Punish or Perish?

Michael Giberson

An article recently published in Science reports on research demonstrating that while many people initially prefer participating in groups without the ability to punish or be punished by others, over time subjects migrate into a group allowing punishment. Or, as the authors — Özgür Gürerk, Bernd Irlenbusch, and Bettina Rockenbach — explain it in their abstract:

We show experimentally that a sanctioning institution is the undisputed winner in a competition with a sanction-free institution. Despite initial aversion, the entire population migrates successively to the sanctioning institution and strongly cooperates, whereas the sanction-free society becomes fully depopulated. The findings demonstrate the competitive advantage of sanctioning institutions and exemplify the emergence and manifestation of social order driven by institutional selection.

Haven’t had a chance to read the article yet, but clearly a trip to the library is in order. A New York Times article reports a reaction from Elinor Ostrom:

“I am very pleased to see this experiment being done and published so prominently,” Dr. Ostrom said, “because we still have many puzzles to solve when it comes to the effect of punishment on behavior.”

Dr. Ostrom has done fieldwork with cooperatives around the world and said she often asked other researchers and students whether they knew of any long-lasting communal group that did not employ a system of punishment. “No one can give me an example,” she said.

The Times also quotes Columbia sociologist Duncan Watts saying, “the mystery, if there is one, is how these institutions evolve in the first place … before it is apparent to anyone that they can resolve the problem of cooperation.”

But as was noted a few paragraphs above in the Times story, subjects who felt taken advantage of experienced “deep frustration and anger.” I suspect that feelings of deep frustration and anger could lead to sanctioning behavior without anyone having first thought through the institutional ramifications.

Erte Xiao, a PhD student at George Mason University and Research Affiliate, Institute for the Study of Labor (IZA), has been doing related reseach for her dissertation. For example, a paper by Erte and Daniel Houser, published in the Proceedings of the National Academy of Sciences, suggests that experimental subjects are more likely to resort to costly punishment strategies when they are prevented from directly expressing negative emotions.

The next research step may be to expand the group choices available to the subjects in the Gürerk, et al., research to include a setting in which no-cost verbal sanctioning is also possible in addition to costly economic sanctioning. Maybe verbal sanctions would be treated as cheap talk and ignored, or maybe they would be non-destructive tools to enhance cooperation.

Hat tip to Stephen Dubner at the Freakonomics blog.


5 thoughts on “Punish or Perish?

  1. Is there a clue here to how a workers’ paradise devolves to a police state?

  2. Maybe, but in the experiments the punishment power was shared by everyone, not held by a favored few.

    By the way, which one would you call a workers’ paradise? In the sanction-free group I guess you could say there is no “boss” to punish you for shirking, and maybe that is one idea of a workers’ paradise. On the other hand, the sanction-free group tends to have more free riders (free loaders?), whereas the group with sanctions exhibits a lot more “from each according to his ability” contribution to the general good.

    Whichever one is the workers’ paradise, the workers’ choice in the experiment was clearly the group with sanctions.

  3. It’s difficult to see from here exactly what “group” means in the experiment. So, I’m guessing that it is a hard bounded unit like a lifeboat or submarine. In that context, an individual’s acts affect everyone, possibly severely. Military units have military discipline – with its sanctions – for good reason.

    In a group that would see individual consequences to actions, a community for example, the business that provides poor service goes out of business. The business owner is affected but, he has no claim on the group and the failure of his business has no dramatic effect on the group. This is not a sanctions regime but, it is effective discipline.

    From each according to his ability, to each according to his value to the group.

  4. I am extremely disappointed how scientists in the social science are now making such bold claims regarding the meaning of their research. As a trained social scientist who understands the statistical and experimental design issues involved with the field, I am appalled that these “scientist” tout their work in the manner they do.

    Any social scientist who has been trained using Keppal and Cook & Campbell would never make the assertions these researchers so easily make about the implications of their findings. The issues of Internal, Construct and External validity should provide sufficient caution in making strong assertions about implications.

    In addition, I NEVER hear any social scientist among peers make bold statements regarding the robustness of the findings or the undisputable nature of the implications. I use the word NEVER very deliberately. We all know that our research has not the power to speak in absolutes. Going a step further, I would say that most research in the social science has SO much in question that a very cautious tone is demanded if one is to be intellectually honest.

    Going back to the recent trend of asserting behavioral theory findings, social-psych experiments as definitive is WRONG from an INTELLECTUAL and SCIENTIFIC perspective. It is nothing more than GRAND STANDING.

    I am CERTAIN that a group of this author’s peers could easily challenge the robustness of this findings on a multitude of levels this piece (Internal, Construct a External Validity fronts).

    In Summary, it saddens me as a social scientist to see fellow scientists acting as attention whores. We are all cheapened by their actions. The end result will be that Americans see us as hacks.

  5. I am extremely disappointed how scientists in the social science are now making such bold claims regarding the meaning of their research. As a trained social scientist who understands the statistical and experimental design issues involved with the field, I am appalled that these “scientist” tout their work in the manner they do.

    Any social scientist who has been trained using Keppal and Cook & Campbell would never make the assertions these researchers so easily make about the implications of their findings. The issues of Internal, Construct and External validity should provide sufficient caution in making strong assertions about implications.

    In addition, I NEVER hear any social scientist among peers make bold statements regarding the robustness of the findings or the undisputable nature of the implications. I use the word NEVER very deliberately. We all know that our research has not the power to speak in absolutes. Going a step further, I would say that most research in the social science has SO much in question that a very cautious tone is demanded if one is to be intellectually honest.

    Going back to the recent trend of asserting behavioral theory findings, social-psych experiments as definitive is WRONG from an INTELLECTUAL and SCIENTIFIC perspective. It is nothing more than GRAND STANDING.

    I am CERTAIN that a group of this author’s peers could easily challenge the robustness of this findings on a multitude of levels this piece (Internal, Construct a External Validity fronts).

    In Summary, it saddens me as a social scientist to see fellow scientists acting as attention whores. We are all cheapened by their actions. The end result will be that Americans see us as hacks.

Comments are closed.