[EM] APR (9): Steve?s 9th dialogue with Juho (Steve)

[Juho Laatu]

> On 16 Dec 2014, at 15:16, steve bosworth <stevebosworth at hotmail.com> wrote:

> S:  I agree with your principles:  “large enough … not too many groups”.  However your 10,000 is needlessly arbitrary and might not deliver what these principles require.  Please consider APR’s formula for determining its “associations”.  This is fully explained in my article on pages 6, Endnote 5, and Flow Chart 2.  What do you think?

The 10'000 is of course arbitrary since it is not designed for any particular society. Each society should pick a number that suits its needs.

In APR the difference that was discussed is having the additional primary.

> J:  In elections with numerous candidates I would favour "blank" ballots that do not list the candidate names at all (one could write candidate numbers in the ballot instead). …
> 
> S:  Why prefer “blank ballots” in preference to APR’s sample ballot that would make it easier for each citizen to use?

I like "blank" style ballots in generall (not in comparison to APR) in elections with high number of candidates e.g. because that makes ballot printing easy, may make also voting easy (depends on the method), and is a neutral approach (no order of parties and candidates). Often there is no need for the amount of paper that tens or hundreds of candidates would require.

> J:  The grouping would exist only to nominate a list of candidates in the election (maybe becomes a proper party if it gets some seats). The newly formed groupings would be treated in the election exactly the same way as the old well established parties.
> 
> 
> S:  Please define what you mean by a “proper party”. 

"Proper party" means here a party that is well established and that may have additional responsibilities and rights when compared to newly formed lightweight associations that may or may net get seats in the election. It is possible that a "proper party" need not collect supporter names to get the right to nominate candidates (since it is already known to be big enough to have that right).

> J:   No specific method in my mind. But for example a basic closed list, with no cutoffs, and proportionality counted at top level, and that allows also small groupings to participate in the election (e.g. after collecting supporter names), would be proportional and not in any arbitrary way, and would allow also small groupings to win seats.
> >
> 
> S:  Please explain this much more completely.

http://en.wikipedia.org/wiki/Closed_list

> Below you say you want to elect the reps from a number of multi-winner districts.  Is this by a “closed  list” party-list system?

I don't think I said so. The closed list method was just an example method as an answer to your question.

> By “no cutoffs” do you mean that any party that receives a percentage of the votes equal to one nth  of n candidate that will be elected to represent a given district will send one rep to that assembly, two nth two reps, etc.?

Roughly so. Also district division may also introduce cutoff-like behaviour (= discrimination of the smallest groupings).

> S:  The primary only determines how many reps will be elected from each association.  Only all the citizen’s votes in the country during the general election will determine the weight that each reps’ vote will have in the assembly.  Thus, neither the Primary nor the resulting associations could prevent, for example, any rep from receiving many fewer than the average number of weighted votes, i.e. when that is how citizens have voted in the general election.  Does this solve the relevant problem in your mind? 

Since the oucome of the primary has a meaningful impact on the outcome of the whole election, this does not solve my problem. Vote buyers, coercers etc. may still have an interest to influence the outcome of the primary.

> > > J: No, APR makes it possible to the voters to prove how they voted. They could prove that e.g. by showing their association specific ballot paper to other people.
> 
> 
> S:  No, no-one can prove how they voted during the general election as I explained in the 4th paragraph below:

I referred to the primary.

> J: To my understanding voting in the primary is not secret.
> 
>  
> S: Correct, did I ever claim otherwise?  Voting in the primary is not secret but any such vote cannot bind how any citizen will vote during the general election.  At the same time, any citizen who might be uncomfortable by seemingly revealing some preference in the primary can simply confine his voting to the general election.  Does this solve the problem you had in mind?

No, the primary is not secret.

> J:  Most methods that are in use today seem simpler than APR.
> 
> 
> S:  You seem to have missed the above reminder:  “APR allows a citizen simply to vote for one candidate if they see its other options as too complex.”  Does not this make APR as simple as other system for all citizens who want it to be this simple?  If you think not, please explain.

For example methods that allow only voting of one candidate are easier to understand.

> S:  Unlike APR, the systems you seem to prefer violate democracy’s principle of “equal respect” by giving some citizens the option of helping to elect a candidate they trust while others are not given this option within their residential district.

I'm not proposing any alternative systems, just discussing the properties of APR, and if those properties could be improved.

Geographic proportionality is usually implemented so that it forces voters to vote candidates of their own district.

> S: You have not yet offered your own justification for insisting on what you call “geographical proportionality”, especially when it denies some citizens the option of helping to elect a rep they positively favour.

I don't insist on geographic proportionality. I did describe why geographic proportionality is often used.

> J: I don't have any "preferred system". Many systems do guarantee approximate proportional representation.  APR introduces some additional exact mathematical nature to this. I don't think most voters require exact mathematics but an overall understanding that the system is fair.
> 
> 
> S:  In comparison to APR, these other systems are not “overall … fair”.  How many citizens have you asked whether they would prefer a system that would guarantee that their vote will continue to count?  How many have answered, “NO”?

I guess you have already noticed that I'm not proposing any systems as alternatives to APR.

> J: The idea that voters may decide which associations are allowed to participate in the actual election may be popular. I'm however not convinced that it would also technically make some stronger bonds between the voters and the parties (i.e. other reasons than the existence of an additional round). …
> 
>  
> S:  If you say that this “participation … may be popular”, does not this suggest to you, therefore, that it should be tried?

I have said that in general I would like to see some weighted vote systems tested somewhre.

> J:  … The added complexity may also make some voters less interested.
> 
>  
> S:  Why would even those who do not want to participate in the primary themselves want to deny it to others?

The society might prefer simple methods (to offer simple and easy to understand methods to the voters).

> S:  Again, as explained earlier, APR offers “simple and effortless voting” for every citizen who may want it.

And a very difficult to understand system to all.

> Currently, at least, I am not satisfied simply to leave these decisions to others, i.e. to “society”.

I see us as experts, discussing the properties of APR. The decisions will be made by the societies (whose members also we might be) that may take different methods into use. Maybe we can help them in their decision making by providing good methods, and good explanations on how they work.

Juho


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20141216/a529a255/attachment-0001.htm>