Immoralistic ChimpsPublished 4 September, 2012
Hello, and welcome back. Yes, it’s been some time since I added a post to this blog, so thanks for checking back in. I had a somewhat frantic summer, and the rigors of travel (and a pair of week-long holidays, I confess) left me with no time to devote the proper attention to the blog, but I plan to resume now that the academic semester is set to begin here in the U.S. I will also add a couple pictures from my summer to add to the excitement of the resumption of posting. I received some suggestions for posts over the summer, and this is my thanks for the suggestions, my apology to those of you who had good ideas while I was away, and my invitation for suggestions now that I’m back.
And, yes, I know that “immoralistic” isn’t a word.
Anyway, jumping right in, an article published online in Proceedings of the National Academy recently caught my attention, entitled “No third party punishment in chimpanzees,” by Riedl et al. (Before I go any further, here is a picture of some bears. See Figure 1.)
In the studies reported, the authors distinguish between two different kinds of punishment. The first kind is vengeance, harming an individual who previously harmed the punisher. The second kind is third party punishment, harming an individual who harmed a third party.
To look at this, Riedl et al. presented chimpanzees with situations in which another chimp took food from the focal chimp, after which, a rope could be pulled to open up a trap door, causing the thief’s food to fall down and away, out of reach. In these second party treatments, when the punisher was dominant to the thief, chimps punished about 40% of the time. In the relevant comparison treatment – when chimps could pull the rope to punish an individual who had stolen from a third chimp – chimps punished about half as much, around 20%: an amount similar to (and not significantly different from) how often they pulled the rope in control conditions in which no theft had taken place. The authors conclude that “chimpanzees do not punish third-party violations of cooperative behavior)” (p. 3).
I think this work is interesting for a number of reasons, but the principle reason is the distinction between 2nd and 3rd party punishment. In some ways, explaining 2nd party punishment from an evolutionary standpoint isn’t all that difficult. If you and I interact many times, then if I harm you when you harm me the first time – and if you infer that I’m likely to respond to similar harms in the future with a similar reaction – then you’re less likely to harm me a second time in order to avoid my vengeance. The logic of deterrence suggests that organisms in ecologies with repeat interactions – in addition to additional constraints – might be expected to be vengeful because of the benefits of teaching others the costs of harming oneself.
Third party punishment, imposing costs on those who have harmed a third party, is the subject of considerable debate in the literature on humans. The deterrence argument is more difficult to make when the harmed individual is not the individual doing the punishing. The deterrence argument is more difficult still to make to the extent that third parties impose costs on individuals who have not harmed third parties. To take one of my favorite examples, punishing people who have sex with corpses maybe deters people from having sex with corpses, but it’s not clear how I – or anyone – is better off with less corpse-sexing going on.
The distinction is also important in the context of multi-party interactions. Take, for instance, Toshio Yamagishi’s classic study published in 1986 – “The provision of a sanctioning system as a public good” – in which subjects could pay a cost to impose a penalty on the lowest contributor to the public good in the group they were in. (See also Ostrom et al., 1992). Is paying for punishment in such studies revenge, imposing costs on an individual who could have contributed to the public good but chose not to? Or, on the other hand, given Yamagishi’s finding that there’s more contributing to the public good when sanctioning is possible, should punishment in public goods games be assumed to be more like third party punishment, insofar as the low contributor also harmed – or, at least, didn’t help – the other two members of the group? Careful methods are needed to distinguish these possibilities (and see work by Jeff Carpenter and colleagues for a good example.)
In any case, the present study suggests that chimpanzees, unlike humans, might not be moralistic punishers, imposing costs on individuals who have harmed a third party. A continuing mystery is why humans are.
Ostrom, E.,Walker, J., & Gardner, R. (1992). Covenants with and without a sword: Self-governance is possible. American Political Science Review, 86, 404–417.
Riedl, J., Jensen, K., Call, J., & Tomasello, M. (2012). No third-party punishment in chimpanzees. Proceedings of the National Academy of Sciences.
Yamagishi, T. (1986). The provision of a sanctioning system as a public good. Journal of Personality and Social Psychology 51(1), 110-16.