September 30th, 2010

Via Henry Farrell at Crooked Timber, a truly terrific quotation from Daniel Kahneman:

Because we tend to reward others when they do well and punish them when they do badly, and because there is regression to the mean, it is part of the human condition that we are statistically punished for rewarding others and rewarded for punishing them.

Read the post for the IDF pilot training story that inspired this. The quotation is from Kahneman´s autobiography on the Nobel site. For a heavier-duty synthesis of his life´s work on cognitive bias, see his 2002 Nobel lecture .

Kahneman and his colleague Amos Tversky (who died before he could be Nobelized with him) have put Francis Bacon´s insight about the inbuilt biases in human understanding – ¨idols of the mind¨ in Bacon´s language – on a rigorous scientific basis. A pity that it took 350 years for anyone to follow up Bacon´s warnings and not just his scientific triumphalism. We would have been spared a lot of persuasive nonsense based on perfect rationality – an idol of the tribe (a misleading short cut) that became an idol of the theatre (a scholastic dogma).

The only rationality we can ever have is bounded.

Share this post:
  • Twitter
  • StumbleUpon
  • Digg
  • Reddit
  • Facebook

7 Responses to “Daniel Kahneman, iconoclast”

  1. Joe S. says:

    Herb Simon did bounded rationality well before Kahneman-Tversky. Of course, nobody listened to him. Part of this is because he didn’t have all the experimental evidence of Kahneman-Tversky. But most of it is because nobody really wanted to listen to him–perfect rationality models were fresh and kicky and (cui bono?) well-funded.

  2. Bruce Wilder says:

    “We would have been spared a lot of persuasive nonsense based on perfect rationality – an idol of the tribe (a misleading short cut) that became an idol of the theatre (a scholastic dogma).”

    Sounds like the story of Economics. Nobel or not, I see little evidence of relief in sight.

  3. paul says:

    Perfect rationality (and all the other variants where you can get completely tractable predictions) is an excellent rationale for building certain kinds of failsafe-free systems.

  4. WIdo Incognitus says:

    It is interesting that what makes Kahneman a destroyer of icons is his defense (sort of) of people as we is.

  5. I think you have cause and effect backwards here.
    Sure, the assumption of perfect rationality was poisonous to economics, but if economics were purely an academic discipline, a kind of anthropology with numbers, that wouldn’t have mattered much.

    The point is that the conclusions that economics could be made to produce were very useful to those in power, and the fundamental poison is that those in power have so much more a say in how society works. If, somehow, the whole of economics is transformed into a discipline that no longer says anything especially useful to those power, it will revert to a marginal social science and those in power will find their justifications elsewhere. Theology had its run as a source of social claims for 1000+ years, biology had its run from Spencer until the Nazis ruined it for everyone else; doubtless another source of justification for the rich will arise. How about physics: “Thermodynamics tells us that an engine can only do useful work if it mediate between two different temperatures; and it is more efficient the greater that temperature difference. So too it is with the economy. A great concentration of wealth is like the furnace of a power plant, making it possible to create the electricity of that plant. Bring everyone to the same level of wealth, and it’s like trying to create electricity with no furnace — ain’t gonna happen!”

    So while Kahneman et al may improve both economics and psychology as intellectual disciplines, I think it is very premature to imagine that they will thereby improve politics and society.

  6. James Wimberley says:

    Nobody has picked up the direct implication of the quote that we have a cognitive bias, via the statistical blind spot, towards overestimating the effects of punishment and underestimating those of reward. This would seem to fit the wildly over-punitive US criminal justice system, which is more sensitive to public opinion (elected judges, prosecutors and sheriffs, and a big role for juries in sentences and tort awards) than that of European countries. In France, the only democratically accountable figure in the entire machinery of criminal justice is the Minister.

  7. Brett Bellmore says:

    “The point is that the conclusions that economics could be made to produce were very useful to those in power, and the fundamental poison is that those in power have so much more a say in how society works.”

    An important observation: Essentially every field funded by the government is warped to produce results more popular with the people running government. This is why for instance, almost all economists assume that virtually all forms of government spending have “multipliers” greater than one: Politicians don’t want economists telling them, (Let alone the voters!) “If you spend more money, you’ll make the nation poorer!”

    Even it it’s true, for many of the things they might spend it on.