NC Media Watch

A quest for reason and accuracy in letters to the editor, guest editorials and other issues of interest to the citizens of Western Nevada County.

Friday, November 25, 2005

Some thoughts on Plato

Quoted in a letter to Editor in the 25nov05 Union:
"Just as it would be madness to settle on medical treatment for the body of a person by taking an opinion poll of the neighbors, so it is irrational to prescribe for the body politic by polling the opinions of the people at large." — Plato
I don’t know exactly where the writer was going with this quote since that was the sum total of his submission. But here are a couple of takes worthy of our consideration.

1. Our main stream media (MSM), both left and right, like to pillory or praise a politician for going with or against the latest poll of the people depending on which axe they’re currently grinding. That’s OK, but the report is almost always cast as the politician either correctly doing what the people want on a given day, or acting counter to the people’s will du jour. The message is always the same - the good politician should be going with the will of the people, the bad guys go against the people. But that’s not what we elected the politician to do else we could write a computer program that would automatically follow the latest poll and cast its vote accordingly. This would save us a lot of grief and money. What we have forgotten (is it even taught in the schools today?) is that we elect a politician with a set of values, beliefs, and reasoning powers and turn him/her loose for the duration. In doing so we express our belief that this individual will do the closest (according to our lights) to the right thing when any future situation rises. And that’s all; we don’t get to micro manage them issue by issue. If we don’t like them, we get to vote their fanny out of there when they again stand before us. This is the way our system is supposed to work no matter what the poll-reading MSM professional journalists tell us in the interval.

2. On the other hand. Recent research in the halls of academia has shown that there is something very real called group wisdom. In short, one can show that the opinions of the individuals can be cooked up in certain ways to yield single prescription that is better than picking the brains of some random individual from the group. Marketing people have been doing something like this for years for designing and positioning products. But now it appears that it even for medical treatments there might be less “madness” in the aggregate opinion of a group. Professor Charles Plott of Caltech is one the pioneers in this area (also very possibly a future Nobel laureate) and today lots of companies are distilling much wisdom from such groups. Of course, it also pays to constitute your group with some care. The electorate today is composed with no care at all. Very little beyond fogging a mirror is a prerequisite for casting a vote, and we frequently hear of elections where even the mirror has been abandoned. So since we can’t carefully compose the group on a per issue basis, maybe Plato’s observation should stand – vote them in, cringe while they ‘serve’, vote them out, and repeat.

Click here for a private e-mail comment. For public comment select comments below.

3 Comments:

Blogger anonymous said...

"Of course, it also pays to constitute your group with some care. The electorate today is composed with no care at all. Very little beyond fogging a mirror is a prerequisite for casting a vote, and we frequently hear of elections where even the mirror has been abandoned."

This is precisely NOT what the 'wisdom of crowds' idea is getting at. People guess best in groups not when there is a great deal of expertise in a group, but instead where there is a large spread in the knowledge held by the group. ie. What you need is not a 'smart' group (as attractive as that might sound to those who consider themselves part of that group) but a group with widely varied strengths and information.

Fri Nov 25, 09:40:00 PM PST  
Blogger George Rebane said...

Not quite O Faceless One. Extracting an aggregate measure of wisdom from an arbitrary crowd still faces the GIGO problem. None yet has derived the diagnosis of a patient’s blood disease, the design of a correct portfolio, or the optimum flight profile for an interceptor from such pollings even if such groups were seeded with an expert or two (signal-to-noise and all that). But H-P has gotten the best sales predictions from aggregating the group wisdom of its expert sales force. Our Founders even worried that the Grand Experiment would not work because of it, and they may yet be right. Thomas Jefferson’s ‘A nation ignorant and free, that never was and never shall be.’ summed best the quandary of an electorate seeking to perpetuate something as general as a republican democracy. To my knowledge Charlie Plott and colleagues are doing the best work in this area. If you want to maintain your claim I suggest you publish it, get your tux spruced up, and make reservations for Stockholm – glory awaits!

gjr

Sun Nov 27, 09:04:00 AM PST  
Blogger anonymous said...

"To my knowledge Charlie Plott and colleagues are doing the best work in this area. If you want to maintain your claim I suggest you publish it, get your tux spruced up, and make reservations for Stockholm – glory awaits!"

I think you are building a strawman here.

My basic point is, as I understand it, that the cleverness of groups stems not from adding deep domain knowledge by adding 'smarter' people, but by the broadening effect you get from adding people with additional bits of knowledge.

I'm not suggesting that people completely ignorant in a subject matter (like goose weights in a weighing contest) add to the correctness of a group's guess.

As for 'smarter' people improving the results of elections, that is a different issue. My own gut feeling is that intelligent people tend to overestimate their abilities in areas outside of their own personal experience (particularly academics), which makes an election a tough nut, especially since it isn't really choosing policy, but is choosing policy via a proxy.

Sun Nov 27, 05:46:00 PM PST  

Post a Comment

<< Home