In the past, we had discussed the reasons why democracy is fundamentally flawed, needs to be overhauled, and what its replacement should look like.
Admittedly, it was a highly abstract topic. Hence why we had also discussed a simple idea of how democracy could be reformed, to look more like the Jury system, as opposed to mob rule.
Today, I’d like to propose another way of reforming Democracy. One that is more complex, but very intuitively simple. One that closely models the way we ourselves make decisions, seek advice & live our lives, every single day.
Imagine a village of 20 people, voting on whether to adopt Amendment #17412. If you’re wondering what Amendment #17412 is all about, welcome to Democracy. Too often, we find ourselves having to vote on the basis of issues that we barely know about, and are certainly not experts in. And this village is no different. Some people are highly knowledgeable about the issue, but others aren’t. Regardless, like good citizens, they all go to the ballot box anyway & vote according to their best knowledge.
And here’s what the results look like:
A seemingly random splattering of yes votes (green) and no votes (purple) all over. We’ve just witnessed democracy in action.
Now suppose we tried something slightly different. Instead of just asking people to vote on Amendment #17412, we also ask them to name someone they know personally, whom they most trust on this issue.
And here’s what that might look like:
A few patterns immediately start emerging. You can see that a few people, A1, B1, C1 & D1, are highly respected by the community. Presumably because they are knowledgeable, wise, and trusted by the people around them.
At this point, a few quick words about Google, and how they were able to make use of the above patterns to unlock new information & achieve unparalleled success. Back in the 1990s, there were numerous companies trying to succeed in the Search Engine marketplace. Lycos, AskJeeves, AltaVista… the list goes on. And yet, sometime in the early 2000s, they all just disappeared and everyone switched to Google. Their search engine, simple enough that it could be built by a couple of students in a garage, was so great that it simply blew the competition away. When you look into how they implemented this groundbreaking search engine, one of the main pillars was the PageRank algorithm.
I could write an entire article about PageRank but to put it simply, it brought democracy to the internet. It gave every single website the ability to vote on any other website, through the act of linking to it. And Google for its part, crawls through the web everyday, tallies up all the ballots, and posts the winners near the top of their search results. It is in many ways very similar to democracy… with one very important twist. Every website gets a different number of votes based on how trusted it is, as measured by the number of other websites that link to it.
To give an example, suppose the Nobel Prize winner Eugene Fama decides tomorrow to start an economics blog. It’s so insightful & interesting, that NYTimes decides to link to it. Suppose that at the same time, Joe Blow, who took a community college class on Economics, also starts an economics blog… and his brother-in-law who works at the Street Sheet links to it as well. We now have 2 blogs, on the same topic, carrying the same keywords, and each having a link from a newspaper. A purely egalitarian search-engine would give them equal visibility. But Google’s knows that NYTimes is a lot more trusted than Street Sheet, since it has accumulated millions of people linking to it. Hence, it gives NYTimes thousands of ballots more than Street Sheet, and this in turn gives Eugene Fama’s blog a huge well-deserved boost over Joe Blow’s.
It’s a very simple concept, and yet, looking at the quality of search results returned by Google as opposed to its now-extinct competitors, it makes all the difference in the world.
Now suppose we applied this same algorithm to our village’s electoral process. Here’s what the results would look like:
You can see that the patterns we noticed earlier, have produced real electoral results. Because individuals A1, B1, C1 & D1 are so highly trusted by their communities, they are given more votes than everyone else. Going one step further, since person D1 is highly trusted by both A1 and C1 who are themselves highly trusted, this gives D1 the most voting power of all. What earlier seemed like a random splattering of votes, now has a real structure to it. The signal to noise ratio of the voting process has been tremendously improved.
The exact algorithm used might be mathematically complex, but the underlying principle is dramatically simple & intuitive. When you find yourself having to make a tough decision, what do you do? Do you go on facebook, create a poll, and ask all your 700 friends to vote on it? Or do you approach a few people whom you & others trust the most? If they themselves are not sure about the answer, you might go one step further, and ask them to refer you to others who are even more knowledgeable on the topic. Ultimately, you would consider the advice given by these few highly trusted individuals, just as valuable as the semi-informed opinions of your other 700 friends put together.
And this is exactly the same principle on which Google’s PageRank algorithm is based, and has worked so successfully. We all intuitively understand this same insight & apply it in our lives everyday. It’s time we started applying it in our political process as well. Doing so can change our electoral map from a noisy random splattering of votes, into a highly sophisticated & structured network of trusted information. We may not all get an equal number of votes, but that’s fine. After all, we’re not all equal experts in governance or public policy. However, we all would get an equal say in deciding who the trusted members of our community should be. It could be your brother, your father, your daughter, your teacher, or your priest. And ultimately, trusting in the wisdom of the wisest individuals in our community, will take us a lot further than polling our entire facebook feed.