First author

Papua New Guinea offers scientists studying social interactions something of an island 'laboratory'. It is home to many small indigenous groups with strong, cohesive identities and distinct cultural and social norms. In addition, their social systems of justice are less centralized than those of modern Western society. Helen Bernhard and her colleagues from the Institute for Empirical Research in Economics at the University of Zurich, Switzerland, went there to test how judgement of altruistic behaviour, such as fair sharing of goods, depends on group affiliation.

The scientists refereed a series of third-party punishment games involving 195 people from two indigenous groups. Games comprised three players. One shared a given amount of money with a second, and the third judged the fairness of the transaction and had the chance to 'punish' the first player for unfair behaviour. The aim was to find out whether the 'judge' favoured members of his or her own group when deciding the fairness of others' behaviour. And group affiliation did have an important role — judges favoured members of their own group, even supporting unfair behaviour (see page 912). Nature caught up with Bernard to discuss this work.

What was it like working with the New Guineans?

They were eager to take part and enjoyed the games, which was very motivating. But after having explained the same game and the same examples more than 200 times, I was also happy when the experiments were over.

Did you feel the New Guineans were studying you as well?

I am sure my presence and the games provided an interesting topic for conversation. But because I grew up in Papua New Guinea and lived there for 16 years, I probably didn't behave like a complete foreigner.

Switzerland is multiethnic. Did this have any influence on your choice of research?

Yes. Given the omnipresence of groups in our modern societies we felt it was necessary to understand the effects of group affiliation on human behaviour. Neglecting this would have left an important and interesting feature of human behaviour unexplained.

Is this work relevant to modern societies?

The New Guinean clans are a special subject pool, so we must be careful about generalizing our results. However, a subsequent study of altruistic punishment in the Swiss army replicated the result that the willingness to punish is much higher if the victim of the violation belongs to the punisher's group. This indicates that our results might also be applicable to modern societies.