[EM] Rob Lanphier's hierarchical scheme

[Abd ul-Rahman Lomax]

At 01:42 PM 8/26/2006, Warren Smith wrote:
>Yes, it does bear some resemblance to A.Lomax's proxy ideas.
>I too have devoted some thought to these ideas.

Warren, it should be noted, appears to have independently invented 
Asset Voting, which is effectively a single-user delegable proxy (DP) 
system used as an election method. It is my favorite method for 
selecting a peer assembly (all members have equal voting power) with 
a minimum of wasted votes.


>However, I suspect Lomax's ideas are better and Lanphier's 
>worse.  Or I understand neither.

I've tried, perhaps too much... :-)

I think that Warren may have misjudged Rob Lanphier's ideas. However, 
let's see.

>Specifically, as far as I understand it, with Lomax's proxies, you can
>select anybody on the planet to be your proxy.

Delegable Proxy is a general concept which could be applied in many 
different ways. However, I do assume that a proxy is freely chosen by 
the client, with a minimum of external restrictions.

>With Lanphier's, you are a member of a group of 6.  The membership 
>of that 6-set
>was not selected by you.  It was selected in some undescribed way by 
>the government,
>perhaps randomly.

Yes, this is a difference. When I wrote that there was a resemblance 
between Lanphier's concept and my early work with DP, I'm referring 
to an original idea of meetings of ten people which elected a 
representative who met with nine similar elected representatives, 
etc., with one meeting per day. If every group of ten elected the one 
whom, of their group, they most trusted, one could hold a 
Presidential election in ten days. I did not keep this idea for very 
long. As I just stated it, it is a majoritarian system, filtering out 
minority opinion near the base level, unless somehow the minorities 
can be organized into the lower level meetings.

Further, I had come to recognize a fundamental problem of electoral 
democracy: if there are elections, there are losers, and, if what is 
being elected are representatives, there are then people not 
represented. A system like Asset Voting, used for full non-party 
proportional representation, does get around this. But Asset Voting 
is really a deliberative method, not a standard aggregative voting method.

But, it hit me, what if, instead of a fixed structure with meetings 
of size N, there were a chaotic structure, a fractal, with meetings 
organized around a chosen representative. In other words, instead of 
starting with meetings and electing representatives, start with 
representatives and let those who choose the rep constitute meetings. 
Where a rep is chosen by too many people to hold a coherent meeting, 
the rep could choose members of his constituency to form a high-level 
meeting, and suggest that those who would like to continue to 
function as his constituency, but indirectly, choose one of these 
elevated members.

Thus the proxy network, with manageable meeting size, is built 
simultaneously from the bottom and the top. The individual members 
always retain full choice, but that choice is made in communication 
with the proxies. Thus our standard suggested DP tool: a proxy list, 
which is simply a list of members who have named a proxy, the name of 
the proxy, and the *accceptance* of the proxy. In the systems we 
envision, the acceptance of the proxy is an important element; 
acceptance is an agreement to communicate directly.

At least that is what I hope clients will come to expect. It is this 
communication that is the essence of DP. Voting is practically a 
detail. DP is a device for distribution *deliberation*, controlling 
noise and the work load of each member.

What I see as a serious shortcoming of Mr. Lanphier's proposal is 
that far too many members must put in far too much effort. DP 
distributes the effort. When it is implemented in the Free 
Association context, we have assumed that all members retain the 
right of direct vote, when polls are taken, but -- and this is 
crucial -- not all members necessarily have the right to address a 
high-level meeting.

When I was developing the concepts, the internet was not a factor. 
However, it all gets easier with the internet; a "meeting" can easily 
be, quite simply, a mailing list. Generally, I'd expect, higher-level 
meetings would be open, and especially the top-level meeting of an 
organization would be open, that is, anyone could join, but upon 
joining, subscribers would be on moderation, and it might even be 
that attempts of such a subscriber to send a message to the list 
would result in a response informing the member that such submissions 
are not accepted, that only qualified members of the list may post, 
but that any qualified member may receive a submission and agree to 
post it. "Qualified member" might mean someone holding a certain 
number of proxies, direct or indirect, or otherwise admitted by vote.

And note that any member could vote. Whatever voting rights members 
have are absolute throughout the structure. So the restrictions on 
posting are those which have been accepted by, at least, a majority 
of members, and which continue to enjoy majority support.

And those who don't like this restriction, for example, are quite 
welcome to form their own list. Because of the DP structure, if there 
are enough such members, they can send their own proxy or proxies to 
the top-level list.

It's noise filtering, the very concern that led Lanphier to consider 
the system he described here.

Lanphier does propose an assignment of people to meetings. Essential 
to good communication is rapport. Skilled people can get beyond that, 
but such skill is rare. It is an essential skill to function at a 
high level, but expecting this skill at base-level meetings is 
unrealistic. Thus Lanphier's system, if I've understood it, is 
vulnerable to premature filtering out of minority opinion. DP allows 
minorities -- of any kind -- to aggregate representation to a level 
commensurate with their numbers, bringing such representation to a 
level where the participants are more capable of integrating 
viewpoints to find consensus.

The FA context makes this practically essential, for FAs don't in 
themselves, have any power. A majority vote in an FA means nothing 
unless there is a *true* majority willing to commit to the decision, 
and with the resources to implement it, because FAs leave power 
(votes in governmental elections, money, access to communication) 
firmly in the hands of the members. A mere majority, opposed by a 
determined minority, has little power, because the resources of the 
two factions may cancel each other out, if they both attempt to act 
in the world according to their factional opinion. However, to the 
extent that they can find consensus, their power becomes multiplied, 
the factions don't waste their resources fighting each other, more 
than necessary.

To find consensus is a difficult process. The fixed hierarchy concept 
(concentration by a factor of 10 in my early thought and by 6 in 
Lanphier's proposal) was, in fact, derived from standard meeting 
process, where such breaking down of large groups into small ones 
came to be a common method in the 1970s and 1980s. Small groups can 
much more easily find consensus, full agreement, than large ones. As 
the group size increases, the time necessary to explore and find 
consensus increases exponentially, typically, until even modest 
intentional communities find that meetings become far too onerous a 
burden, except for the truly dedicated.

But consensus *can* often be found, if it is valued and pursued. DP 
is a technique for organizing into manageable groups *without* 
top-level management of the process.

>I may have just misrepresented Lomax or Lanphier or both.   We shall 
>continue on
>blithlely anyway.  In fact, to avoid calling them "Lanphier's" and 
>"Lomax's" schemes
>from now on we shall call them the "Govt's" vs the "People's" 
>selection schemes.

I'll suggest that Lanphier's suggestion involves a set of rules, plus 
the machinery to implement those rules, which could be governmental. 
There are other possibilities. Let's call it a Top-Down Planned 
Hierarchy, using elections and election analysis to assign members to 
meetings within a fixed structure. By contrast, what I envision is, 
quite simply, Delegable Proxy, for the concept of the proxy implies 
free choice by the client, and it likewise implies a consent of the 
proxy to so act.

However, a great deal will depend on the culture within which DP is 
operating. Large numbers of people choosing Clint Eastwood as their 
proxy -- for some reason Mr. Eastwood seems always to be the 
personality who comes to mind in this context, but nothing should be 
construed from that regarding my opinion of Mr. Eastwood -- with, 
perhaps, Mr. Eastwood accepting automatically (or a system which does 
not even bother to require acceptance) -- does not create the 
deliberative network that will make DP truly functional, it merely 
concentrates power, if used in a control structure. In the FA 
context, it wouldn't mean much of anything.

>Now either way, the scheme continues hierarchically, reducing the population
>by a factor of 6 (or whatever) each stage, until at the end we have 
>some manaegably small
>legislature which makes decisions.

This was my original concept, as applied to a single decision, the 
election of the President of the United States. It was thought that, 
at each point, the selection process involved those selecting having 
the opportunity to actually meet and discuss issues with those 
becoming members of the set of those who could be elected, and at 
each point, on a small scale.

This concept, note, does not contemplate direct voting. FA/DP does. 
That is, any member of the FA who wishes may directly vote on any 
issue, even at a top level meeting. But I assume that, in a mature 
organization, the vast majority of votes would be cast through proxies.

Direct voting might be retained in governmental DP. I see no harm in 
it; I think that, in governmental as well as in FA DP, direct voting 
will remain the exception. Direct voting is a way in which a member 
may easily make an exception to trusting the proxy. "I trust John 
very much, but he just has a bug about this topic, so I'll leave him 
as my proxy, but I'll cast my vote when I'm concerned."

And, of course, that member might himself or herself be a proxy 
representing many members.

>Now.   The problem (I think) with Govt-selection, is (a) it is 
>highly manipulable
>by some faction that gets control then amplifies their 
>control.  This is kind of
>like gerrymandering.

Yes, there would be that danger, and it wouldn't be easy to detect. 
DP is practically invulnerable to this. And the FA context, outside 
of control structures like governments, would make attempts to 
manipulate, say by bribing proxies, highly expensive and probably futile.

>   And (b) if the govt-selection is done randomly, then
>there is no gerrymandering, but it may lead to a horribly effective form of
>massive-conformity, wherein ideas that are not "mainstream" are systematically
>reduced each stage, resulting in exponential decrease by the time 
>the top of the
>hierarchy is reached.   It is kind of like the median voter is 
>selected each time to get
>promoted one hierarchy level, and that causes, in a big hierachy, 
>the extremes of the
>idea-spectrum to be totally filtered out extremely effectively.
>(Try a computer sim if you do not believe me.  The high levels of 
>the hierarchy will
>have vastly reduced fraction of extremes.)

My analysis exactly.

>With, however, people-selection, a bunch of Wackos can get together 
>and promote
>one of them to the next level, then they try to do so again next 
>level, etc, thus getting
>Wacko representation even at high levels, and without
>filtering out Wacko ideas.

If the member registration is honest or validated, wacko views can 
only be represented by the number of votes held by wacko voters.

Thinking of the DP process as involving "promotions" is ... off the 
point. I don't envision any fixed (imposed) set of meeting rules; I 
think, from my experience with FAs, that meetings will evolve their 
own rules as needed. Wacko ideas will get due consideration, and not 
much more than that. They will rise to the level where they may be 
exposed and their defects considered in depth, *plus* whatever value 
exists in them -- there is usually some value -- can be extracted and 
become part of an emerging consensus.

>That is good.  On the other hand it might go too far and lead to artificial
>extremeness-amplification as we go up the hierarchy.

I don't think so. I think that the structure, as envisioned, will 
naturally find a close balance. It *is* a noise-filtering structure, 
but with so many points of entry that new ideas will have a fair chance.

As I've often written, if I can't convince my proxy, whom I chose 
because I trust his or her judgement, that my idea is a good one, 
maybe it isn't such a good idea. And if, on top of that, I can't 
convince anyone else (who could take it to another proxy), maybe it 
*really* is a lousy idea.

While there will be bypasses, communication in DP will generally take 
place in small groups, and most essentially between client and direct 
proxy. I'll have my proxy's phone number (and he or she will have mine).

The wacko idea can only advance if, given an opportunity for full 
discussion, it can continue to convince those who learn about it that 
it is worthy of higher consideration.

>Also, note that if each 7-set (here assuming 7X reduction each level)
>is 4 Wackos and 3 normal people,
>then we get vastly more Wackos at high levels in the Hierarchy than if each
>7-set had been 7 Wackos.  This is the effect of "gerrymandering" I 
>was talking about.
>Such gerrymandering could vastly manipulate things if externally 
>applied.  If however
>people-selection is employed, then the factions that "trick" their members
>into arranging just the right gerrymandering (i.e. 4 Wackos
>each stage, and 3 dumb normals who don't understand what is being done to them
>thanks to a con job) will get exponentially tremendously more power at the
>high levels.  So there is a premium on recruiting dumb people.
>So people-selection may also lead to problems.
>But I suspect fewer problems.

Let me put it this way: I certainly hope that the culture would come 
to suspect anyone who comes to you and says, "Let me represent you."

Rather, I'd hope to see people seek out their proxy. Find the person 
involved in the organization whom you trust most and ask him or her 
to be your proxy. That person may well be too busy, but, in a mature 
organization, would have a means of suggesting to you someone who 
might be able to serve you as a proxy. You'd end up with your 
trustworthy person acting on your behalf at a high level, but also 
with someone chosen by that person with whom you can directly communicate.

It is *very* important to realize that I'm proposing DP specifically 
in Free Associations. DP could have many other applications, but 
there are many untried aspects to DP, many possible ways in which 
unexpected consequences to arise. The FA context has a huge built-in 
safeguard: FAs are thoroughly libertarian, non-coercive 
organizations. There can be no tyranny of the majority in an FA. If 
that wacko faction manages to take over the FA, the sane people, who 
already have a proxy network in place, simply reconstitute themselves 
as an independent organization, and can cheerfully ignore the wackiness.

FAs, thus, can fission easily. But, guess what? The structure is the 
outcome. FAs fissioning simply extends the possibilities, for two 
fissioned FAs are still easily connected, all it takes is one proxy 
from one acting in the other. Similarly, separate FAs, without any 
top-level decision, can easily merge *in effect*, simply by extending 
cross-membership, and, again, it only takes a few members to do this. 
This is why the DP structure, combined with the FA concept, could be 
so revolutionary.

You'd think that Libertarians would jump for it. And, indeed, some of 
them might be. FA/DP is rigorously libertarian, that is, the 
organization is utterly non-coercive, it does not even take your dues 
and spend it as decided by a majority, except for minimal operating 
expenses, it certainly does not spend your dues on controversial 
measures. FA/DP organizations are not gatekeepers; they facilitate 
but they cannot control.

(Many organizations might claim this, but, in fact, structurally 
there are a small number of people who control general communication 
between members. The members, if they thought the organization were 
out-of-control or abusive, could not, except with extreme effort and 
struggle, reform themselves as a new organization. But FA/DP members 
could easily do this, and the fact that they could do this is the 
greatest protection against hijacking of an FA/DP organization by a 
special interest.)

(But, of course Libertarians, that is, the Libertarian Party, are 
advocating the application of libertarian principles in government. 
What I say to them is, if those principles will work in government, 
if they are practical there, why not use them in a voluntary, peer 
organization, where, by design, there can be no coercion? FA/DP is 
"libertarian" but an FA/DP organization would have no opinion, as an 
organization, as to whether or not governments should be Libertarian. 
But they *will* be testing the concepts....)