[Election-Methods] delegate cascade

[Abd ul-Rahman Lomax]

At 11:09 AM 7/26/2008, Michael Allan wrote:
>(May I use the list as a scratch pad, to record ideas?  There's a
>theory somewhere in these technical pieces, and maybe it connects with
>the social architecture being mooted in other threads.)
>http://groups.dowire.org/r/topic/4yAf4tghgQ5b53pGbPDRpZ
>http://groups.dowire.org/r/topic/XfrpCKqPLfZVDbcSnRQHY

I could also suggest the list fa-dp at yahoogroups.com, which really 
could use some activity. This would be exactly on point there.


>I've been wondering how the previous argument would apply to an
>election in which the issue was an office (executive, jurist,
>bureaucrat) as opposed to a norm (policy, law, plan).
>
>    (4)
>
>
>      \      \  |  /      /         \  |  /
>       \      \ | /      /           \ | /
>
>    ---  E -->  C  <-- D  ---     ---  X  ---
>
>       /      / |  \     \           / |  \
>      /      /  |   \     \         /  |   \
>
>
>This is the bottom view of an election, showing the two roots (C and
>X).  Higher branches and leaves are largely omitted.
>
>If the issue here is a norm (such as the Public Health Bill) then
>there is no sense in which the election will ever produce a "winner".
>C and X are never going to become traditional legislators.  The
>incoming vote flow will never confer on them a *power* to legislate.
>It will only give them temporary license to *assemble* legislative
>ideas that are flowing toward them.  This is another sense in which
>the formal vote flow must correspond to actual need.  As long as C and
>X are successfully assembling the incoming flow of text, they will
>receive a corresponding flow of votes; otherwise the vote flow will
>shift elsewhere.  In this sense, C and X are just tools of the
>electorate.

In a sense. In another sense, they are leaders. In AA: "Our leaders 
are but trusted servants, they do not govern."


>Moreover they are disposable tools.  Figure 4 is just a snapshot in
>time.  It shows a particular stage of norm construction at which C and
>X happen to be the best assemblers - the best tools for the job.  It
>is unlikely they will remain the best tools at all stages of the
>construction.  This is especially true since the election will never
>terminate.  Although a "final" legislative draft (say C's) might be
>pulled from the root of the tree at some point and promulagated as a
>law, the election will nevertheless continue uninterrupted.  No norm
>is ever final - it always remains open to the possibility of amendment
>or even withdrawal.  So the electorate always has an interest in it.

In Asset Voting, it's possible that not even officer elections have a 
term. Rather, it would be a parliamentary system. The electors hold 
votes that they received in the secret ballot phase, which would 
presumably take place periodically. So that "office" has a term. But, 
note, those are unconditional elections. The voter simply chooses the 
elector. The elector could be the voter himself or herself, under 
some conditions. (Under conditions where security is a severe 
problem, there might be some restrictions, so that one would only 
know one's own vote count if one received more than some bottom 
threshhold of votes, which might be low, like five or so.... but 
these are governmental applications and this kind of thing isn't an 
issue in FAs, I'd say.)

The electoral college that is created could be a standing one, i.e., 
when seats are created by it in the Assembly, the votes could be 
withdrawn. For stability, I'd not have the seat's rights in 
deliberation terminate immediately (absent a vote of the Assembly to 
terminate, presumably for some kind of abuse). Rather, the voting 
power of the seat diminishes, possibly even to zero or whatever votes 
the seat had from the election, and a new member is possibly seated 
(so the assembly size might vary a bit.

Officers, as distinct from seats, then, would serve at the pleasure 
of the Assembly, as with any servant.

But, again, this is a utopian scheme. It's not far away from what we 
have, but what can be implemented immediately is the FA/DP concept, 
outside of government, for many different possible purposes. One 
project was a parent association for a Waldorf School. Again, highly 
influential people, supposedly, were interested. And, again, ... 
nothing happened. People will say, "What a great idea!" and then do 
nothing. At least they will do nothing the first time they hear about 
it, and maybe the second or third. Sooner or later, though, two or 
three people will realize that they can Just Do It. And they will, 
and then others will join.


>Technically speaking, it appears that the formal flow of votes at any
>moment is the *medium* that connects input (i) to output (o):
>
>   (i) actual interest of the electorate in the norm
>
>   (o) formal expression of the norm, as a text
>
>But how does this apply to an election in which the issue is a
>political office?  How would the formal medium of assent relate to the
>exercise of executive power?  Or juridical?  What precisely are the
>inputs and outputs to be mediated?

To me, it is a question of who is the most trusted. Generally, the 
question of the officer is the same question as the question of who 
chooses the officer. It's a question of delegation. One of the 
biggest responsibilities of a major officer is the delegation of 
responsibility; thus, I claim, the most essential skill involved in 
serving as an officer is the same skill involved in choosing who will 
serve as an officer. Theoretically, if we trust that A would serve 
well in the office, we should trust that A, if A is not going to 
serve, would be a reasonable person to make the choice. For a limited 
time, not for all time, like a King. Now, this election.

So, yes, an ad-hoc structure assembled from votes, where every voter 
votes -- say, secretly -- for one person, and then these single 
persons may reassign those votes by choosing a personal 
representative, could work very well to rapidly assemble a winning 
combination of votes. In some cases, those initial choices would 
self-assemble directly to a majority without further ado. In others, 
I assume, there would be horsetrading, logrolling, and all the rest. 
(What is being pictured here is a single-winner election using Asset 
Voting plus a delegable proxy structure to assemble the votes. The DP 
structure is open, all vote flows in it are public, and reassignable 
at any time. Electors are not obligated to name a proxy, but ... tell 
me, if you knew that a "candidate" would not name a proxy, would you 
vote for that candidate? I can say that I wouldn't. Someone who 
trusts only themselves is not the kind of person who should be 
running a government! So the DP system *might* be in place prior to 
the public election. It might even be required by law.... there are 
zillions of possibilities.

I prefer to work it all out in Free Associations first, where it is 
fail-safe, before putting much effort into control structures where 
governmental power is at stake, or substantial collected assets.


>I wonder especially how vote delegation relates to power delegation.
>If my neighbour has leadership qualities, and I vote for her as Mayor
>(my delegate), do I thereby assent to her becoming the Mayor's
>lieutenant in my neighbourhood?  Might the Mayor delegate actual power
>to her, if the need ever arises?

Sure. Why not? With Asset Voting, I see the electoral college 
becoming an extension, from one perspective, of the Assembly. But it 
is also a broad contact network. You want to get a message to a Seat 
in the Assembly? You know who your vote elected. That Seat will 
likely have many electors who gave the Seat the votes to fill the 
quota. You also know, definitely, who you voted for. Likely this is 
someone you know. So you have an identified communication path to the 
Seat, with likely rapport of some kind. Nobody knows for sure who you 
voted for, but you. But that doesn't really matter all that much. The 
electoral college becomes a penumbra around the Assembly, individual 
electors sometimes actually voting where they take an interest in the 
particular vote. Some of these electors will eventually gain seats, 
and will already be familiar with assembly business. Former seat 
holders, perhaps no longer wishing to be so active, may still remain 
as electors. Some of these might even have courtesy rights in the 
Assembly, i.e, the routine right to attend and speak could be 
granted. The structure becomes flexible and, quite likely more stable 
than present electoral structures, where power hinges on what party 
is in the majority.

Have you noticed that political parties no longer are necessary in 
order to assemble a quota of votes? That this produces proportional 
representation without parties or even any specifically defined 
groups that are being represented? (That is, if people vote strictly 
by party, the representation would be strictly by party. If they 
voted strictly by location, the representation would be strictly by 
location. In reality, it will be all of these, some in one case and 
some in another, and it is, on the one hand, largely unpredictable, 
but, on the other hand, probably quite stable. Would the person you 
most trust change frequently? Asset Voting allows you to vote for 
that person, without your vote being wasted. No other system does 
that with any reliability.

Lewis Carroll, when he first described Asset in 1884, was thinking of 
the "common man," who might only know a single candidate to trust. He 
was concerned about exhausted votes in STV, and realized that if the 
candidate could reassign the votes "as if they were his property," 
the exhausted vote problem would be solved. Quite a better solution, 
don't you think, than the Australian one, where, if a voter does not 
rank all the candidates, the ballot is considered spoiled and is 
void? It is, in fact, a blindingly obvious solution, so obvious that 
when I first read about STV -- only a few years ago -- I thought that 
the votes were transferred by the candidates. Isn't that the most 
obvious way to do it?

Apparently not!