Rational Irrationality: A Self-Proving Theory?

No Treason has written about rational ignorance and rational irrationality before, but it’s instructive and amusing to see what happens when those concepts are presented in a forum with a more mainstream readership:

Some interesting press about a very interesting new book by Bryan Caplan, an economics professor at George Mason University: The Myth of the Rational Voter: Why Democracies Choose Bad Policies.

Have a read through the comment thread there and count the Reason Magazine weblog commenters who are doing their damndest to wish away those two concepts. You’d think that once those ideas were presented, they would pretty much spell the end of the libertarian movement.

Then again, people pay no penalty for ignoring those ideas…

Tip: Jay Jardine

19 thoughts on “Rational Irrationality: A Self-Proving Theory?”

  1. There is a simple solution: don’t give equal representation to everyone.

    But how can that still be democratic, you ask? Have a look:
    http://www.metagovernment.org

    This new government rigs it so that the more you care — the more involved you want to be in a particular form of policy — the more say you have in that policy arena. *Provided* that you have something to say that respected people find respectable.

    What a simple and elegant solution to the disaster that is representative democracy.

  2. Andrew,

    The intention of the metagovernment and its associated websites is to gradually make a set of laws so superior to the status quo — and so compellingly, unprecedentedly democratic — that the people will want to change to this form of government.

    Superior according to whose values? You’re making a terrible mistake in assuming that the electorate wants justice and efficiency from government. Voters don’t want “better” laws – they want their imagined enemies crushed, their fears comforted, and most of all they want loot: manna from gub’mint Heaven, taken from other people and given to them personally by this year’s elected Santa.

    In any case, you’ve still got the classic libertarian movement public goods problem. You’re trying to produce a better government, but I get the same exact benefits from your efforts that you do, even though I’m doing absolutely nothing to help you.

    If you succeed, well then, I’ll help myself to the fruits of your labors – “thanks for the better government, sucker!” If you fail (overwhelmingly likely, considering what your models SlashDot and WikiPedia have produced to date), I still get the benefits of my plans, because I haven’t been wasting my life on yours. See how all that works? No matter what happens, I get all the benefits of your work with none of the costs.

    I could of course go on to further point out the 100% failure rate of libertarian movements to date, how a perception of expertise among an internet peer group doesn’t necessarily translate into actual, non-virtual (aka “real”) expertise, and the fact that your project will tend to become populated with folks more interested in social networking and backslapping from peers (both of which are private goods, note) than in producing a better government, but three paragraphs of truth are probably enough for now.

  3. Your benefit-for-nothing analysis doesn’t work with a recursively scored participatory system.

    if you choose not to participate, then yes, you get the benefits of the process… but not necessarily to your liking. If you want the government to act the way you want it to, then you must participate, or you lose your say.

    This has the opposite effect of most other systems: the more you care, the more say you get in how government works.

    As to whose value system, the answer is the consensus. It is a radical departure from current systems, but not unworkable. If a consensus cannot be reached on a particular issue, then there simply is no law on that issue (just like how in the U.S. there is no law against abortion, because consensus hasn’t been reached). Over time, consensus WILL build on important issues. It will have to. And unimportant issues will be relegated to debates, not laws.

    As for the shortcomings of slashdot and Wikipedia, they are not very sophisticated systems. Slashdot has a two-level, very primitive scoring system, and Wikipedia has no scoring system. This project has a recursive, multifaceted scoring system with both sums and factors.

    Bash it all you want. It’ll happen without you, if need be. The technologies all exist for governance by internet technology, and it is only a matter of time before people start to move to that model (unless we move to complete totalitarianism first, which is a distinct possibility, I grant).

    Now here’s the real question: who do you want running the new form of governance? A recursively scored system run by a minimalist nonprofit whose goal is to make the scoring system balanced, or… MySpace (ie, Rupert Murdoch)? Your choice.

  4. Andrew,

    This Metagovernment scheme reminds me of nothing so much as another attempt to build a perpetual motion device. Ultimately you can’t get around the public goods problem for government any more then you can get around the laws of thermodynamics.

  5. Hm. This libertarian philosophy is a little new to me, so forgive my ignorance. As I understand it, you are describing what Wikipedia calls the “free rider problem,” right? http://en.wikipedia.org/wiki/Free_rider_problem

    So nobody has an incentive to help out if they don’t reap the full benefits of their actions?

    First, if that is the case, then how does Wikipedia exist? Or Firefox? Or Linux? Or any of the other screamingly successful open source projects? People could have done development for themselves, but instead they shared their work and benefited from the sharing. Yet other people benefit much more than the original authors ever will. Nonetheless, http://en.wikipedia.org/wiki/Ant is one of the most informative things I’ve ever read. Why?

    Second, the scoring system described by the metagovernment project means that the more you participate, the more say you get in any particular issue. So if you really want to see a change in gun control laws, you can work a lot on that issue, develop a reputation, and become the most influential person in gun control laws. There is a direct incentive to participate heavily on issues you care about.

    Conversely, someone who does not participate in governance gets NO say on anything. In representative democracy, the most uneducated, Paris-Hilton-adoring idiot gets exactly as much say as the most educated, engaged professor. In open source governance, the more you participate, the more influence you get.

    How is that worse than voting for someone who allegedly is supposed to represent your interests on a thousand different issues? Especially if your candidate doesn’t win?

  6. So nobody has an incentive to help out if they don’t reap the full benefits of their actions?

    I would instead characterize it to say that folks tend to invest their efforts in a manner that they calculate that they’ll benefit from.

    First, if that is the case, then how does Wikipedia exist? Or Firefox? Or Linux?

    Why do people contribute to those things? Because it makes them feel good, maybe. Good feelings are a private good. Peer group acceptance, possibly – another private good. The public goods produced are a byproduct of the private goods being produced. Of course these aren’t very valuable private goods to most folks.

    A very good example of a faux public good is religious charity. It’s really a private good with a byproduct of a public good. The private good being produced is the fact that religious charity gets believers into Heaven – that’s why folks pony up. The public good (the charity) is a byproduct of that private good.

    So my question for you is, what private good does your scheme produce? It isn’t a better government, because I’m fairly confident that we’re all going to eat whatever the next election puts on our plates.

  7. The private good being produced is the fact that religious charity gets believers into Heaven – that’s why folks pony up.

    Ain’t no good works that get you into heaven. Is the love of our Heavenly Fodder that get you in heaven. You do good works because you go to heaven; you don’t go to heaven because you do good works. Dig it?

  8. The public goods are successfully produced in some cases. Linux and Firefox are fair examples. But the core producers are generally a small group of developers who have found a way to bundle their the public good with private goods. I doubt Linus Torvalds is hurting for money, even though nobody has to pay for Linux.

    The problem with “Metagovernment” is that that those who govern will have every incentive to wield government to produce private goods for themselves. Those serving themselves will have far more incentive to persevere than those serving others.

    It’s naive to think any rules can be instituted to prevent that.

  9. The problem with “Metagovernment” is that that those who govern will have every incentive to wield government to produce private goods for themselves. Those serving themselves will have far more incentive to persevere than those serving others.

    Sounds like the status quo; except that in open source governance, everyone gets to participate.

    I still don’t see the problem here. If you don’t want to participate, then you opt out of getting any private good except for by chance. If you opt in, then you have a chance to get what you want, provided the rest of the interested community lets you.

    The only way the rules help is to give incentives toward synthesis over conflict. So that if half the people want an arsenal of uzis and the other half wants gun control, they have to come up with some way to get something that will work for both sides.

    Anyone who doesn’t have an opinion one way or the other doesn’t have to participate.

  10. The only way the rules help is to give incentives toward synthesis over conflict. So that if half the people want an arsenal of uzis and the other half wants gun control, they have to come up with some way to get something that will work for both sides.

    But the winning side gets to stuff their platform down the losers’ throats. Seems to me that this devolves almost instantly into warring pressure groups – just like now.

  11. But synthesis is not about winning sides and losing sides. It’s about two seemingly conflicting sides coming up with something that works for both of them. Here’s an easier example:

    Almost nobody really wants to see lots of women having abortions all the time. Everyone is pro-life.

    And almost nobody wants the government to control women’s bodies. Everyone is pro-choice.

    So it’s really a false dichotomy to have this pro-life vs. pro-choice debate. There’s no genuine conflict here.

    But that doesn’t work in a two-party system, so we have this false conflict which benefits… the politicians.

    Get rid of the politicians, reward synthesis, and everyone wins.

  12. But synthesis is not about winning sides and losing sides. It’s about two seemingly conflicting sides coming up with something that works for both of them.

    And government isn’t about synthesis. It’s about winners getting to stuff their values down the losers’ throats.

    Either your metagovernment is going to stuff its collective values down everyone’s throats or it isn’t. If it is, then its incentive structure is the same as what we have now.

    If it isn’t going to force its collective values on everyone, well then, what good is it to the average voter? Answer is none at all. As I said earlier, the average voter wants his enemies destroyed, his fears comforted, and his pockets lined with loot, all at public expense.

    So which is it – will your metagovernment enforce its collective conclusions or not?

  13. I realize this is a few days off, but I wanted to post it here for posterity’s sake.

    I think the Metagovernment is a very interesting idea, one which I will be following with great interest. However, I just wanted to say that I feel that it’s design benefit – that those with the most interest in a topic will be the ‘most influential voters’ on that topic – is actually a detractor rather than a benefactor. It merely mimics the situation right now!

    When a CA state measure came up to increase taxes on oil companies, these companies spent hundreds of millions of dollars to advertise and create doubt in the minds of voters. They spent money to defeat the legislation with advertising, and won. The voters didn’t vote against it because they thought it was a bad idea, they voted against it because the opposition to the bill ‘spent’ more time and money on the issue, and persuaded them of all sorts of things that weren’t really part of the bill.

    I fail to see yet how a wikigovernment will improve that situation. From my perspective, the way to improve an entire society’s access to government is to switch to a direct democracy with a tightly regulated ‘equal-access’ media situation.

    If you want to get a law done up, you start a measure. If it doesn’t pass, you continue to work with your opponents to find a consensus that the large portion of society will want to support. Also, you tightly regulate how much people can ‘spend’ versus their opponents to support an issue. This way, the merits of the measure can be the topic at hand, not the advertising budget of the policy groups.

  14. So which is it – will your metagovernment enforce its collective conclusions or not?

    One of the fundamental principles listed on the metagovernment website is: “Without consensus, there is no law.” So, no, it doesn’t look like there are any enforced conclusions unless there is a pretty compelling consensus.

    As for the average voter, well, it’s up to them to decide if they care or not. If they decide not to care, then they get no vote at all on that issue.

    The people who really don’t care enough to do anything about government are also the people who are not going to be motivated enough to stop the new form of government from taking hold over time.

  15. The voters didn’t vote against it because they thought it was a bad idea, they voted against it because the opposition to the bill ’spent’ more time and money on the issue, and persuaded them of all sorts of things that weren’t really part of the bill.

    Those people all had equal votes. In a scored system, the more informed people would be more likely to have higher scores, and would also be less swayed by fifteen-second scare-ads.

    The concept of recursively scored voting has never really been tried because it takes enormous and continuous computation of everyone’s score in relation to everyone else’s score. Fortunately, we have computers now, and this task is suddenly extremely easy.

Leave a Reply

Your email address will not be published. Required fields are marked *