Skip to main content

Verified by Psychology Today

Scott M. James
Scott M. James Ph.D.
Punishment

Living by the Bronze Rule

Natural selection's solution to cheating your neighbor.

How much would you pay to have someone killed?

At first blush, the question is offensive: 'I would never pay to harm another!' But the truth is almost certainly this: It depends. Imagine your child is brutally raped and murdered. Because you live in lawless state, however, the criminal goes unpunished. Suppose you happen to know people (who know people) who, for the right price, can "make stuff happen." Are you so sure you wouldn't consider the proposition?

While we may aspire to the Golden Rule, we are probably best described by the Bronze Rule: Do unto others as they have done unto you. Evidence from behavioral economics is revealing. People, it turns out, routinely pay to punish.

Take the Ultimatum Game. An individual receives $10 and is told to split that money, however he would like, with another (anonymous) participant. The "receiver" then can either accept the offer or reject it. If she rejects it, no one gets anything. The median offer is around $4. From an economic standpoint, this is of course irrational since any amount is better than no amount; why leave with nothing when you can at least leave with $3? But what the evidence shows (and what participants implicitly know) is that $0 can in fact be better than $3. For rejecting the low-ball offer punishes the "proposer" by stripping him of his assets. Participants are quite willing to pay to punish. (One study in Indonesia found that participants were willing to give up the equivalent of two weeks' wages in order to punish.)

Even more striking is a variation on the game involving a third-party observer. The observer receives $20 and is required only to observe several rounds of what's called the Dictator Game: The Dictator makes an offer and the receiver has no choice but to accept the offer. But the observer has this option: if he so chooses, he can pay to reduce the Dictator's take. Here's the remarkable thing: Instead of walking away with $20, most observers are compelled to (pay to) punish a complete stranger. And for what? For shorting another complete stranger! It's an itch we can't help but scratch.

But why? The evolutionary account seems plausible: Among early humans, the threat of punishment functioned as a constraint that benefited everyone, for each gained more from the general trust of his peers than from cheating his neighbor and risking punishment.

This is exactly what is revealed by Public Goods experiments. In these experiments, participants have the option of investing their money in some public project, which offers a guaranteed return (of, say, 20%). However, no participant knows what others are investing. Moreover, returns are distributed equally among participants--regardless of how much one has invested. Hence, if everyone invests heavily, each maximizes her returns. But investing heavily is risky since if I'm the only one to invest heavily, then others benefit at my loss. Behavioral economists Ernest Fehr and Simon Gachter ran one experiment where punishment was available after individual investments were revealed and one where punishment was not available. In the latter experiment, public investment substantially decreased over time as people quickly lost trust in their associates. In the former experiment, punishment (and eventually just the threat) substantially increased public investment over time. So while people pay to punish, punishment appears to pay.

Up to a point, however. Psychologist Kevin Carlsmith and his colleagues showed that revenge is actually not sweet. Carlsmith's subjects were enticed into investing heavily in a public goods game by a confederate. That same confederate then cheated. One group of subjects had the option to retaliate; another didn't. Those who retaliated "reported a significantly worse mood" and ruminated longer over the slight than those didn't retaliate but who had expressed a desire to.

The upshot then is slightly paradoxical: we (together) do better when we pay to punish, but you (individually) do worse when you pay to punish. So turn the other cheek, so long as you're sure society won't.

advertisement
About the Author
Scott M. James

Scott M. James is Assistant Professor of Philosophy in the Department of Philosophy and Religion at the University of North Carolina.

More from Scott M. James Ph.D.
More from Psychology Today
More from Scott M. James Ph.D.
More from Psychology Today