>"The heart of the problem is not how we vote for officials - it's that we vote for officials, instead of getting to vote on issues.

>Americans are proud of being "governed by the people", yet a citizen has no effective way to have any influence on any particular issue! If it's very important to me to promote gay rights or environmental responsibility, I'm supposed to vote for a Democrat? How effective is that?

>We need to ditch representative democracy if we want democracy. (The next question is whether we want democracy.)"---PhilGoetz

The main problem with direct democracy is that we are reliant on "the people", who may be ill-informed and not make correct choices on issue questions. With a representative democracy, you may have intelligent and rational actors who would make better policy choices. PhilGoetz may disagree though, and believe that it is important to enfranschie "the people" in policymaking...

Rather than rely on philosophical discussion based on values, I propose an experiment to find out if PhilGoetz' Direct Democracy works.

I start up a simulation (which I will not name so that you don't play the simulation ahead of time). I will give you Policy Questions based on the simulation, where you will simply vote Yes or No. Majority rules. (To make it more interesting, I'll have each vote represent a random "interest group", with control over entire voting blocs.) Anybody can change their vote at any time. If people don't have the time to vote, then can develop a "profile" which would allow them to vote in proxy. Voting will end after a specific period of time, or the moment the vote crosses over the majority threshold, and stays over for a required period of time.

The simulation will end in a war against an NPC country. If you "win" this war, you win the simulation, the Direct Democracy works, and then future experiments could lead to people comparing the effectiveness of different types of "democracy" in creating good policy. If you "lose" the war, the government is destroyed, and you lose the simulation. Direct Democracy may has some problems and need to be modified or abanonded.

I will limit the amount of information I will give you. I'll only give enough information so you understand how the simulation works, but no more. It's up to you to decide what is the best thing to do.


New Comment
12 comments, sorted by Click to highlight new comments since: Today at 1:28 AM

I'm confused. How is a direct democracy voted on solely by LW people going to demonstrate whether they work in general or not? Not only are people here more informed on average, but from the okCupid commenting in the Open Thread I get the feeling that we're a relatively homogenous bunch.

You don't know the rules of the simulation, you only know enough to be able to make desicions within it. I think that should limit the 'informedness' of the voters in question.

And I am merely testing to see if direct democracy works...at all. If it doesn't work well here, then it does act as evidence against working in a less "homogenous" society. (I think. Maybe a less "homogenous" society might have a better functioning direct democracy, with new ideas, but I still think one test is better than none.)

But.. if you limit the ability to be informed, isn't this just a group version of trying to guess the teacher's password?

Anyway, I'm interested in participating out of curiosity to see how this experiment will be run.


Same here. Thought to be perfectly honest this seems more like a game than a serious experiment.

Are you planning to also have us vote for representatives to see if "if representative democracy works...at all."?

As I see it--which is very foggy right now--this experiment might make for an interesting game (and I like interesting games), but I don't see how it at all tests whether or not the principle of direct democracy works.

You haven't specified any controls, any verifiers, or really anything that should make us think that the votes inside the simulation represent even one potential form of direct democracy, much less the general class of all direct democracies we care to study.

Still, it might be fun, so I'll play.

The problem with democracy isn't that people can't vote with enough granularity. See wiki: public choice theory and one of our reluctant heros, Patri, speaking at Mises Brasil for a small taste of a very large and well developed argument against democracy.

If people don't have the time to vote, then can develop a "profile" which would allow them to vote in proxy.

This idea is interesting. If we had a direct democracy in the US with this option available, almost certainly most people would use it for the majority of issues, since they don't have time to vote on everything. Then, most of the political machinations would probably end up being about deciding what profile categories any given bill fits in. Could such a thing work? Would it work better than the current system?

For the profiles, it would be cool if one could specify some issues granularly and some issues broadly. For example, Vince Voter doesn't know very much about environmental issues but is generally in favor of more environmental protection, and he specifies a single "more protection" choice in that category. However, Vince is both knowledgable and concerned about financial regulations, so he could specify his proxy opinion in several subcategories such as national debt reduction, bank regulation, and so on.

Minor editing suggestion: use the quote format in the article editor to make PhilGoetz's text stand out more. I had a hard time spotting the end of the quote as it is now.

Does PhilGoetz provide an argument against Robert Michels' Iron Law of Oligarchy? If Michels is right, then pure democracy (in a literal sense) is incompatible with large scale organizations (like governments or political parties).

I am interested. When you say that the simulation would end in a war, do you mean economic, actual warfare, etc?

Also would it be more accurate to have the voting blocks be the ones to propose a given policy?

By war, I mean actual war that can result in the destruction of the direct democracy. Your victory in the actual war is dependent on Quality of Life, Economic Strength, and Army Strength.

I think it would be more accurate for voting blocs to propose given policies (representing referendums and initivates). It will also save me the effort for coming up with the actual policy questions myself; the voting blocs will do it for me.

I didn't see this post until now. What makes whether you win the war, correlate with whether people voted "correctly" on the questions? Do the questions have answers that the simulation knows, either when they are asked or afterwards?

New to LessWrong?