An experimental contribution to the theory of customary (international) law

In their majority, public international lawyers postulate that for a new rule of customary law to originate, two conditions must be fulfilled: there must be consistent practice, and it must be shown that this practice is motivated by the belief that such behaviour is required in law. Maurice Mendelson (Recueil des Cours 272 (1998) 155) has challenged this view. He believes that the majority view ignores the fundamentally incomplete nature of public international law. He claims that the new rule emerges because mere practice leads to convergent expectations. This paper uses data from student experiments with a linear public good to show that behaviour con-verges even absent verbal communication; that convergence is guided by mean contributions in the previous round, which serve as an implicit norm; that freeriding on this implicit norm is re-garded as illegitimate; that cooperation can be stabilised at a high level if 'reprisals' are permitted. Hence the mechanism of norm formation proposed by Maurice Mendelson is fully borne out by the experimental data.