6 ms·

> You want third parties, you need ranked choice voting or something similar. Arrow's Impossibility Theorem : "When voters have three or more distinct alternat

by espeed 7y ago

> You want third parties, you need ranked choice voting or something similar. Arrow's Impossibility Theorem : "When voters have three or more distinct alternatives (options), no ranked voting electoral system can convert the ranked preferences of individuals into a community-wide (complete and transitive) ranking while also meeting a specified set of criteria: unrestricted domain, non-dictatorship, Pareto efficiency, and independence of irrelevant alternatives" [1]. [1] https://en.wikipedia.org/wiki/Arrow's_impossibility_theorem https://en.wikipedia.org/wiki/Arrow's_impossibility_theorem

davidw 7y ago
Certainly - that doesn't mean that we can't have a system that's a bit better than we currently have, though. It just means there's not a perfect system.
- nsajko 7y ago
  No, it just means there is no ranked/ordinal voting system that satisfies Arrows criteria. It says nothing about score/cardinal voting. Arrow: "I’m a little inclined to think that score systems where you categorize in maybe three or four classes probably (in spite of what I said about manipulation) is probably the best."
  - ClayShentrup 7y ago
    Nailed it.
azernik 7y ago
Still doesn't disadvantage third parties, and is usually at least fairly close to public desires.
bzbz 7y ago
Look at the theorem you linked. What is its point? Its “dictator” exists in majority vote as well, and is merely anyone who has not determined their vote yet in a close election, e.g. the “swing states.” Correct me if I’m misreading, but how does this favor majority vote over ranked voting?
- espeed 7y ago
  It just says 2 choices work -- like in binary decision tree, you can always choose the best of two -- but with 3 or more choices, you lose this property -- sort-of like in rock-paper-scissors [1], we can all agree that... 1. ROCK is better than scissors 2. SCISSORS is better than paper 3. PAPER is better than rock ?? PAPER is better than rock ?? ?? ROCK is better than rock ?? DOES NOT COMPUTE This in-transitive loop is called Holonomy [2], and it Does Not Compute in the classical sense -- it can put you in a state where it's impossible to win the majority vote -- no one can win the popular vote and so no one agrees who the winner is, ever -- and thus there can be no mandate in an intransitive relation [3]. See Eric Weinstein discuss this concept in the context of Gauge Theory on Joe Rogan's podcast #1203: https://podscribe.app/feeds/http-joeroganexpjoeroganlibsynprocom-rss/episodes/664ea75440dc4ec486a4e72d852a51ad#01:06:51 https://podscribe.app/feeds/http-joeroganexpjoeroganlibsynpr... [1] https://en.wikipedia.org/wiki/Rock–paper–scissors https://en.wikipedia.org/wiki/Rock–paper–scissors [2] https://en.wikipedia.org/wiki/Holonomy https://en.wikipedia.org/wiki/Holonomy [3] https://en.wikipedia.org/wiki/Transitive_relation#Intransitivity https://en.wikipedia.org/wiki/Transitive_relation#Intransiti...
paulddraper 7y ago
Nor can you exactly know an object's momentum and position, nor construct a non-trivial wholly sound type system, nor create an ML algorithm that learns all data sets. There's no free lunches, but there are better and worse lunches.
tunesmith 7y ago
Arrow's is one of those theorems that is useful for its own purposes, but completely abused in a pop culture sense. All it proves is that you can't simultaneously meet those particular criteria, but it doesn't establish that those criteria are required in any sense. The IIA criteria is especially problematic. A lot of vote arguments comes down to a lack of consensus on what we're even trying to solve for. For instance, say that you have a two-candidate race. 50.1% pick A, but their support is lukewarm. 49.9% pick B, but they're extremely passionate. Who should win? We can't really even get a consensus agreement on that question. I think that when it comes to American democracy and one-person-one-vote, the answer should be A, but plenty of others disagree with me. As for the Perot scenario, you'd need both some form of ranked choice voting (not necessarily IRV), and the elimination of the electoral college. Having RCV with an electoral college doesn't help because if you split the EC, the election still goes to the House.
- slowmovintarget 7y ago
  > but they're extremely passionate. Who should win? Being angry, even in a constructive fashion, shouldn't merit extra consideration. A vote is a signal of magnitude 1. Anything else means a person who is angrier gets more consideration. Now we have a market incentivizing anger, which is not how things are supposed to work. My opinion: lack of consensus on that is caused by lack of critical thinking.
  - tunesmith 7y ago
    I agree that moving away from one-person-one-vote creates a market that incentivizes passion, whether angry or not. Even false passion. The people that disagree tend to talk about "greater social utility". I just don't think there's a way to reliably measure degree of passion without falling into self-worth questions. Like, someone wise might want to avoid confrontation, someone idiotic might have a trumped-up sense of self-importance. (No pun intended honestly.) One-person-one-vote is the best way we've come up with to avoid that, treating every voter as having equal worth.
BuckRogers 7y ago
Perfect is the enemy of better.
unethical_ban 7y ago
Agree with all the other comments - no voting system can be "perfect", defined by a sufficient number of sometimes contradictory desirable attributes of the system. Literally anything would be better than FPTP voting for single member districts than what we have now. And, the very act of fixing the system would be a political action, showing that systems can be changed when they are damaged or outdated. Sure, approval voting might be better than IRV. But as I said elsewhere, the first thing to do when disagreeing with particulars of electoral reform is to agree loudly with the general motion - electoral reform.
ClayShentrup 7y ago
1. Arrow's Theorem only applies to ranked methods, not rated methods like Score Voting and Approval Voting. http://scorevoting.net/ArrowThm http://scorevoting.net/ArrowThm 2. Arrow's Theorem really has nothing to do with escaping duopoly. 3. If you want to escape duopoly, you need Score Voting or Approval Voting. https://web.archive.org/web/20180922145553/https://asitoughttobe.com/2010/07/18/score-voting/ https://web.archive.org/web/20180922145553/https://asitought...

by espeed 7y ago

davidw 7y ago

Certainly - that doesn't mean that we can't have a system that's a bit better than we currently have, though. It just means there's not a perfect system.

nsajko 7y ago
No, it just means there is no ranked/ordinal voting system that satisfies Arrows criteria. It says nothing about score/cardinal voting. Arrow: "I’m a little inclined to think that score systems where you categorize in maybe three or four classes probably (in spite of what I said about manipulation) is probably the best."
- ClayShentrup 7y ago
  Nailed it.

azernik 7y ago

Still doesn't disadvantage third parties, and is usually at least fairly close to public desires.

bzbz 7y ago

Look at the theorem you linked. What is its point? Its “dictator” exists in majority vote as well, and is merely anyone who has not determined their vote yet in a close election, e.g. the “swing states.” Correct me if I’m misreading, but how does this favor majority vote over ranked voting?

espeed 7y ago
It just says 2 choices work -- like in binary decision tree, you can always choose the best of two -- but with 3 or more choices, you lose this property -- sort-of like in rock-paper-scissors [1], we can all agree that... 1. ROCK is better than scissors 2. SCISSORS is better than paper 3. PAPER is better than rock ?? PAPER is better than rock ?? ?? ROCK is better than rock ?? DOES NOT COMPUTE This in-transitive loop is called Holonomy [2], and it Does Not Compute in the classical sense -- it can put you in a state where it's impossible to win the majority vote -- no one can win the popular vote and so no one agrees who the winner is, ever -- and thus there can be no mandate in an intransitive relation [3]. See Eric Weinstein discuss this concept in the context of Gauge Theory on Joe Rogan's podcast #1203: https://podscribe.app/feeds/http-joeroganexpjoeroganlibsynprocom-rss/episodes/664ea75440dc4ec486a4e72d852a51ad#01:06:51 https://podscribe.app/feeds/http-joeroganexpjoeroganlibsynpr... [1] https://en.wikipedia.org/wiki/Rock–paper–scissors https://en.wikipedia.org/wiki/Rock–paper–scissors [2] https://en.wikipedia.org/wiki/Holonomy https://en.wikipedia.org/wiki/Holonomy [3] https://en.wikipedia.org/wiki/Transitive_relation#Intransitivity https://en.wikipedia.org/wiki/Transitive_relation#Intransiti...

paulddraper 7y ago

Nor can you exactly know an object's momentum and position, nor construct a non-trivial wholly sound type system, nor create an ML algorithm that learns all data sets. There's no free lunches, but there are better and worse lunches.

tunesmith 7y ago

Arrow's is one of those theorems that is useful for its own purposes, but completely abused in a pop culture sense. All it proves is that you can't simultaneously meet those particular criteria, but it doesn't establish that those criteria are required in any sense. The IIA criteria is especially problematic. A lot of vote arguments comes down to a lack of consensus on what we're even trying to solve for. For instance, say that you have a two-candidate race. 50.1% pick A, but their support is lukewarm. 49.9% pick B, but they're extremely passionate. Who should win? We can't really even get a consensus agreement on that question. I think that when it comes to American democracy and one-person-one-vote, the answer should be A, but plenty of others disagree with me. As for the Perot scenario, you'd need both some form of ranked choice voting (not necessarily IRV), and the elimination of the electoral college. Having RCV with an electoral college doesn't help because if you split the EC, the election still goes to the House.

slowmovintarget 7y ago
> but they're extremely passionate. Who should win? Being angry, even in a constructive fashion, shouldn't merit extra consideration. A vote is a signal of magnitude 1. Anything else means a person who is angrier gets more consideration. Now we have a market incentivizing anger, which is not how things are supposed to work. My opinion: lack of consensus on that is caused by lack of critical thinking.
- tunesmith 7y ago
  I agree that moving away from one-person-one-vote creates a market that incentivizes passion, whether angry or not. Even false passion. The people that disagree tend to talk about "greater social utility". I just don't think there's a way to reliably measure degree of passion without falling into self-worth questions. Like, someone wise might want to avoid confrontation, someone idiotic might have a trumped-up sense of self-importance. (No pun intended honestly.) One-person-one-vote is the best way we've come up with to avoid that, treating every voter as having equal worth.

BuckRogers 7y ago

Perfect is the enemy of better.

unethical_ban 7y ago

Agree with all the other comments - no voting system can be "perfect", defined by a sufficient number of sometimes contradictory desirable attributes of the system. Literally anything would be better than FPTP voting for single member districts than what we have now. And, the very act of fixing the system would be a political action, showing that systems can be changed when they are damaged or outdated. Sure, approval voting might be better than IRV. But as I said elsewhere, the first thing to do when disagreeing with particulars of electoral reform is to agree loudly with the general motion - electoral reform.

ClayShentrup 7y ago

1. Arrow's Theorem only applies to ranked methods, not rated methods like Score Voting and Approval Voting. http://scorevoting.net/ArrowThm http://scorevoting.net/ArrowThm 2. Arrow's Theorem really has nothing to do with escaping duopoly. 3. If you want to escape duopoly, you need Score Voting or Approval Voting. https://web.archive.org/web/20180922145553/https://asitoughttobe.com/2010/07/18/score-voting/ https://web.archive.org/web/20180922145553/https://asitought...