[EM] language/framing quibble

[Kristofer Munsterhjelm]

Fred Gohlke wrote:
> Good Afternoon, Kristofer
> 
> re: "The rationale (for protecting an opinion not held by the
>      majority of the electorate) is that it enables compromise."
> 
> I submit that the essence of the Practical Democracy concept is 
> compromise.  Three people, exchanging views on a variety of public 
> issues and choosing the spokesperson who most closely represents the 
> attitudes of the group, will work out to the best solution possible.  In 
> most cases, 'n' won't win, and neither will 'y'.  Instead, superior 
> alternatives will be found.

There are two ways to regard the results of an election method. The 
first is how accurate the representatives display the properties or 
opinions of the people, and the second is how good a job the 
representatives do.

You appear to be focusing on the latter, while I've been focusing on the 
former. My reason for doing so is that because politics is a process, if 
a method has biases, then these may amplify in ways that can cause odd 
results over time. A good example of that is Duverger's law for 
plurality; the bias is vote-splitting, and it causes two party 
domination. Even if the country was nearly filled with paragons of 
virtue, partisanship would happen nevertheless, simply because of the 
long term feedback of the system. Therefore I try to find situations 
where your method loses information, and see whether they could cause 
long-term problems.

 From a representativeness point of view, there are two potential 
sources of trouble, that have been referred to earlier.
The first is that while compromise is good, it might be that the 
compromise isn't transferrable. That covers the "primary", "successive 
shaving-off", and "differing compromise" objections. Mainly, the 
intuitive idea is that if a group of three (or some other number) agrees 
upon a compromise, then by necessity, that compromise can't contain all 
possible opinions of the group. If this happens in many places, the 
compromise of compromises will differ from the compromise of all, had 
they been somehow able to elect directly, simply because the minorities 
add up to a sizable fraction that could have influenced matters if they 
had been more lucky.
The second is that if there are multiple opinions (not just yes and no), 
then the triad can't hold an opinion that's held strongly by less than a 
third of the people. There's simply not enough room, unless the 
compromise candidates hold this opinion weakly, so that at least some 
aspect is retained. Raph referred to this when he said that 1/7 is a 
pretty high bar, and that was with a council size of 7; certainly, most 
parties (assume they're opinion clusters, I'm not talking about the 
negative effects of partisanship here) in PR democracies are elected by 
less than 1/3 of the people, which would be the worst case barrier (for 
this argument) with a council size of 3. This effect could cause 
majorities or compromise candidates to become more extreme as long as 
they know they're part of a majority, so that they can overrule the 
minority: the internal strategy would then be "how much of the others' 
views can I squeeze out without alienating the other two in my triad".

This does, of course, not discount the importance of having good 
politicians. If the legislature turns into a battlefield or get twisted 
by corruption, it doesn't matter how representative the politicians are. 
The degree to which representation matters differ with the application 
of the method: if the legislature can recall the executive, having a 
majoritarian executive is of less problem than having an 
unrepresentative legislature. An extreme end point in this case would be 
judges: you wouldn't really care about whether judges are liberal or 
conservative if they're good enough to maintain their neutrality 
whatever their political views.
Thus, if your method is used for executive or judicial positions, the 
potential adverse situations may not occur because disproportionality 
doesn't add any significant measure of chaos.

> re: "The compromise on a national level might be different from
>      the compromise on a local level, meaning that the entire
>      spectrum should be preserved to the extent that it is
>      possible."
> 
> That is an implementation concern.  The original draft of the concept 
> was done for the State of New Jersey (US) using the 2004 voting-eligible 
> population of 5,637,378 people.  It anticipated that, at certain levels, 
> those not selected to advance to state or national offices would 
> constitute a parallel process for local and county offices.  The issue 
> was not seeking ideological representation but selection of the best, 
> brightest and most trustworthy people for public office.

Ah, I meant "local" as in lower level triads and "national" as in higher 
level triads, from the council democracy idea of having councils 
correspond to geographic locations. I thought this would be the case for 
the randomization algorithm for your triads, or at least for groups of 
triads, but that need not be the case, so I might have confused you there.

> re: "Otherwise, you can get effects similar to primaries where
>      the primary electors elect those that are a compromise
>      within their own ranks, and then the general election turns
>      out to have candidates that are more extremely placed than
>      the voters."
> 
> I don't believe the methods are comparable in any way.  Parties control 
> the selection of candidates for public office.  They are chosen for 
> their bias and their lack of integrity, not for their ability to serve 
> the public interest.  That creates a situation in which corruption is 
> inevitable.

Assume for the sake of the example that parties are honest. They aren't, 
but that's not what the primary effect I mentioned is about. To simplify 
further, say that there's only one political axis (left-wing to 
right-wing). The two parties occupy positions some distance away from 
center, say at 0.20 and 0.75. Then the compromise of the left-wing party 
will be at 0.20, and that of the right-wing party will be at 0.75; these 
are both more extremely placed than the true compromise at 0.5.

> re: ("A wise electorate will realize their best interests are
>       served by electing people with the wit and wisdom to listen
>       to, consider, and, when appropriate, accept fresh points of
>       view.")
> 
>      "Yes, but to do so, they need the big picture."
> 
> Anyone who achieves selection to, for example, our Congress, is 
> guaranteed, not only to have 'the big picture' but to be able to 
> enunciate it in so compelling a manner that even those who seek the same 
> seat are convinced.  If the selected person is deficient in any way, the 
> others will be sure the weakness is made clear before the choice is made.

I'm getting ahead of myself, but one thing I observed from the 
simulations is that the higher level councils often have not just a 
majority, but consensus. Do you think this could lead the councilmembers 
to consider their own positions to be held by more than is actually the 
case? Since the positions are central compromises (assuming, again for 
the sake of the example, negligible disproportionality effects), it 
would be better than for the system to produce a false consensus of a 
minority view, but it could still be distorting. I'm not sure, so I ask.

> re: "What I meant is that even if you could magic up an election
>      method, there will be som reduction of minority opinion.
>      There simply isn't enough room in a 200-seat legislature (to
>      use example numbers) to perfectly represent opinions that
>      are held by less than a 200th of the people ..."
> 
> That is a fact.  We must keep in mind that we elect the 200 people in 
> that legislature because we want them to make the best decisions for the 
> entire electorate regarding issues that arise during their term.  If an 
> issue arises that affects a minority we want them to consider the matter 
> carefully and arrive at the best resolution possible for all of us ... 
> regardless of anyone's ideology.
> 
> 
> 
> re: "... if the method tries, then some opinion held by a greater
>      share will suffer.  On this I think we agree ..."
> 
> We do.

> re: "The majority /of that council/. That need not be the
>      majority of the people at large. If the real majority is
>      thinly spread, it can get successively shaved off until
>      nothing remains."
> 
> That may be.  I haven't examined the point carefully because my focus is 
> on electing better decision makers.  There is no doubt that there will 
> be issues that are not clear-cut.  To resolve them, we need to change 
> the way we maintain our laws.  I could describe one way of doing so but 
> would rather not digress unless you consider it important.

That reduction (that may happen) seems to be a result of the election 
method. A direct PR election, for instance, would not have this problem 
(though it may have others). Therefore, you could either readjust the 
election method, or compensate by some other means; I suppose it's the 
latter that you're referring to when you say that we'd need to change 
the way we maintain our laws.

> re: "... if a candidate says "Okay, I'll try to compromise" and
>      gets the votes of the rest of the triad, and then escalate,
>      then what's keeping the candidate from turning on his
>      promise?  Presumably you'd expect most people to be honest,
>      but there's still an uncertainty, and that uncertainty
>      appears at every level."
> 
> That is, and will always be, a risk in representative government. As I 
> said in the outline:
> 
>   "This is a distillation process, biased in favor of the most
>    upright and capable of our citizens.  It cannot guarantee that
>    unprincipled individuals will never be selected ... such a
>    goal would be unrealistic ... but it does insure that they are
>    the exception rather than the rule."

In an ordinary election, that risk occurs once. Among the triads, it 
occurs once at each level. Unless the triad negotiation phase is very 
expensive, I think that a body elected by its method should have a 
recall mechanism to balance that weakness.

Some council democratic ideas (with councils tied to geographical 
coverage) have recalls at each level. I think that would be too fragile, 
though, unless (and perhaps even if) the recall is only accepted if it 
also contains an alternative.

> re: "Majority flip frac is the fraction of the times that the
>      last triad had a majority for one position where that
>      position was in a minority among the people."
> 
> Wahoooo!  Ya got me!
> 
> Awww, I'm joking.
> 
> I confess that I don't understand the math involved but I think I've got 
> a slight glimmer of the picture.  Let me also say this.  I REALLY wish I 
> could work with math like that.  What little I can see in what you've 
> done is exciting.
> 
> I guess I'd better check what I think I see:
> 
> Are you saying that when 60% of the total population holds a given 
> opinion, the chances are that 99.9983% of the final triads will hold 
> that opinion?  I'm not clear on the number of levels this entails, but I 
> don't greatly care because I assume it's a reasonable number.

I'm saying that when 60% of the population holds a given opinion, on 
average, 99.9983% of the final triad members hold that opinion. "cp avg" 
is the average support in the final triad. I'm also saying that if 52.5% 
of the population holds a given opinion, 4.1% of the time, a majority of 
the final triad members hold the opposing opinion (a "majority flip"). 
That's not so bad, and it actually surprised me; I thought the 
quantization error would be much greater.

> I wonder if it would be OK for me to mention the danger in trusting 
> simulations when dealing with humans.  The incredible financial crisis 
> that threatens us, right now, is as good an example of the danger as I 
> can think of.

That is OK, and I know there is a danger in taking a model too far. You 
can end up with ideologies like "the market can do no wrong, because a 
perfectly competitive model in an environment with no externality is 
Pareto efficient"... where the real world has imperfect competition 
facing externalities. Still, a simulation is better than nothing at all. 
The hard part is knowing how far you can generalize the results.

> I think, in trying to visualize the system, it's better to think about 
> having three people you know meet, charged with the responsibility for 
> resolving some issue.  In municipal terms, for example, for them to 
> decide whether a stop sign, a stop light or no traffic control should be 
> placed at the intersection of Maple and Vine.  Even though there's an 
> excellent chance that they will have divergent views on the matter, it 
> is hard to imagine them not making a sincere effort to reach the best 
> decision possible for the community.

That's true. The first representativeness issue regards successive 
triads, and so wouldn't fit with that visualization. The second 
representativeness issue would be similar to there being four traffic 
solutions; one has to remain unrepresented unless some can argue for 
more than one solution.

> re: "Let's then hope that the members can combine the opinions
>      better than the limitations of the system squeeze out
>      minority opinions that may be influential."
> 
> Even more (in my opinion), let us hope the process gives us people of 
> probity and intellect; the kind of people we can rely on to consider all 
> opinions objectively.

Both are important, yes.