Hacker News new | ask | show | jobs
by mkramlich 5134 days ago
Two citizens go to the voting booth and choose between electing candidates 1 or 2 to the Presidency. Call these citizens A and B. But at some point previous to this day both of these citizens had taken a test to gauge general intelligence. The test is not perfect but is better at ferreting out signal than having nothing at all. Here's that test (for sake of argument):

The test describes an apparent murder case. Here are the known facts. The victim was found dead by bullet wound. Seven different witnesses claim they saw suspect X enter the victim's house carrying a gun, then heard a gunshot, then saw suspect X leave the house and flee the scene. One neighborhood resident, a 90 year old man with dementia, told an officer he thinks suspect Y did it. When asked his reasons, he said, "Because he's a bad man!" though admits he did not personally see or hear anything directly related to the murder scene or time. The test then asks the test taker: without knowing for sure, in an omniscient way, who committed the murder, and without any additional data or evidence, who do you think is more likely to have commited the murder, suspect X or Y?

Citizens A and B take this test. Citizen A indicates X. Citizen B indicates Y.

Now someone chimes in on the Internet to say that tests are unfair and it's impossible to test for intelligence, that the government can skew tests, etc. Then election day comes. Citizen A votes for candidate 1. B votes for candidate 2. Since everyone's vote weighs the same, these two votes effectively cancel each other out.

Do you think that's a smart system? Do you think that system produces better or more "fair" results than a system in which each vote can be weighed differently? I've already obviously said which I think to be true. I just wanted to give a concrete example of how we can devise tests or filtering mechanisms that can score or filter for desirable qualities, and that the current system allows dumb votes to cancel out smart votes. The current system can be even worse than that if there are more "dumb" people than smart ones (where for sake of argument assume dumb/smart is a generalization across a wide set of qualities and signals.) To continue my example, what if there were a lot more people who thought like citizen B, than A? Do you think that would produce smarter results for the country? Or dumber results?

And yes we can objectively measure and compare between two people to determine who is more logical or well-read. Again, not in some perfect ideal way in every single possible situation -- the world is messy and imperfect -- but in enough cases and in broad enough strokes that yes we can better filter for signal over noise. I know it is possible.

Also the government chooses special criteria and devises lengthy and arbitrary rules already, all the time, for all sorts of things. You can't do D unless E, F and G are true. You're not allowed to do J unless you've previously done K and L. Lots of precedent, both in government and in the private business sector, clubs, pretty much everywhere in society. But for some reason when we go to the voting booth (figuratively speaking) everyone's vote counts the same. When serious money is on the line or someone's life we don't go take a poll of every Tom, Dick and Harry in your neighborhood, instead we go to experts or at least to people who we have a reasonable expectation will give us smarter, better advice than the average Joe. Suddenly when millions of lives are on the line, and trillions of dollars in spending (when it comes to voting on national-level political issues or candidates) we abandon that system and dramatically loosen and worsen our standards. I say it's a hilariously bad design. So obviously bad that perhaps it just might be by design and intent. But regardless, I know we can do better. So let's try.

Thank you for your feedback though!