[Election-Methods] Asset Voting can be a proxy system

[Abd ul-Rahman Lomax]

At 05:55 AM 8/15/2007, Jan Kok wrote:
>On 8/14/07, Abd ul-Rahman Lomax <abd at lomaxdesign.com> wrote:
>
> > Proportional representation through elections could seem an
> > attractive solution, except that, in fact, it would only begin to
> > touch the problem. Asset Voting, while it looks like an election, is
> > actually a proxy system.
>
>Very interesting!!! What Abd has observed is that the person you would
>turn over your "assets" to, is probably the same person you would name
>as your proxy.

Asset Voting *is* a proxy system, though it was not conceived as 
such. Rather, when Warren put Asset together, he was, I think, 
assuming that the only "candidates" would be those seriously running 
for the office. This would work, but when Warren first started 
writing to me about Asset, and I pointed out that it was like 
Delegable Proxy, I also noticed that restricting the set of those who 
voters can vote for to be "candidates" as we now see them was 
artificial. Rather, since Asset does not waste votes, it becomes 
entirely possible for *anyone* to "run for office," even if there is 
no hope of winning. And voters could write them in. The only problem 
is technical: obviously it would be impractical to have thousands of 
"candidates" on the ballot, and, in my experience, many write-in 
votes are not counted at present because election officials do not 
bother to exactly identify the person named.

(I saw an election where the write-in pretty clearly got the most 
votes, but was not even named among those getting votes. How did I 
know about the votes? Well, it was my wife who was the last-minute 
candidate, and I know I voted for her, she voted for herself, she 
says, and we knew of at least two neighbors who voted for her. Those 
would not all go wrong! And an official who was present during the 
counting remembered her name coming up.... If we had seriously cared, 
we could have challenged the result (which was that nobody was 
elected, there were three or four single-vote persons, and, since the 
Board of Selectmen had the authority to make interim appointments, 
and they offered to appoint my wife, there it stood .... but I 
learned something about elections.)

So I have suggested that it be required that those willing to serve 
as electors -- technically "candidates" -- must register as such, for 
a very small fee that would cover printing costs for a directory of 
potential electors. Cheaply printed.

In any case, you could really vote for anyone, given the small 
restriction noted (prior consent and registration of the candidate, 
which could be done a matter of days before the election).

This really starts to look like delegable proxy, and when we consider 
how thousands of electors might cooperate to create seats, 
delegability becomes a fairly obvious and important element. Each 
elector may name a proxy. And here, common law would suggest that 
they have a perfect right to do so, unless specifically prohibited by law.

But even the original conception had the candidates acting as proxies 
for the voters. They serve as proxies in creating seats, and they 
serve as proxies if seated. In the latter case, they may have been 
directly chosen, as in standard proxy, or they may have benefited 
from reassigned votes, i.e., as in delegable proxy.

Once you have identified voters, i.e., the electors, those holding 
Assets, it then becomes possible to go ahead and create a peer 
Assembly, but to allow the electors to retain voting rights.

I've written about this often, actually. The crucial realization is 
that the limitations of elected-representative democracy are 
justified by two things: the increasing difficulty of deliberation 
when meetings get large, and the frequent votes involved, which, 
where there is direct democracy, increasingly creates a participation 
bias, where those more highly motivated to participate gain 
inequitable power. If the system were efficient, that power would 
actually be a kind of preference strength measurement, but it is not 
efficient, the difficulty rises exponentially with meeting size.

However, separating the right to vote during deliberative process -- 
which is consent, and, properly, is restricted to Yes/No questions -- 
from the right to do things that take up everyone's time, such as 
introducing motions or speaking to the assembly, solves the problem 
of scale. But to allow simultaneous representation by members with 
seats and direct voting by electors requires that the voters who 
created a given seat be clearly identified, so that there is not double voting.

Asset Voting is Delegable Proxy with a secret ballot front end (which 
has long been a proposal, see the Election Methods wiki), and a peer 
assembly back end that serves as a proportional representation 
council, the only twist where it differs from standard DP is that all 
members routinely exercise the same voting power. Which may be 
counted as one vote each, but it is really Q votes each, Q being the 
quorum for election.

What this does is to keep the electors functional and responsible 
continuously between elections. And the voters know exactly whom they 
voted for, and the electors also know *where* their votes came from, 
i.e., precincts. Thus a much closer relationship between voter and 
elector becomes possible.

Further, I've noted that electors, in assigning votes that they hold, 
may choose to assign them in precinct blocks. (There is a little 
fuzziness to this, but it could be done so that only rarely was there 
any substantial unclarity on the part of the voter as to whom they 
elected; where there was some unclarity, it would simply be that 
their vote helped elect two members.)

This we address one of the huge issues that Richie and CVD were 
concerned about: voter apathy, based on the sense, quite realistic, 
actually, that their vote doesn't count. Their vote would count and 
they would see the effect. Right now, your vote does not count unless 
you vote for a winner. If you didn't, you would have seen the same 
result staying home. And even if you vote for a winner, you 
*probably* would have seen the same result.

> > Because the electors in Asset Voting are public voters, it also
> > becomes possible to have, for them, direct democracy.
>
>Darn! Abd, have you mentioned this before? I think this is a
>significant, new idea. Delegable Proxy superimposed on Asset Voting.

I've been writing about it for years, Jan. Yes, I write about a lot. 
I think about a lot. "Delegable Proxy" does not have to be a formal 
part of the system, it is a way in which the electors could 
efficiently coordinate. In the end, the electors register their votes 
to create seats, those actions have nothing to do with DP, except 
that DP may have informed them as to how to do it effectively.

Then direct voting becomes possible in the assembly. It also becomes 
possible to recall members, the original set of voters is still there 
and available. And we discussed, some time back, on EM or the Range 
list, rules for this. I suggested that there be *immediate* effect of 
the withdrawal of votes by an elector, reducing the voting power of 
the member, but *not* an immediate effect on the member holding a 
seat. Again, what is routinely considered to be the same thing -- a 
seat and voting power -- can actually be separated, not only without 
harm, but preserving the power of direct democracy that is ordinarily 
considered to be lost in representative democracy.

(If a member is guilty of misconduct, it would be the assembly which 
would remove the member from holding a seat, and then the votes of 
the electors can be collected to create a new seat; it's also 
possible to allow, under some conditions, fractional seats. As long 
as there are not too many of them, they do no harm. I would consider 
this to be, properly, part of the assembly Rules, under the 
jurisdiction of the Assembly. But the right of electors to vote 
directly would be a matter of law. And this, then, protects the 
electors from actions by the assembly to restrict their voting 
privileges, since they would get to vote on those actions! Generally, 
an assembly has the right to make its own rules....)

>So, as a voter in, say, a state congressional district,
>
>If I want to be able to vote directly in state assembly votes, I can
>declare myself as a public elector.

You got it. While you are doing it, you might also let your friends 
know you are doing it. If they trust you, they might give you their 
vote as well. And, of course, if they are concerned about issues, 
they then have someone to talk to who can do a little more than just 
talk. And you, of course, will become connected with others higher up 
in the structure; I do not know how many electors there would come to 
be, but it becomes possible that members with seats could actually 
meet with their electors.

Delegable Proxy is a *communications* structure, many miss the point 
thinking of it only as a way of voting.

>If I want to be listed on the ballot, I can gather signatures. The top
>10, say, signature gatherers get listed.

Maybe. It's not really so important any more. I have some preference 
for *no* names being listed on the ballot! It's well known that 
simply being on the ballot gets you some votes. Especially if you are 
listed first. Instead, there would be a method for entering, perhaps, 
a candidate code, taken from the booklet I mentioned. But the 
candidate list could also be on an electronic voting machine.

If the Assembly has 50 seats, we've been assuming, how many 
candidates would be on the ballot? In Asset, generally, there are no 
fixed districts, districts, if any, are created by the electors if 
they transfer votes in precinct blocks. There are very good reasons 
that they would generally want to do so, but some members will have a 
state-wide "district." And districts overlap, so more than one member 
would generally represent a given precinct.

Think thousands of candidates in a state. At least.

>During the negotiation phase of Asset Voting, I would pass my assets
>to some local activist that I trust to represent me. At the same time,
>I would probably designate that person as my proxy.

Actually, I don't think so. You would designate that person your 
proxy for the purpose of negotiations. In the end, you would get a 
recommendation for how to assign your vote. Your authority is not 
given away. You either follow the recommendation, or you don't. How 
you decide is up to you. As an elector, you are a sovereign voter!

Even if you never assign your votes, if direct voting is allowed, you 
still can vote in the assembly, and can use any member of the 
assembly willing to serve you for the purpose of introducing motions 
or presenting arguments. But I'd think it quite advisable to assign 
the votes. You can take them back later if you think you made a mistake....

Officially, there is only your action of assigning your vote to 
create a seat that counts. Until those votes are actually used to 
create a seat, your action is unconditially revocable. Once a seat is 
created with them, there would be some process required, and some 
latency. But it need not be much!

>When votes are taken in the Assembly, my proxy can vote for me, or I
>can vote directly. In either case, our votes override the votes cast
>for us by our representative.

Yes, exactly.

>In order for this to be practical and not slow down proceedings, I
>think we would need voting by Internet, SMS, phone with touch-tone, or
>similar. There are problems with security that need to be worked out.
>There is also the problem of Internet delays, busy signals, etc. that
>could prevent one from voting within the time limit. There may be no
>good solutions to that, but if one is paranoid, one could perhaps vote
>early before voting on a motion officially begins. One should be
>allowed to change one's mind and re-vote.

Sure. That's standard practice, actually. Remote voting is quite 
practical, and, remember, these are all public actions. It is *not* 
secret voting, most of the serious security problems arise when the 
voting is secret, because how can you check? If you are an elector, 
and you vote, and someone who voted for you looks at that vote and 
thinks it wacky, they can call you up! You have no proof that they 
voted for you, but, unless you are getting these calls all day long, 
you will, I presume, explain. Or say, "My Gosh! I didn't cast that 
vote, I was on vacation! My account must have been hacked! And you can fix it."

Electors, I expect, even though they don't have seats, will pay far 
more attention to what is going on in the Assembly. Further, those 
they seat may use them as extended staff.... each member becomes, as 
it were, the top proxy in an organization, with distributed intelligence.

I've been saying that we need to look at the *general* problem, not 
just at the details of election methods. The general problem is the 
central problem of government itself. What is the purpose of government?

As I see it, it is to facilitate communication, cooperation, and 
coordination of members of society toward the common welfare. It is 
what evolution is bringing us to, and early governmental forms were 
successful, to the degree that they were, because they solved this 
problem. But more efficient and effective solutions are certainly possible.

And the good news is that we have most of the elements in place and 
it does not require a massive redesign, just a little informal 
structure and some *small* changes in law. The latter are actually optional!

Notice that DP in the Asset Voting picture, aside from the two-level 
vote assignment process that would be official and according to law, 
is only used informally and voluntarily. Yet it is what would make it 
efficient for large numbers of electors with small numbers of votes 
each to collect their votes in a way that works.

We can, of course, do the same thing completely outside the official 
system. If we knew how to vote, we could essentially do whatever we 
agree upon. So what is stopping us?

The answer is fairly simple, and the solution is likewise, though by 
the nature of the problem and the inertia involved, it will take 
time. It might take surprising little time, though. I think I will 
*probably* see it in my lifetime, and I'm 63. With certain 
contingencies, we could see effective FA/DP by 2008. As you know, 
Jan, there are rumblings and trials going on, all very small-scale.

But very small scale is how I expect it to start. A large-scale 
*conversion* would accelerate the process, but, as you also know, 
that's quite unlikely because of the persistence of inequities 
effect, which affects people who are very well motivated almost as 
much as it affects those who are simply greedy for power.

(If you have a position of authority in, say, a nonprofit 
organization, and you believe that you want what is best for society, 
how will you react to a proposal to subject your power to the 
continuous vote of millions of uninformed people? You would be, shall 
we say, wary. And you might be right to be wary. If it is not done 
right, shifting to direct democracy could be quite a disaster. You 
would have to really understand the whole proposal, the whole FA/DP 
system, before you would get what it actually means. And that isn't 
easy, it seems to take even quite intelligent people a couple of 
years of exposure to the ideas for it to start to get through, there 
are so many assumptions we have about what is possible and impossible.)