms, can get away with only so much. As they become more repressive, or as they regulate more harshly, other sovereigns, or other rule-sets, become competitors. At some point it is easier for citizens to leave than to put up with the burdens of regulation[13], or easier to evade the law than to comply with it.

Because such moves are costly in real space, sovereigns, at least in the short run, can get away with a lot. But in cyberspace, moving is not so hard. If you do not like the rule-set of your MMOGs, you can change games. If you do not like the amount of advertising on one Internet portal, then in two seconds you can change your default portal. Life in cyberspace is about joining without ever leaving your home. If the group you join does not treat you as you want to be treated, you can leave. Because competitive pressure is greater in cyberspace, governments and other propagators of rule-sets must behave like firms in a competitive market.

This is an important and interesting conception of governance. Important because it describes governance in cyberspace; interesting because it perhaps shows the purpose and limits of citizen-sovereignty in real space. It argues for a world of volunteers, one where rules are not imposed but selected. It is a world that minimizes the unconcented-to-power of any particular government, by making governments competitors for citizens. It is government like McDonald’s or Coca-Cola — eager to please, fearful of revolt.

There are reasons, however, to be skeptical about this view. First, consider the claim that exit costs are lower in cyberspace than in real space. When you switch to a different ISP or Internet portal, you no doubt confront a different set of “rules”, and these rules no doubt compete for your attention. This is just like going from one restaurant or shopping mall to another. There are competing rule-sets; they are among several factors you consider in choosing an ISP; and to the extent that there is easy movement among these rule-sets, this movement is undoubtedly a competition among them. Some ISPs, of course, try to make this movement difficult. If you’ve been a member of AOL for ten years, and you decide you want to switch, AOL doesn’t make that change easy by providing, for example, a simple ability to forward your e-mail. But as people recognize this restriction imposed by AOL, they’ll choose other ISPs. If the competition is real, the rule-set will compete.

Communities, however, are different. Consider the “competition” among, say, MMOGs. You join an MMOG and spend months building a character in that community. You also collect assets — buildings you’ve built, or weapons you’ve acquired. Both resources are a kind of capital. The set of relationships you’ve developed are the social capital; all the stuff you have is the physical capital.

If you then become dissatisfied with life in your chosen MMOGs, you can leave. But leaving is costly. You can’t transfer the social capital you’ve built, and, depending upon the game, you may not be able to transfer the physical capital either. Like choosing to join a different frequent flyer program, the choice to join a different MMOG is a decision to waste certain assets. And that fact will weaken the competition among these rule-sets.

I don’t mean to overstate the point. Indeed, as markets have developed for selling assets within MMOGs, and the nature of the games has become standardized, some argue that it is becoming much easier to move from one game to another. In real space you also can’t easily transfer social capital from one community to another. Friends are not fungible, even if they can give you connections at your new home. But physical assets in real space are transferable. I can sell what I don’t want and move what I do. Always. In MMOGs, not always.

Paradoxically, then, we might say that it may be harder to change communities in cyberspace than it is in real space. It is harder because you must give up everything in a move from one cyber-community to another, whereas in real space you can bring much of it with you.[14] Communities in cyberspace may in the short run have more power over their citizens (regarding social capital) than real-space communities do.

This means that the picture of competing rule-sets in cyberspace is more complex than Post suggests. The pressure on competition is potentially greater in turn. That might motivate a desire in cyberspace communities to shift toward citizen-sovereignty, but, again, there’s not much evidence of that shift yet.

There is a second, more fundamental skepticism. Even if we could construct cyberspace on the model of the market — so that we relate to spaces in cyberspace the way we relate to toothpaste in real space — there are strong reasons not to. As life moves online, and more and more citizens from states X, Y, and Z come to interact in cyberspaces A, B, and C, these cyberspaces may well need to develop the kind of responsibility and attention that develops (ideally) within a democracy. Or, put differently, if cyberspace wants to be considered its own legitimate sovereign, and thus deserving of some measure of independence and respect, it must become more clearly a citizen-sovereignty.

This same dynamic happens in real space. There are many institutions that are not “sovereign” in the sense that they control how people live, but are “sovereign” in the sense that within the institution, they control how people behave. Universities, social clubs, churches, and corporations are the obvious examples of institutions that gain a kind of autonomy from ordinary government. This autonomy can be thick or thin. And my suggestion is that it gets thicker the more the institution reflects values of citizen-sovereignty.

This kind of sovereignty is expressed in the law through doctrines of immunity. A corporation has certain immunities, but that depends upon it fitting a particular corporate form. Churches have a certain immunity, but it is increasingly challenged as its governance becomes more alien.

Communities in cyberspace will earn a similar immunity more quickly if they reflect citizen-sovereign values rather than merchant-sovereign values. The more responsible the communities become, the more likely real-space governments will defer to their norms through doctrines like immunity.

This maturation — if it is that — is obviously a long way down the road. It depends upon an increasing self-recognition by members of these cyberspace communities that they are, in a sense, separate, or complementary communities. It depends upon an increasing recognition among noncommunity members that there’s something distinctive about these communities. Some are optimistic that this will happen. As Dan Hunter and Greg Lastowka write:

We’ve seen something similar to this progression in our own history. There was a time when the United States was really “these united States”, a time when the dominant political reality was local and there were real differences of culture and values between New York and Virginia. Despite these differences, in 1789 these states united to establish a relatively thin national government. This government was to be minimal and limited; it had a number of narrow, strictly articulated purposes, beyond which it was not to go.

These limits made sense in the limited community that the United States was. At the time there was very little that the states shared as a nation. They shared a history of defeating the strongest army in the world and a purpose of growing across an almost endless continent[16], but they did not share a social or political life. Life was local, exchange was relatively rare, and in such a world limited national government made sense.

Nevertheless, there were national questions to be articulated and resolved. Slavery, for example, was a mark on our country as a whole, even though the practice was limited to a few states. There had been arguments at the founding about whether slavery should be left to local regulation. But the Constitution was founded on a compromise about that question. Congress was not permitted to address the question of the “importation” of slaves until 1808.[17] After that, it could, and people, increasingly, said that it should. Slavery continued, however, to be a stain on the moral standing of our nation. Congress could eliminate it in the territories at least, and some argued that it should do so in the southern states as well.

Opponents to this call for Congress to cleanse our nation of slavery were of two sorts. One type supported the institution of slavery and believed it was central to southern life. They are not my focus here. My focus is a second type — those who, with perfect integrity and candor, argued that slavery was a local issue, not a national issue; that the framers had understood it not to be a national issue; and that the national government should let it alone.

However true that claim might have been in 1791 or 1828, it became less plausible over time. As the nation became socially and economically more integrated, the plausibility of saying “I am a Virginian first” declined, and the significance of being a citizen of the nation as a whole increased.[18]

This change came about not through some political decision but as a result of a changing economic and social reality. Our sense of being members of a national community increased until, at a certain stage, it became impossible to deny our national citizenship. A war produced that recognition. The Fourteenth Amendment wrote it into the Constitution; economic and social intercourse made it completely real. And as this change took hold, the claim that issues like slavery were local became absurd.

The very same process is happening to us now, internationally, and cyberspace is making an important contribution. It has been slowly gaining momentum, of course, since the end of World War II, but the Internet has wildly accelerated the pace. Ordinary citizens are connected internationally and can make international transactions as never before. The presence of a community that is beyond any individual state is increasingly undeniable.

As this international community develops in cyberspace, its citizens will find it increasingly difficult to stand neutral in this international space. Just as a principled sort of citizen in 1791 might have said that slavery in Virginia was irrelevant to a citizen in Maine, so in 1991 the control of speech in Singapore may have been irrelevant to a citizen of the United States. But just as the claim about slavery’s local relevance became implausible in the course of the nineteenth century, the claim about speech on the Net will become equally implausible in the 21 st century. Cyberspace is an international community; there are constitutional questions for it to answer; and we cannot simply stand back from this international space and say that these questions are local issues.

At least, we could not say that once we effectively invaded this international space with the Internet of 1995. We put into the world an architecture that facilitated extraordinarily free speech and extraordinary privacy; that enabled secure communications through a protocol that permitted encryption; and that encouraged free communications through a protocol that resisted censorship. That was the speech architecture that the Net gave the world — that we gave the world.

Now we are changing that architecture. We are enabling commerce in a way we did not before; we are contemplating the regulation of encryption; we are facilitating identity and content control. We are remaking the values of the Net, and the question is: Can we commit ourselves to neutrality in this reconstruction of the architecture of the Net?

I don’t think that we can. Or should. Or will. We can no more stand neutral on the question of whether the Net should enable centralized control of speech than Americans could stand neutral on the question of slavery in 1861. We should understand that we are part of a worldwide political battle; that we have views about what rights should be guaranteed to all humans, regardless of their nationality; and that we should be ready to press these views in this new political space opened up by the Net.

I am not arguing for world government. Indeed, the impossibility of such an idea is the focus of much of the next chapter. My argument instead is that we must take responsibility for the politics we are building into this architecture, for this architecture is a sovereign governing the community that lives in that space. We must consider the politics of the architectures of the life there.

I have argued that we should understand the code in cyberspace to be its own sort of regulatory regime, and that this code can sometimes be in competition with the law’s regulatory regime. For example, we saw how copyright law could be inconsistent with the regulatory regime of trusted systems. My argument is that we should understand these to be two regulatory regimes in competition with each other. We need a way to choose between them. We need a way to decide which should prevail.

As this system of regulation by code develops, it will contain its own norms, which it will express in its structures or in the rules it imposes. If the predictions of law and economics are correct, these norms will no doubt be efficient, and they may well be just. But to the extent that justice does not track efficiency, they will be efficient and unjust. The question will then be: How do we react to this gap?

There is an important pattern in this competition between code and law. Law, at least as it regulates international relations, is the product of extended negotiations. Countries must come to an agreement about how law will regulate and about any norms that they will impose on private ordering. As their work relates to cyberspace in particular, this agreement is quite significant. It will require the nations of the world to come to a common understanding about this space and to develop a common strategy for dealing with its regulation.

Conflicts

Here are two stories about the power of sovereignty, one you’re likely to have heard of, and the other not.

On Being “in” Cyberspace

Cyberspace is a place.[13] People live there. They experience all the sorts of things that they experience in real space there, and some experience more. They experience this, not as isolated individuals playing some high-tech computer game, but as part of groups, in communities, among strangers, and among people they come to know and sometimes like — or love.

While they are in that place, cyberspace, they are also here. They are at a terminal screen, eating chips, ignoring the phone. They are downstairs on the computer, late at night, while their husbands are asleep. They are at work, at cyber cafes, and in computer labs. They live this life there, while here, and then at some point in the day they jack out and are only here. They rise from the machine in a bit of a daze, and turn around. They have returned.

So where are they when they are in cyberspace?

We have this desire to pick: We want to say that they are either in cyberspace or in real space. We have this desire because we want to know which space is responsible. Which space has jurisdiction over them? Which space rules?

The answer is both. Whenever anyone is in cyberspace, she is also here, in real space. Whenever one is subject to the norms of a cyberspace community, one is also living within a community in real space. You are always in both places if you are there, and the norms of both places apply. The problem for law is to work out how the norms of the two communities are to apply given that the subject to whom they apply may be in both places at once.

Think again about Jake Baker. The problem with Jake was not that he went to a different place where the norms were different. The problem was that he was simultaneously in a Michigan dorm room and on the Net. He was subject to the norm of civility in the dorm, and he was subject to the norm of indecency in cyberspace. He was subject, that is, to two sets of norms as he sat in that single chair.

So whose norms would apply? How would real-space governments deal with the conflict between these two communities?

Some examples might help to set a context in which that question might be answered. Ordinarily, when you go to Europe you do not bring the federal government with you. You do not carry along a set of rules for Americans while in Europe. In Germany you are generally subject to German law. The United States ordinarily has very little reason to worry about regulating your behavior there.

But sometimes the U.S. government does have a reason to regulate American citizens abroad. When it does, nothing in international law can stop it.[14] For example, there are jurisdictions where pedophilia is not adequately regulated, and for a time they became target tourist spots for pedophiles from around the world. The U.S. government passed a law in 1994 to forbid Americans from engaging in child sex while outside the United States, even in jurisdictions where child sex is permitted.[15]

What justification could there have been for such a law? Obviously, the sense of Congress was that if a person engages in such behavior in a foreign country, they are more likely to do it here as well. If they visit a community where the norms permit such behavior, they are more likely to carry those norms back to their life here. Thus, while the American government generally doesn’t much care what you do elsewhere, it does begin to care when what you do elsewhere has an effect on your life here.

Regulations like this are the exception, of course. But they are the exception because the practice of passing into alternative, or alien, communities in real space is also the exception. The frictions of real-space life make it less likely that the norms of an alien culture will bleed into our own; the distance between us and alien cultures is so great that very few can afford to have a life in both places.

But the Net changes this. As the Baker case suggests, and as any number of other cases will press, with cyberspace these other communities are no longer elsewhere. They can be brought home, or more frighteningly, into the home. Real-space communities no longer have the buffer of friction to protect them. Another community can now capture the attention of their citizens without their citizens’ ever leaving their living room. People may be in both places at the same time. One affects the other. As Edward Castronova writes, “synthetic worlds are becoming important because events inside them can have effects outside them.”[16] The question for government is how far to allow these effects to go.

Now this question has really three different parts — two old, and one new. The old part is how a far a government will allow foreign influences to affect its culture and its people. Cultures at one time isolated are later invaded when the barriers to invasion fall. Think about the plea from Europeans to stop the invasion of American culture, which pours over satellite television into the living rooms of European citizens.[17] Or even more extreme, the Middle East. These places have long fought to protect their culture from certain alien influences, and that fight becomes much more difficult once the Internet becomes ubiquitous.

The second old part is the question of how, or whether, a government will protect its citizens against foreign practices or rules that are inconsistent with its own. For example, the copyright law of France strongly protects the “moral rights” of French authors. If a French author enters into a contract with an American publisher, and that contract does not adequately protect the “moral rights” of the French citizen, how will the French respond?

But the third question — and the new part — is the issue raised by the ability for citizens to live in the alien culture while still at home. This is something more than merely watching foreign television. The alternatives offered by TV are alternatives of the imagination. The interactive life of cyberspace offers alternative ways of living (or at least some cyberspaces do).

My focus in this chapter is not on the first question, which many call cultural imperialism. It is instead upon the conflicts that will be manifested by the second and third. It may well be true that there have always been conflicts between the rules of different governments. It may always have been that those conflicts have bled into particular local disputes. Cyberspace has exploded this third stage of the debate. What was once the exception will become the rule. Behavior was once governed ordinarily within one jurisdiction, or within two coordinating jurisdictions. Now it will systematically be governed within multiple, noncoordinating jurisdictions. How can law handle this?

The integration of cyberspace will produce a profound increase in the incidence of these conflicts. It will produce a kind of conflict that has never happened before: a conflict arising from individuals from different jurisdictions living together in one space while living in these different jurisdictions.

This question has produced a ferocious argument between two extremes. At one end is the work of David Post and David Johnson. Johnson and Post argue that the multiplicity of jurisdictions in which your behavior is subject to regulation (since anything you do in cyberspace has an effect in every other context) should mean that much behavior is presumptively not subject to regulation anywhere. Anywhere, that is, save cyberspace.[18] The inconsistency of any other solution, they argue, would be absurd. Rather than embracing the absurd, we should embrace something far more sensible: life in cyberspace, as Milan Kundera might put it, is life elsewhere.

At the other extreme is the work of scholars such as Jack Goldsmith and Tim Wu, who claim there is nothing new here — at least new from the perspective private international law.[19] For many years the law has worked through these conflicts of authority. Cyberspace may increase the incidence of these conflicts, but it does not change their nature. Old structures may have to be molded to fit this new form, but the pattern of the old will suffice.

While both sides embrace partial truths, in my view both are mistaken. It is true, as Johnson and Post argue, that there is something new here. But what is new is not a difference in kind, only a difference in degree. And it is true, as Goldsmith and Wu argues, that we have always had disputes of this form. But we have not had conflicts at this level. We have not had a time when we could say that people are actually living in two places at once, with no principle of supremacy between them. This is the challenge that we will face in the future.

This duality is a problem because the legal tools we have used to resolve these questions before were not designed to deal with conflicts among citizens. They were designed to deal with conflicts among institutions, or relatively sophisticated actors. They are rules made for businesses interacting with businesses, or businesses interacting with governments. They were not designed for disputes between citizens.

Jessica Litman makes an analogous point in her work on copyright.[20] For much of the last century, Litman argues, copyright has worked fairly well as a compromise between publishers and authors. It is a law that has largely been applied to institutions. Individuals were essentially outside copyright’s purview since individuals didn’t really “publish.”

The Internet, of course, changes all this. Now everyone is a publisher. And Litman argues (convincingly, in my view) that copyright’s rules do not necessarily work well when applied to individuals.[21] The ideal rules for individuals may not necessarily be the ideal rules for institutions. The rules of copyright need to be reformed to make them better suited to a world where individuals are publishers.

The same is true of conflicts between sovereigns. The rules for dealing with these conflicts work well when the parties are repeat players — corporations that must do business in two places, for example, or individuals who constantly travel between two places. These people can take steps to conform their behavior to the limited range of contexts in which they live, and the existing rules help them to that end. But it does not follow (as it does not follow in the context of copyright) that the same mix of rules would work best in a world where anyone could be a multinational.

The solution to this change will not come from insisting either that everything is the same or that everything is different. It will take more work than that. When a large number of citizens live in two different places, and when one of those places is not solely within the jurisdiction of a particular sovereign, then what kinds of claims should one sovereign be able to make on others, and what kinds of claims can these sovereigns make on cyberspace?

This question is not yet answered. It is another latent ambiguity in our Constitution’s past — but in this case there is no founding international constitutional moment. Even if there had been, it would not have answered this question. At the founding ordinary people were not routinely living in multiple noncoordinating jurisdictions. This is something new.

Possible Resolutions

That there will be conflicts in how governments want their citizens to behave is certain. What is not yet certain is how these conflicts will be resolved. In this section, I map three separate strategies. The first was the dream of the early Internet. The second is the reality that many nations increasingly see today. And the third is the world we will slowly become.