By Thomas Schultz, SNSF Research Professor.

Law does not necessarily have to be something made by states. Perhaps this is what it should be. But there is nothing inherent in law that makes it creatable only by states. The world had law before it had states. At the end of the day, law is whatever we want to recognise as law. Today, some situations where norms are not created by states seem to cry out for the right to be recognised as law. eBay is a good example.

Millions of transactions are concluded on eBay every day. Some of them, quite normally, lead to disputes. In eBay’s context, this presented a challenge. Imagine you buy a digital camera worth 500 francs from someone in Hong Kong. The camera arrives but the pictures it takes are all distorted (and it’s not because of who you photograph). The seller says he will not t help you, claiming you must have dropped it.

Millions of transactions are concluded on eBay every day. Some of them, quite normally, lead to disputes. In eBay’s context, this presented a challenge. Imagine you buy a digital camera worth 500 francs from someone in Hong Kong. The camera arrives but the pictures it takes are all distorted (and it’s not because of who you photograph). The seller says he will not t help you, claiming you must have dropped it.

What do you do? File a lawsuit? This makes no economic sense. Go to a private form of dispute resolution, such as arbitration? Given the flexibility and scalability of arbitration, this seems more promising. Yet the classic view would still require that the arbitrator apply the law, and ‘the law’ in that view is the law of a state, for instance that of the place where you live, but possibly, depending on the circumstances, it could also be the law of Hong Kong. Simply examining which law should in fact be applied and what it says exactly will most likely create legal fees worth more than the camera. In theory, the law has a solution for such a case, but in practice there isn’t anyone to apply it. In theory the law applies; in practice it is a lawless situation. Or is it?

eBay, in essence, fixed this problem by reconsidering what the notion of law means. (Actually, that is not true. The people at eBay did not engage in analytic legal philosophy. But what they did effectively comes down to the same result.) To see the point, let us quickly roll back to the origins of eBay.

Initially eBay was a very small marketplace on which a group of friends sold a few random items. Stories about misbehaviour in that context could easily be shared with other members of the group, allowing the usual mechanisms to enforce social norms to kick in (such as excluding people from a group).

Things became more complicated when the marketplace grew to a few thousand users: it was no longer possible to share stories about misbehaviour with the entire group. What eBay did was to put together a group of six individuals called “The Posse” (a posse is historically a group of people on whom a sheriff would call to enforce the law). Social norms and best practices had meanwhile emerged informally on the marketplace. The Posse’s job was to monitor compliance with these norms. In cases of violation, they posted the names of offenders on electronic bulletin boards.

eBay continued to grow. As the community reached millions of users, The Posse was no longer sufficient. Social norms and the usual social control mechanisms within a group were just not enough. New users were not aware of the social norms that had developed on eBay. Monitoring the behaviour of this crowd was no longer feasible. Too many people were listed on the bulletin boards dedicated to misbehaviour; the list could not possibly be checked before each transaction.

In reaction eBay introduced two things to compensate for the effective lawlessness of the place.

The first was a set of written user policies, posted on the Internet, which ‘codified’ the social norms that had emerged. They progressively became more developed and more precise, and regulated more and more aspects of eBay “life”. eBay users could even elect representatives who would directly discuss possible amendments to these rules with eBay employees.

The second was a cheap, computer-assisted dispute resolution mechanism that progressively learned from past cases. It referred to nothing but the user policies to resolve the cases.

A few years later, a group of researchers asked eBay users who had satisfactorily resolved their dispute using this system what they thought the applicable law was — that is, they inquired what law eBay users thought determined their rights on the marketplace. They expected that Americans would say US law and Germans German law. But the overwhelming answer was “eBay law”. This was so even when these eBay rules openly breached the national law (for example German law) that a court of law would have applied.

A good law professor would have taken the occasion to lecture them on the fact that the law that determines their rights is still the applicable law of a state; it is not because a transaction takes place online that the law no longer applies. Or perhaps not. Perhaps the law professor sees that his or her theory only works in theory — even if that theory relates to something as fundamental as the definition of law itself.

This article was published in Globe, the Graduate Institute Review and in Le Temps on 19 March 2015.