Essay talk:Party and Representative Voting

A major defect in this system

You will still get a two party system out of this, gracias a Duvergers law Worzelpete (talk) 02:55, 29 November 2016 (UTC)

Actually, that only holds for FPTP+SMD systems. (That's not to mention the fact that Duverger is frankly contradicted by the UK, India, and Canada.) It's possible to use other systems -- such as ranked choice voting -- to select representatives, who are then weighted by PARV. In fact, it's even possible to have multi-member districts and PARV. The only aspect essential to PARV is the weighting. Fuzzy. Cat. Potato! (talk/stalk) 03:40, 29 November 2016 (UTC)

What's wrong with simple proportional representation? Worzelpete (talk) 22:28, 1 December 2016 (UTC)

(1) Removes local representatives [and thus some level of accountability], (2) hands off too much power to the parties that put together the lists themselves [which makes it easier to kick out legislators who don't vote the party line], and (3) will always result in a (small amount of) wasted votes, as the final legislative seat must be "fit" to the overall vote. (Imagine there's 100 legislative seats; Aardvark Party gets 50.1% vote, Beetle gets 49.9% vote. If both parties get 50 seats, then Beetle is slightly overrepresented and Aardvark has had ~2% of its votes wasted. If Aardvark Party gets 51 seats, then it's no longer proportional.) By directly weighting votes via the popular vote, this is avoided. (In other words: PR is too granular and thus must "ignore" some votes; every single vote counts in PARV.) FᴜᴢᴢʏCᴀᴛPᴏᴛᴀᴛᴏ, Esϙᴜɪʀᴇ (talk/stalk)

You can have Proportional representation and still chose individual candidates. Several places have a system like that. You can split your vote(s) among candidates of several parties and give the rest to the list you like best (in order). So let's say there are 20 seats to fill; Each party nominates 20 (or less) candidates and each voter gets 20 votes. You can either vote "straight party" (in which case you tick the box for the party list) or you can "split the vote" in which case you give a number of votes to individual candidates regardless of party. So say we have three parties. I am a partisan of Party A, but candidate X of party C is a friend of mine or I like his politics or I have a bet running he'll get in. So I vote for candidate X with three votes and give the remaining 17 to Party A by ticking the box for the list. Or if I want some individual person from Party A above the rest of the lot, I can give (up to) three votes to Party A and still tick the list, in which case the first fourteen people on the list get one vote each. It is a bit complicated at first, but it is one of the most democratic systems there is that still ensures you can vote for the guy you like and know. As for local representatives... That's why bicameralism exists. One chamber is elected at large and proportionally while the other chamber is elected by district (you can also have districts with more than one member if you think that works better). Also, a design flaw of your system is that it does not enable representatives to have a secret vote. I know there arguments against a secret vote for representatives, but there are also arguments for it - especially if a vote of no confidence against someone with authoritarian tendencies (or mob connections) is held. Worzelpete (talk) 23:49, 2 December 2016 (UTC)

On (1): I understand that systems like that exist. However, (as far as I understand) they require multi-member districts (or else you could not select multiple candidates), which reduces locality. And I would like to agree with you that the system is, even if not substantially, more complicated. All else equal, least complicated is surely superior?

On (2): That would seem to violate proportional representation, would it not? Why prefer the bicameral system of a PR chamber and a non-PR chamber -- especially when the PR chamber violates localism and the non-PR chamber violates PR? And, as before: All else equal, least complicated is surely superior?

On (3), I believe you forgot to respond.

On (4), I don't believe that's entirely correct. Let us suppose that only the weighted vote total of for and against was reported (eg, 57.87% for, 41.80% against, 0.33% abstaining). How could one identify individual legislators from this data? Yes, one could possibly determine how many members of each party voted for and against (eg, if the number is a multiple of one party's unique & prime weight, then only that party voted; if not, then members of multiple parties voted). But that's hardly useful for an authoritarian in the vast majority of circumstances, isn't it? FU22YC47P07470 (talk/stalk) 00:58, 3 December 2016 (UTC)

I think bicameralism has three main upsides: First, it makes it harder for one (slim) majority to run roughshod over all opposition. Second it enables to include both a system good at representing "the will of the people" (at large) and a system focused on "local concerns" - no political system will ever be perfect in capturing both and both are legitimate to a certain extent. Third, bicameralism enables one body to have a "second look" (That is what the US senate was supposed to be for; a more somber and deliberative body with more elderly and experienced members who don't let their passions get the better of them) - Sometimes a law is just a bad idea and having someone to catch those laws is always a good idea. I dislike the fact that you cannot kick out individual members in a party list based proportional system (hence my proposed fix which already exists in some places) and I do think that there should be some way for Place ABC to have "their representative" in the capital. Unfortunately no place I know of has a bicameral system where one chamber is strictly proportional (and does not employ first past the post) and the other is strictly localized. It would also decrease the democratic deficit of chambers like the US Senate where Wyoming gets the same two votes that California does. Worzelpete (talk) 22:15, 3 December 2016 (UTC)