Some time ago I came across a book which, I dare say, revived my faith in Humanity. We all know that news tend to be pessimistic, in the sense that they depicture people's mistakes more often (and far brighter) than what they do well. Moreover, positive remarks are usually restricted to the very best in any given activity; it's no surprise that we, the John Does of this world, do not find them flattering. Adding up to the pot, I've read so much about irrationalities and biases in our thinking that made me wonder why did we manage to survive so far. This Wisdom of Crowds, by James Surowiecki, is heart warming in the way it reveals the potential we all share. If not alone, then together.
The main idea behind the book is in fact quite simple: given a (not necessarily large) number of people, their average prediction/decision is almost alwaTys better than most of their individual ones. Most strikingly, it is also quite often better than the one picked by the smartest of them! Note that, by average, I mean aggregated in an adequate way; similarly, the term smart was also employed quite loosely but I believe you got the picture.
From guessing the number of beans in a jar, predicting sports outcomes, evaluating shares in stock markets or finding a lost submarine the author presents us with a wide and interesting variety of corroborative examples. Oddly enough, the very first example depicted in the book came from one trying to demonstrate its very opposite: the stupidity of crowds.
Obviously, there are some underlying assumptions. The wisdom of crows does not apply to every possible subject, although it appears to be remarkably robust to environment complexity and lack of information. There are conditions that must be met, such as independence of thought/action, the existence of an adequate aggregation function and coherent responses (that is, people must act in conformity with their beliefs and not try to play politics). Its application to decision making also raises other issues, such as responsibility and accountability. Still, it is clear that, if used with parcimony and awareness, the wisdom of crowds has plenty of applications.
The last chapter talks about the implications of the wisdom and its assumptions in democracy, and the reasons why it often fails. Hopefully by now I've made a compelling enough argument for you to go buy the book. If not, well, I guess I'm not as good a writer as Surowiecki is.
Filipe Baptista de Morais
Subscrever:
Enviar feedback (Atom)
Interesting. I wonder in what way this so-called "wisdom of crowds" affects public vote in Switzerland (check "direct democracy"). Could you summarize the last chapter's conclusions?
ResponderEliminarWell, I'd rather have my readers read it themselves, but I can make some short comments.
ResponderEliminarThe authors have an optimistic view of the democratic system, opposed to the traditional economical/game theory perspective which expects everyone to behave in its own interest only.
Problem is, people are horrible misinformed. Moreover, they are often put aside of the political debate, which is usually only a privilege of but a few experts and political figures. A political scientist, James Fishkin, has presented and experimented a couple of (somewhat utopian) ideas to overcome this; try looking for "Deliberative Polling" and "Deliberation Day" along with his name.
Another important limitation to the wisdom of crowds is the far-from-immediate feedback of political decisions; this makes it really hard to evaluate a decision's outcome.
Finally, since there is no objective definition and/or measurement of the "common interest" there is also no objective answer to decision making in politics. It is thus quite hard to say whether people are doing a good job voting. It does seem obvious that, were we more informed and less biased, we could make better decisions (whatever those are), leading us back to ideas such as Fishkin's. Of course if we were econs those would not be needed since people can look for information themselves. Thing is, we don't. It's not really a matter of lack of interest, but more a weird trait of human nature. For more, much more on this check Nudge by Thaler and Sunstein.
Finally, although there are no means to measure the quality of democratic decisions, there does not seem to be any reason to believe that any other systems (dictatorship, aristocracy,etc...) would fare any better. The authors go a bit further. They had previously mentioned cooperation and coordination problems which do not have definitive and general solutions. The problem at hand ("How do we live together? How can living together work to our mutual benefit?") can thus be seen as the ultimate problem of cooperation and coordination. Democacry can then arise not as a means for finding answers, but as an answer in itself. Quoting the final sentences,
"The decision that democracies make may not demonstrate the wisdom of the crowd. The decision to make them democratically does".
There is yet another question, which is how to exert democracy? Government organization and poll structures are, in my opinion, far from being closed issues, although discussing any alteration to them is certainly frowned upon and considered Tabu. But this is completely out of the scope of both the book and this post.
ResponderEliminar