[EM] Trees by Proxy

[Abd ul-Rahman Lomax]

At 04:07 AM 3/28/2007, Dave Ketchum wrote:
>On Tue, 27 Mar 2007 11:32:01 -0400 Abd ul-Rahman Lomax wrote:
>
>>Sorry. He referred to a specific time before the effectiveness of a 
>>revocation of proxy. That establishes a minimum term of office.
>
>Not quite, assuming that is minimum time for a legislator to BE a 
>legislator.  Try:
>      Tomorrow AM proxies get filed for A.
>      Tomorrow 1 PM same givers file for B in place of A.
>      Next day same givers file for C in place of B.
>
>Since proxies take effect at midnight, ten days after filing, A
>never gets to be a legislator via these proxies.

Sorry, *once a legislator has been elected,* and has been sitting for 
some time (at least ten days), then he has a minimum term of office, 
which is ten days from the date of revocation. That's what I meant. 
Yes, it could

By the way, the result that Ketchum gave, that A would never take 
office, was not at all clear from the rules he stated. If the 
revocation isn't effective until 10 days after filing, A would take 
office for one day. That *is* what he wrote. However, generally, 
proxies in corporate practice are not effective until the time they 
are *used* -- i.e., in a vote -- and they may be revoked at any time. 
If the vote has not occurred, and the secretary receives the 
revocation in time to avoid using the proxy, the proxy must not be 
used. This is the corporate practice, and is the generally understood 
meaning of "at any time."


>B holds office for ONE day before being replaced by C (assuming B 
>and C hold no proxies other than these).
>
>>
>>>  A proxy giver can change the proxy AT ANY TIME.
>We are fighting over the meaning of words.  Hopefully the above
>example will clarify what I am proposing.

If words aren't important, Mr. Ketchum, don't fight over them.

I would consider this an amended proposal. It *still* establishes, 
for legislators who have been seated and are serving, a minimum term 
of office of ten days. During that ten days, the legislature could 
authorize World War III. The danger of this only became apparent, 
say, nine days before the vote. Personally, I'd prefer not to have my 
vote cast for the war. Why can't I revoke my proxy immediately?

If this were an elected legislature, with fixed terms (or a 
cumbersome and almost totally unusable recall process), I can 
understand that I can't.

But this is a proxy legislature being described. Now, below, Ketchum 
starts to try to describe why.


>>Here is what Ketchum wrote originally:
>>
>>>Borrow proxies fresh from corporate stockholder usage.  Their
>>>effectiveness starts at midnight 10 days after filing; ends 10
>
>>>days after a replacement is filed or signer dies.
>
>A thought for Abd:  Assume you thought of being a county legislator. 
>Suppose you then got a call:  "Hey, Abd, proxies have been filed so 
>you have a job - be down here in TEN MINUTES to take your seat, ready to vote."

First of all, this is an insane scenario. Remember, I have my own 
proxy filed. Is this delegable proxy? If it is, *my proxy* gets 
whatever votes have just been given to me, until and unless I show up.

Secondly, remember, I propose direct voting. So if I'm interested in 
serving on the legislature -- which in this case means I have floor 
rights -- *I'm already there, voting, routinely, so that people can 
see how I vote. By this time, I'm voting quite a few votes, almost as 
many as it takes to get my seat and be able to debate and enter 
motions. In fact, I'm already doing this, but not directly, I'm going 
to my chosen legislator and he is doing it for me.

I gave an example of corporate proxy rules. In those rules, proxies 
can be revoked at any time. Those rules do not provide, and I have 
not specified, that proxies take effect immediately upon notice. 
Rather the rules take effect the next day if the proxies are filed by 
midnight. So, sorry, not 10 minutes. And one of the benefits of 
proxies is that the task involved can be delegated. Legislators 
already, in some contexts, can cast votes by proxy. So why not continue that?

Ketchum has here given an example of a possible problem from 
immediate effectiveness of proxies. I'm suggesting that it isn't a 
problem at all, not if the rules are appropriate. And I am far less 
concerned about delay in a proxy becoming effective than I am about 
the reverse. If my vote isn't counted, that is a small problem. If my 
vote is counted against my specific wishes, that is a large problem. 
And we were talking about revocation of proxies, not of the 
effectiveness of new ones.

>Presently we elect most legislators in Nov., they take office a few 
>weeks later, and we are stuck with them for a year or more.

Yes. So Ketchum's proxy legislature is apparently an improvement over 
existing practice. But that doesn't say much, when we are comparing 
alternate reforms. IRV is an improvement over Plurality. Does this 
mean that we should all be out there advocating IRV? I don't think 
so. There are better and simpler reforms.

And that is what I'm proposing. Ketchum, so far, has not alleged one 
cogent reason for maintaining proxies beyond the consent of the voter.

If it's an insignificant detail, why does he so stubbornly insist 
upon his revocation period? I think it *is* significant. A proxy 
should be a proxy!

>Switching to election by proxy is an enormous change and deserves 
>giving it a chance without turning upside down things that are not a 
>necessary part of that.

The 10 day delay is an entirely new introduction, arbitrarily set, 
for what reason? Why not allow what Kechtum actually claimed at one 
point, that proxies can be revoked "at any time."

Obviously, there is a necessary time that it takes to register a 
proxy revocation. Whatever time is necessary, that is what it should be!

Similarly, there is a necessary time for proxy assignment. Again, 
whatever time is necessary, that is what it should be!

And this has nothing to do, necessarily, with floor rights. The 
assembly can set whatever latency it wants with regard to those 
rights, it will not harm legislative outcomes. If a legislator is 
participating improperly, the legislature can sanction him or her, it 
has nothing to do with how many proxies are held. Ketchum is simply 
making it *more* complicated than necessary, and for what reason? 
What harm is alleged from allowing a proxy to be revoked at will? 
None that I've seen. The example above was the first stab at it, and 
the quite minor and transient harm is not from revocation, it is from 
assignment without warning.

Basically, by not thinking through what it could mean to have 
delegable proxies, Ketchum is forced to deal with problems that are 
artifacts of the limited way in which he has implemented his proxy legislature.

As to the practicality argument, the charge that by eliminating the 
10 day -- or whatever -- latency, the measure would be less likely to 
be implemented, it is pretty hard to get less likely than impossible. 
And it is impossible right now. Proxy democracy will not happen with 
public institutions until and unless it sees wider use in NGOs, as 
well as in smaller political contexts. But if we want to think of sky 
pie, we might as well imagine a delicious one!

In the NGOs, there is no reason to delay proxy assignments or 
revocations. There would be a proxy list, and a voter can change his 
or her proxy assignment on that list at any time. A matter of 
minutes. When a vote is to be expanded, the list is copied at that 
time and is used. So it is static only during the time it takes to 
analyze a vote *which has already been cast.*

This would mean that a proxy would not necessarily know how many 
votes he or she is casting until afterwards. I see no harm in that at all.

The problems with proxy revocation arose in Ketchum's mind because he 
thinks of floor rights as being assigned based on number of proxies 
held. That's quite a reasonable way to do it, but *this aspect* need 
not change with every proxy change. There is little or no harm in a 
legislator who has lost a few proxies continuing to have floor 
rights, and there is little or no harm in one who has gained some to 
see a delay in gaining floor rights. These times need not be long, 
but, generally, it would be more important to gain a seat quickly 
than to lose one quickly. (This could mean that the legislative size 
could increase temporarily. Again, except under very unusual 
circumstances, little or no harm. And the body itself can address the 
problem if needed, it could, for example, delay new seats a little or 
speed up the loss of a seat. *None of this affects votes.* Just the 
right to address the assembly and enter motions, and any motion worth 
its salt is going to get introduced anyway, the up and coming 
legislator would just go to another and ask that it be introduced. If 
he can't find another, he would have a terrific time finding a 
second, so it's moot!

>There is nothing magic about exactly ten days, and this would 
>deserve review as part of planning to make the switch.

>Also, at least in New York, our Constitution specifies that VOTERS 
>must approve before certain legislative actions can take place.  Our 
>laws restrict others.

For this comment to be at all relevant, Ketchum must be referring to 
our discussions regarding the rules of assemblies. What he's 
mentioned now is not such a constitutional restriction. He's confused 
legislative *action* with legislative *process*. Action is the 
outcome of process. The legislature cannot, for example, assume new 
powers simpy by voting for it. But if the legislature is authorized 
to regulate, let's say, driver's licenses, it can make its decisions 
by whatever process it chooses.

>Legislatures properly enact their own rules for details, but do have 
>to live under the above.

Again, if process is a "detail," then sure. But Ketchum is avoiding 
acknowledging that the constitution does not bind the Legislature as 
to its own rules and process. It binds its powers, as well as its election.

>Abd mentions corporate activity, apparently not noticing that stockholders
>generally meet to do their activity once per year.

Generally. So? The institution of the proxy is quite general. From 
what did Ketchum derive that I hadn't noticed this? I specifically 
mentioned the Annual Meeting, if not here, then I have many times....

Once again, it seems that a detail is being raised as if it were 
significant. First of all, stockholders can call special meetings, 
generally. And, in theory, stockholders could continuously supervise 
the corporation. It isn't done, for practical reasons, but they'd 
have the right to do this if they wanted to. In small corporations, it happens.



>A note about some stockholders:  Last year one corporation's 
>stockholders proposed that majority voting was plenty enough, and 
>agreed to that at the annual meeting.  NY's Business Corporation Law 
>requires approval of 2/3 of outstanding shares for certain 
>activity.  So this year we shall see if that 2/3 will agree to 
>changing 2/3 to majority for the future.
>
>>
>>>  I specify a delay between filing the change and it's taking 
>>> effect because I see the legislature needing to know the effects 
>>> coming up in time to make adjustments - including such as new 
>>> legislators preparing to be legislators before the day they start doing.
>>What adjustments? What "preparation?" Proxy assignments affect the 
>>*votes* of legislators, not how they prepare to vote. Yes, we have 
>>proposed that floor rights depend on the number of proxies held, 
>>but this is an all-or-nothing assignment, and is presumably based 
>>on status at a certain point in time. Changing the *number* of 
>>votes held by a legislator with floor rights need not have any 
>>immediate effect on those rights.
>
>
>He did not get ANY rights until he got to hold enough proxies.

That's a proposed rule. Not a bad one. But it does not follow from 
this rule that he *loses* rights immediately just because he loses a 
few votes. My language was precise, and apparently wasted. "need not 
have any immediate effect on those rights."

>   I do propose getting voting rights on less proxies than for floor 
> rights - which usually would mean paying attention to floor 
> activity in order to vote intelligently before getting floor rights.

Direct voting accomplishes quite the same thing, especially if it is 
delegable proxy.

>>This is quite separate from my proposals to allow direct voting. 
>>Direct voting is a means whereby a voter can, measure-by-measure, 
>>overrule his proxy without revoking the proxy. The desirability of 
>>this is so obvious that I'm not even going to argue it any more....
>
>Another complication that is best left out of the primary proposal 
>though, perhaps, considered as an add on.

It is considered an "add on" because the roots of democracy in direct 
participation have been lost. To be sure, our present democracies 
have largely evolved, not from direct democracy with an increase of 
scale, but from top-down grants of rights. So, from this point of 
view, some special privilege is being granted.... when, in fact, it 
should be a basic right.

As I've mentioned, if it were property rights that were involved, the 
right to be represented by proxy -- chosen at will and revocable at 
will -- would be common law. This is why proxy democracy is 
ubiquitous in business. It was not a new invention, it simply 
followed from the law of property.

"Of the people, by the people, and for the people," but do we own the 
government? Apparently not.