Besides the question of why humans cooperate more than other creatures (with the possible exception of social insects), the follow-up question is particularly interesting: How can cooperation be increased? For as impressive as human cooperation is – just think of global trade, division of labour or the UN – equally common are failed alliances, mistrust between individuals or nations, military conflicts, trade wars, tariff barriers, and so on.

There is a beautiful experiment that exists in many variations and reflects social dilemmas at their best and simplest. Of course, the caveats that apply to all experiments apply here as well: Be careful about drawing conclusions from artificial, extremely simple and abstract environments (these very experiments) to complex real-world environments!

Therefore, I would like to limit myself to the most robust results that have been repeated and tested in many variations – they can (almost) be trusted. However, it is still an open question whether they should be regarded as universal behavioural tendencies – this has been questioned recently (Gächter 2010; Henrich 2010).

These two articles pointed to considerable cultural differences in people’s cooperative behaviour after all. It is problematic, for example, that in cooperation experiments so far almost only Western (American) psychology and economics students have been tested (WEIRD people = Western, Educated, Industrialised, Rich, and Democratic). But it is also so practical! The students sit at the university right in front of researchers noses and usually have to do what the professors says.

However, if one takes the trouble – which these two gentlemen and their teams have done – to repeat the same tests all over the world, from Yemen to Amazonian peoples, then one comes across differences that are most interesting. For example, to start with the basics, not even optical illusions are perceived in the same way! The Müller-Lyer illusion is simply not one for some people….

And to come to cooperation: In most standard games (ultimatum games, dictator games, public goods games), we weird people are actually a weird sample, mostly at the edge of the range of variation spanned worldwide. That is, we are just not representative of “humanity”. But back to the topic: How do you increase cooperation among people?

The most effective means is also the simplest: as soon as people are allowed to talk to each other – and not sit anonymously and silently in computer cubicles at their PCs – the rate of cooperation increases enormously. Only two key ingredients are needed: Someone to take the leader’s part and a smarty-pants to come up with the most efficient “solution” for the task. And voilà – a well-cooperating group is ready!

Two other mechanisms that are well known in society are also very effective: punishment and reputation (also often combined). As soon as there is a possibility of punishment, it starts; everyone uses punishment to show the other: “That’s not how it’s done! Please contribute to the petrol money for the joint holiday trip”. Or even the popular education in road traffic… This quickly ensures that potential freeriders increase their contribution to the community. There are only two stupid things about this. Firstly, economists no longer understand the world, because why would anyone stoop to spending their hard-earned money on punishing others? Because that is a social dilemma of the second order (better to punish others, and not I have to do it). And secondly, punishments destroy part of the jointly generated good. In other words, they are not particularly efficient, since wealth is destroyed in two ways. The punisher has to invest something and has costs, and the punished gets something deducted.

Reputation works better, as everyone knows who dutifully counts their stars on Ebay and attaches great importance to who has rated which product on the internet and in what way. So if there is a chance to show how altruistic and good I am in a situation, many people do it (“Do good and talk about it”), and most others register that (“Look, the neighbour has donated to the Red Cross again!”) and react accordingly. Then a sorting also takes place: With that person (high reputation = altruist), I would like to enter into a long relationship; the freerider (bad reputation = egoist), on the other hand, I leave to the left.

So, those were the three most important mechanisms for increasing the cooperation rate – at least in the laboratory in public goods games. I’ll write about the others and how practicable they are in real life next time.

The two articles worth reading:

Gächter et al. (2010): Culture and cooperation. Philosophical Transactions of The Royal Society of London: Biological Sciences 365: 2651-2661

Henrich et al. (2010): The weirdest people in the world? Behavioural and Brain Sciences 33: 61-135