More or less, because both sides have to agree to the process. Then the market favours those arbiters that manage to maintain a reputation for being unbiased and fair.
This still doesn’t select for rationality precisely. But it degenerates into a different system to that of a lawyer system.
Yes, but if a side can hire a rationalist to argue their case before the judge, then that rationalist will degenerate into a lawyer. (And how could you forbid assistance in arguments, precisely? Offline assistance at least will always be present.)
And since the lawyer-like rationalists can be paid as much as the richest party can afford, while the arbiter’s fees are probably capped (so that anyone can ask for arbitration), the market will select the best performing lawyers and reward them with the greatest fees, and the best rationalists who seek money (which is such a cliched rational thing to do :-) will prefer being lawyers and not judges.
Edit: added: the market will also select the judges who are least swayed by lawyers. It still needs to be shown that the market will have good information as to whether a judge had decided because the real rational evidence leaned one way, or because a smart lawyer had spun it appropriately. It’s not clear to me what this will collapse to, or whether there’s one inevitable outcome at all.
Yes, but if a side can hire a rationalist to argue their case before the judge, then that rationalist will degenerate into a lawyer. (And how could you forbid assistance in arguments, precisely? Offline assistance at least will always be present.)
Would a lawyer by any other name still speak bullshit? Yes. But why are we talking about lawyers and judges?
And since the lawyer-like rationalists can be paid as much as the richest party can afford, while the arbiter’s fees are probably capped (so that anyone can ask for arbitration), the market will select the best performing lawyers and reward them with the greatest fees, and the best rationalists who seek money (which is such a cliched rational thing to do :-) will prefer being lawyers and not judge
You have explained well the reason that capping is a terrible idea. Now it is time to update the ‘probably capped’ part.
The market will also select the judges who are least swayed by lawyers. It still needs to be shown that the market will have good information as to whether a judge had decided because the real rational evidence leaned one way, or because a smart lawyer had spun it appropriately. It’s not clear to me what this will collapse to, or whether there’s one inevitable outcome at all.
It’s not clear to me either. I also add that I rather doubt that the market, even with full information, would select for the most rational decisionmakers. That’s just not what it wants.
Would a lawyer by any other name still speak bullshit? Yes. But why are we talking about lawyers and judges?
Because I think that in the proposed scenario, where people hire master rationalists to arbitrate disputes, these arbitrators and other rationalists who would be hired by each side independently for advice would start behaving like judges and lawyers do, respectively. (Although case law probably wouldn’t become important.)
You have explained well the reason that capping is a terrible idea. Now it is time to update the ‘probably capped’ part.
I didn’t mean they would be capped by guild rules or something like that, but rather, that the effective market prices would stay low. I’ve no proof of this, economics is not my strong suit, but here are the reasons I think that’s likely to happen:
If arbitration is so expensive that some people can’t afford it (and it needs to be affordable by the poorer party in a conflict), that’s an untapped market someone could profit from. Whenever your argument is with a poor party, you have to have an arbitrator whose fee is at most twice the fee the other party can or is willing to afford, and there’s no effective low limit here. (State-provided judges and loser pays winner’s fees do have something going for them.)
Being a better arbitrator doesn’t require direct investment of money on part of the arbitrator. So a good arbitrator who’s not getting enough work can lower his prices.
Third parties interested in seeing a dispute resolved—if only to achieve peace and unity—might contribute money towards the fee, or send volunteer arbitrators, in exchange for the parties to the dispute agreeing to arbitration. Finally, competition between arbitrators (for money and work) would eventually draw the fees down, assuming a reasonable supply of arbitrators.
What would make people choose an arbitrator that wasn’t the cheapest available? Assuming some kind of minimal standard or accreditation (e.g., LW karma > 1000), an arbitrator is inferior if he cannot properly comprehend your rational argument or might be swayed by your opponent’s master rationalist lawyer. You then have a choice: invest your money in a costlier and fairer arbitrator, or in a better lawyer so you can sway the cheap arbitrator to your own side. I do hope that one dollar buys more unswayingness than swaying-power, but with humans you never know.
How does someone prove she’s a good arbitrator in the first place? Wouldn’t you need another, more senior arbitrator to decide on that and to handle appeals? Either there’s a hierarchy, in which case the lowest ranked arbitrators are cheap (because it’s their own entry price in the business); or there’s a less centralized web of trust, and if it’s fragmented enough the whole idea of universally trusted arbitration is undermined.
I see, so rationalist arbitration of a dispute raises the stakes in status and neutral-party persuasion and that would lead to a market for lawyers? In what sense would this be damaging/ harmful?
Edit: Obviously it would be really bad if lawyers were causing the arbitrators to make bad decisions. But presumably the arbitrators are trained to avoid biasing information, fallacies etc. If you want to persuade and arbitrator you have to argue well—referencing verifiable facts, not inflaming emotions or appealing to fallacies. If all online disagreements did that than the internet would be a much better place! If arbitration leads to better standards all the better. Now, the system might be unfair to those who couldn’t afford a lawyer and so can’t present as much data to further their cause. But 1) presumably the arbitrator does some independent fact checking and 2) there are already huge class barriers in arguments. Arguments an average high school drop outs and an average Phd are already totally one sided. The presence of arbitration wouldn’t change this.
Now, the system might be unfair to those who couldn’t afford a lawyer and so can’t present as much data to further their cause.
If that’s so then there’s no point to arbitration. He with the best lawyer wins.
Put it this way: take, in the general (average) case, any decision made by an arbitrator. For simplicity, suppose it’s “A is right, B is wrong.” Now suppose party B had employed the services of the very best rationalist ever to live as a lawyer. What is the probability the arbitrator would have given the opposite judgment instead? How high a probability are you willing to accept before giving up on the system? And how high a probability do you estimate, in practice?
The purpose of arbitration isn’t to establish the truth of a question. If that were the case there would be no reason for the arbitrator to even listen to the disagreeing parties. She would be better off just going off and looking for the answer on her own. This would also take much, much longer since she wouldn’t want to leave any information out of the calculation.
Rather, the purpose of arbitration is to facilitate agreement. Not just any agreement but a kind of pseudo- Aumann style agreement between the two parties. The idea is that since people aren’t natural Bayesian calculators and have all kinds of biases and incentives that keep them from agreeing they’ll hire one to do the calculating for them. This means we want the result to be skewed toward the side with better arguments. If the side with weak arguments doesn’t end up closer to the side with strong arguments then we’re doing it wrong. This is true even if one side puts a lot more time or money into their arguments. Otherwise you’d have to conclude that arguing never has a point because the outcomes of arguments are skewed toward those who are the smartest, have done the most research and thought up the best arguments.
If agreement is more important to you than objective truth, than sure, that method will work. I just happen to think a system that optimizes for agreement at the expense of truth and facts tends to lead to a lot of pain in the end. You end up with Jesuits masterfully arguing the number of angels that can dance on the head of a pin.
Not just any agreement but a kind of pseudo- Aumann style agreement between the two parties.
Edit: If a rationalist is hired to arbitrate a dispute between two Jesuits regarding the number of angels she isn’t going to start complaining that there are no angels. That isn’t what she was hired to do. If the Jesuits want to read some atheistic arguments they can find those on their own. The task of the arbitrator is applying rationalist method to whatever shared premises the disputing parties have. But the system as a whole still tends toward truth because an arbitration between a Jesuit and an atheist will generate a ruling in favor of atheism (assuming the Jesuit believes in God because of evidence and not Kierkegaardian faith or “grace”).
Think we’ve got some fundamental disagreements here about just what it is that rationalists do. You cannot just hire them to argue anything. The ideal rationalist is the one who only ends up arguing true beliefs, and who, when presented with anything else, throws up their hands and says “How am I supposed to make that sound plausible?”
The ideal rationalist is the one who only ends up arguing true beliefs, and who, when presented with anything else, throws up their hands and says “How am I supposed to make that sound plausible?”
I reply: I’m paying you a lot of money. You’ll find a way.
When I say or hear “rationality”, I think of the tool, not of the noble “ideal rationalist” whose only pursuit is truth, not money or other personal interest.
That which can be used to argue for any side is not distinguishing evidence, whether “that” is a strategy, a person, an outlook on life, whatever.
Rationality is winning. I’m hiring a master rationalist to make me win my court case. What’s not to like?
A rational debate and agreeing on objective truth may be what the arbitrage system wants. But what the individual disputant wants, in the end, in an important enough court case, is to win. If I have to game the system to win, I will. (It doesn’t help when we create legal entities like corporations, which are liable to get into many more trials and also to treat many more trials as all-out war where winning is paramount.)
More or less, because both sides have to agree to the process. Then the market favours those arbiters that manage to maintain a reputation for being unbiased and fair.
This still doesn’t select for rationality precisely. But it degenerates into a different system to that of a lawyer system.
Yes, but if a side can hire a rationalist to argue their case before the judge, then that rationalist will degenerate into a lawyer. (And how could you forbid assistance in arguments, precisely? Offline assistance at least will always be present.)
And since the lawyer-like rationalists can be paid as much as the richest party can afford, while the arbiter’s fees are probably capped (so that anyone can ask for arbitration), the market will select the best performing lawyers and reward them with the greatest fees, and the best rationalists who seek money (which is such a cliched rational thing to do :-) will prefer being lawyers and not judges.
Edit: added: the market will also select the judges who are least swayed by lawyers. It still needs to be shown that the market will have good information as to whether a judge had decided because the real rational evidence leaned one way, or because a smart lawyer had spun it appropriately. It’s not clear to me what this will collapse to, or whether there’s one inevitable outcome at all.
Would a lawyer by any other name still speak bullshit? Yes. But why are we talking about lawyers and judges?
You have explained well the reason that capping is a terrible idea. Now it is time to update the ‘probably capped’ part.
It’s not clear to me either. I also add that I rather doubt that the market, even with full information, would select for the most rational decisionmakers. That’s just not what it wants.
Because I think that in the proposed scenario, where people hire master rationalists to arbitrate disputes, these arbitrators and other rationalists who would be hired by each side independently for advice would start behaving like judges and lawyers do, respectively. (Although case law probably wouldn’t become important.)
I didn’t mean they would be capped by guild rules or something like that, but rather, that the effective market prices would stay low. I’ve no proof of this, economics is not my strong suit, but here are the reasons I think that’s likely to happen:
If arbitration is so expensive that some people can’t afford it (and it needs to be affordable by the poorer party in a conflict), that’s an untapped market someone could profit from. Whenever your argument is with a poor party, you have to have an arbitrator whose fee is at most twice the fee the other party can or is willing to afford, and there’s no effective low limit here. (State-provided judges and loser pays winner’s fees do have something going for them.)
Being a better arbitrator doesn’t require direct investment of money on part of the arbitrator. So a good arbitrator who’s not getting enough work can lower his prices.
Third parties interested in seeing a dispute resolved—if only to achieve peace and unity—might contribute money towards the fee, or send volunteer arbitrators, in exchange for the parties to the dispute agreeing to arbitration. Finally, competition between arbitrators (for money and work) would eventually draw the fees down, assuming a reasonable supply of arbitrators.
What would make people choose an arbitrator that wasn’t the cheapest available? Assuming some kind of minimal standard or accreditation (e.g., LW karma > 1000), an arbitrator is inferior if he cannot properly comprehend your rational argument or might be swayed by your opponent’s master rationalist lawyer. You then have a choice: invest your money in a costlier and fairer arbitrator, or in a better lawyer so you can sway the cheap arbitrator to your own side. I do hope that one dollar buys more unswayingness than swaying-power, but with humans you never know.
How does someone prove she’s a good arbitrator in the first place? Wouldn’t you need another, more senior arbitrator to decide on that and to handle appeals? Either there’s a hierarchy, in which case the lowest ranked arbitrators are cheap (because it’s their own entry price in the business); or there’s a less centralized web of trust, and if it’s fragmented enough the whole idea of universally trusted arbitration is undermined.
I see, so rationalist arbitration of a dispute raises the stakes in status and neutral-party persuasion and that would lead to a market for lawyers? In what sense would this be damaging/ harmful?
Edit: Obviously it would be really bad if lawyers were causing the arbitrators to make bad decisions. But presumably the arbitrators are trained to avoid biasing information, fallacies etc. If you want to persuade and arbitrator you have to argue well—referencing verifiable facts, not inflaming emotions or appealing to fallacies. If all online disagreements did that than the internet would be a much better place! If arbitration leads to better standards all the better. Now, the system might be unfair to those who couldn’t afford a lawyer and so can’t present as much data to further their cause. But 1) presumably the arbitrator does some independent fact checking and 2) there are already huge class barriers in arguments. Arguments an average high school drop outs and an average Phd are already totally one sided. The presence of arbitration wouldn’t change this.
If that’s so then there’s no point to arbitration. He with the best lawyer wins.
Put it this way: take, in the general (average) case, any decision made by an arbitrator. For simplicity, suppose it’s “A is right, B is wrong.” Now suppose party B had employed the services of the very best rationalist ever to live as a lawyer. What is the probability the arbitrator would have given the opposite judgment instead? How high a probability are you willing to accept before giving up on the system? And how high a probability do you estimate, in practice?
The purpose of arbitration isn’t to establish the truth of a question. If that were the case there would be no reason for the arbitrator to even listen to the disagreeing parties. She would be better off just going off and looking for the answer on her own. This would also take much, much longer since she wouldn’t want to leave any information out of the calculation.
Rather, the purpose of arbitration is to facilitate agreement. Not just any agreement but a kind of pseudo- Aumann style agreement between the two parties. The idea is that since people aren’t natural Bayesian calculators and have all kinds of biases and incentives that keep them from agreeing they’ll hire one to do the calculating for them. This means we want the result to be skewed toward the side with better arguments. If the side with weak arguments doesn’t end up closer to the side with strong arguments then we’re doing it wrong. This is true even if one side puts a lot more time or money into their arguments. Otherwise you’d have to conclude that arguing never has a point because the outcomes of arguments are skewed toward those who are the smartest, have done the most research and thought up the best arguments.
If agreement is more important to you than objective truth, than sure, that method will work. I just happen to think a system that optimizes for agreement at the expense of truth and facts tends to lead to a lot of pain in the end. You end up with Jesuits masterfully arguing the number of angels that can dance on the head of a pin.
Uh… like I said:
Edit: If a rationalist is hired to arbitrate a dispute between two Jesuits regarding the number of angels she isn’t going to start complaining that there are no angels. That isn’t what she was hired to do. If the Jesuits want to read some atheistic arguments they can find those on their own. The task of the arbitrator is applying rationalist method to whatever shared premises the disputing parties have. But the system as a whole still tends toward truth because an arbitration between a Jesuit and an atheist will generate a ruling in favor of atheism (assuming the Jesuit believes in God because of evidence and not Kierkegaardian faith or “grace”).
Think we’ve got some fundamental disagreements here about just what it is that rationalists do. You cannot just hire them to argue anything. The ideal rationalist is the one who only ends up arguing true beliefs, and who, when presented with anything else, throws up their hands and says “How am I supposed to make that sound plausible?”
That which can be used to argue for any side is not distinguishing evidence, whether “that” is a strategy, a person, an outlook on life, whatever.
I reply: I’m paying you a lot of money. You’ll find a way.
When I say or hear “rationality”, I think of the tool, not of the noble “ideal rationalist” whose only pursuit is truth, not money or other personal interest.
Rationality is winning. I’m hiring a master rationalist to make me win my court case. What’s not to like?
A rational debate and agreeing on objective truth may be what the arbitrage system wants. But what the individual disputant wants, in the end, in an important enough court case, is to win. If I have to game the system to win, I will. (It doesn’t help when we create legal entities like corporations, which are liable to get into many more trials and also to treat many more trials as all-out war where winning is paramount.)