VoyForums
[ Show ]
Support VoyForums
[ Shrink ]
VoyForums Announcement: Programming and providing support for this service has been a labor of love since 1997. We are one of the few services online who values our users' privacy, and have never sold your information. We have even fought hard to defend your privacy in legal cases; however, we've done it with almost no financial support -- paying out of pocket to continue providing the service. Due to the issues imposed on us by advertisers, we also stopped hosting most ads on the forums many years ago. We hope you appreciate our efforts.

Show your support by donating any amount. (Note: We are still technically a for-profit company, so your contribution is not tax-deductible.) PayPal Acct: Feedback:

Donate to VoyForums (PayPal):

[ Login ] [ Contact Forum Admin ] [ Main index ] [ Post a new message ] [ Search | Check update time | Archives: 1 ]

[ Next Thread | Previous Thread | Next Message | Previous Message ]

Date Posted: Tue 2003-08-19 19:44:54
Author: arendt
Subject: Re: RE: updated version of NK-govt - PART 1
In reply to: arendt (reposting redeye) 's message, "RE: updated version of NK-govt" on Tue 2003-08-19 19:30:00

> Basically, the brain areas for
>vision form a
>feedback hierarchy. The lower parts of the hierarchy do the
>simple analysis -
>this is a line, that is a circle. The next level up creates
>objects, and separates
>figure from ground - it produces what the Gestalt
>psychologists understood
>seventy years ago. Higher than that, and you get object
>tracking. At each
>level, the spatial resolution goes down and the level of
>abstraction goes up.

::You're talking about how grandmother cells don't really exist but
::rather are high-level phenomena created by simpler cells, right?

What you say is true, but the target of this whole section
is to introduce the idea that feedback (down from the higher
levels) is just as important as feedforward (up from the sensory
inputs). This is a new and somewhat controversial theory.
In even more detail, it is proposed that this system performs
Hierarchical Bayesian Inference. (As a mathematics guy, I
figure you know what this is or can find a reference.)

>What is newly understood is that this is not just a
>feedforward hierarchy. It
>is also a feedback hierarchy. For example, have you ever
>looked at a trick
>picture or an optical illusion? You can focus your attention
>and force your
>brain to see the image in a completely different way.
>Basically, the higher
>levels are laying down a template and telling the lower
>levels: connect these
>lines, ignore that feature, this goes in front of that.
>
>When all the levels are in equilibrium, we have a stable
>picture of the world.
>This is the science I want to incorporate.
>
>Since I have been working with powers of two, my first crack
>at this hierarchy
>was a pyramid that reduced four at each level. When I ran the
>numbers, I
>realized that the existing two level hierarchy was actually a
>three level hierarchy,
>if you include the cabinet members. Look at the numbers.

::Right... but the problem is that the more decentralized a system ::becomes, the harder it is to coordinate its parts.
::Take constitutional amendments: they are much harder and take ::much longer to pass than in any other country,
::because the USA is the only constitutional democracy I know of ::that requires the states to ratify amendments.
::
::To some degree, when your resolution becomes low enough and ::your level becomes high enough, you see the
::brain as a single entity. I don't know if neurophysiologists do that; ::I know that the basic social models we use
::treat every human being as a single individual.
::
::Now, you can argue that the model you're proposing doesn't have ::to look from so far because the other people
::are other nations. But the problem here is twofold: one, other ::nations will indeed see the USA as one, rather
::than ten or 40 or 160 or 50, and thus some president will be ::needed, if only for foreign affairs; and two, the
::moment you look at the state as a single organism, you're in ::danger of making the mistake of viewing the
::people as subordinate to it just like brain cells and lobes are ::subordinate to the whole person. In Gödel, Escher,
::Bach, there's a dialog in which an anteater reports that he can ::talk with ant colonies, which agree to his eating
::some of their ants. Since on the ethical level the state should ::never be an ant colony, you're on a dangerous
::ground here.

The metaphor of the "body politic" has been around for 500 years. Notice that in our body, we start to
feel pain at the point when even one cell is being damaged or killed. We are very sensitive. The system
I want will have such sensitivity.

It seems you have a problem with viewing government as many subsytems (too complicated) and also
with viewing it as one organism (potential callous disregard of humans). I think politicians understand
these issues and try to strike a proper balance. We do have a Bill of (personal) Rights to protect individuals.

>This says:
>
> 160/210 = 76% critters at level 0
> 40/210 = 19% critters at level 1
> 10/210 = 4.76% critters at level 2
>
>vs today:
> 435/550 = 79% at L0
> 100/550 = 18%1 at L1
> 15/550 = 2.7% at L2 (15 cabinet secretaries)
>
>Each cabinet dept has 1 L2 + 4 L1 + 16 L0 legislatures = 21 x
>100 = 2100 staff

::100 legislators per legislature is too small, IMO, especially for the ::higher-level legislatures (the L2's and the popular
::L1's). 100 might be reasonable for the L0's, emphasis on might; ::however, the L1's need at least 200 and the L2's
::might even need 400. You may argue that 400 is too high for a ::body to function, but the L2's will only be there to
::get feedforward from the L1's, which I presume will be the ones ::with real power. 200 is, I guess, the highest a
::legislature can have and function; I know Finland's parliament ::has 200 members, and the country has less than
::3 million voters, so if 200 were so dysfunctional the parliament ::would probably be in the low 100's and the v/r
::ratio would still be small. So basically, every cabinet department ::has 400 + 200*4 + 200*16 = 4400 staff. This
::doubling is crucial to ensure a good v/r ratio without completely ::disenfracnchising people by overspecialization.

When you mention the "popular" L1s, you raise an issue that needs more work. I have waved my hands about how
people would automatically distribute themselves appropriately over all legislatures; but it is a very real problem
that people would gravitate to the "sexy" issues like defense or taxes or foreign policy. i defer this point to your
later mention of the Harvard curiculum.

Could you give me a cite for your 200 number? My dabbling in sociology indicates that about 150 people is all
that you can really know beyond superficially unless you are a savant in personal relations, like Clinton seemed
to be.

And, you have inverted the pyramid. L2 has more critters than L1, which has more than L0. Why? The idea is that
you need to spend more compute cycles on the details than on the high level stuff. Also, the high level stuff is
more idiosyncratic and personal, while the lower level stuff is more rote, requiring less skill and more sweat.
What business organization has more chiefs than indians, more executive officers than line workers?

I chose to keep legislature size constant and let the pyramid create the manpower gradient. For each 100 L2
legislators, I have 400 L1 and 1600 L0 legislators.

>So, now what? I have fifteen independent "executive branch"
>presidents.
>Well, I think, at this level, those presidents should play by
>parliamentary
>rules and "form a government" by electing one of their number
>the actual
>and true "president".

::Sounds reasonable for an executive, but not for a legislature. ::While SLs are a good idea for both branches, there
::still needs to be some legislature independent of the executive. ::Take corporations as an analogy: the more
::decentralized the board of directors is, the more power the CEO ::has over them. While the CEO of the state
::needs some powers for his own - e.g. signing treaties and ::declaring a state of emergency - there needs to
::be a single board of directors that oversees him.

I never considered SLs in the executive branch. My fear of homunculi and grandmother cells has focussed me
on creating a structure in which executive power coalesces around a particular task and then reforms around
a new task, as the spotlight of attention moves. Authority is *situational*, not absolute. My thinking has been
that the cabinet secretaries can take the lead in specific policy areas. In an environmental crisis, you want the
environmental secretary in charge. But, what is the mechanism to prevent each secretary claiming each crisis
is in his baliwick?

Here is as good a place as any to try to describe the "executive" branch in this architecture, where the cabinet
officers are elected on their own. As I said, there is a parliamentary style formation of government. In this,
the secretaries who are in the same party as the selected president will obviously follow his lead. If the people
of other parties in the governing coalition disagree, they can withdraw, provoking a vote of confidence. Otherwise,
the president is in charge. But, how does the environmental secretary get to deal with the environmental crisis?
Well, in the end, the president decides. But, when he decides, he gets out of the way. He delegates his authority,
but retains a super-veto. The secretary gets to run the show, but at any point the president can say "no". He
can't run his own counter-policy, but he can say "no". If a secretary gets vetoed enough times, he might lose
the next election for being ineffective. Ditto for a president.

So, just when does the president actually get to do something? And, should we have your homuncular unicameral
legislature? To answer this, I need a long detour. Sorry.



First, if you haven't read it, I highly recommend "The User Illusion" (TUI) by Tor Norretranders . This popularization of
results in brain research covers many important topics: the low bandwidth of consciousness, the half second delay
of consciousness with respect to actual time, the effects of set and setting, and much more.

In the brain, consciousness appears to be in charge, but this is "the user illusion". In fact, the low level processors
have to process sensory data into a "simulation". During that processing, a lot of decisions are made, and consciousness
is presented with a "digest" of the information, with a lot of bias already built in. TUIs shorthand for this is:
sense, simulate, see. Its sort of like what happens when Fox News turns reality into horseshit, and the public
thinks its tasty.

It sounds like you want the unicameral legislature to be the consciousness that thinks its in control. Now, TUI
spends a chapter or two on what control consciousness actually has. In TUIs thinking, it has the power to veto
unconscious actions that are in the pipeline to be executed. These actions will occur in about 0.5 seconds if not
vetoed. So, if you are intellectualizing, this small delay is totally irrelevant, and consciousness really is in
control. But, if you are playing soccer at world-class level, your consciousness is mostly a bystander. But, for
all of that analysis, TUI begs off on identifying a specific brain location for consciousness, which is why I am
resistant to the homuncular legislature.



The question of timescale in government has bothered me for a while. Why should the press jump all over the
politicians for immediate (and I mean, right now, this minute) answers, when it takes hours or days to deliberate
an appropriate response, if said response is not already rehearsed, like a military or police or disaster relief
response. The press is going to get BS for an answer. No surprise. You can't turn on the light while the film is
developing. It is the same issue as trying to think about how you are playing the piano - you will mess up.

At the other end of the spectrum, how the hell can you take six months or more to run legislation through Congress?
It begs for interference, log rolling, poison-pill amendments, pork, agenda manipulation, etc.

I would like to mandate some standard scopes and timescales for legislation and press response. That way, the
press will back off, but the politicians will be on a tougher hook. They had 24 hours or 72 hours to get out an
answer, and all they delivered was smoke. For legislation, they had two weeks to deliver some small detail of
some larger bill, on its own to be reconciled later. Why couldn't they decide on that detail? ETC. Such guidelines
will help manage the interchange of information among the various levels of legislatures by presenting reliable
timings that allow efficient scheduling.




To finally get to the president's powers, first, he has the veto of all cabinet level decisions.
This gives him control over all policies in exactly the manner that consciousness controls
unconsciousness in the brain. It is appropriate.

As the leader of the governing coalition, he has the power, through the cabinet secretaries
to introduce legislation into any L2 legislature.



If there is any homuncular legislature in my design, it has to be the budget legislature. The budget is a holistic
thing. You have to both set its absolute size and divide it among departments. This kind of decision is fundamentally
executive; and yet, the power of the purse is the major power of the legislature.

Since the budget is the starting point of the "output" side of government, it should begin at the
higher levels, just as motor control begins at the higher levels of the brain (except for hard-wired reflex
actions or stuff so fast it has to be learned, unconscious reflex).

So, the President formulates a budget and delivers it to the L2 budget committee. The L2 budget committee
parcels out each departments budget to an L1 budget subcommittee. At this point, those L1 subcommittees
communicate with their L2 legislatures to see if it is acceptable. If not, the L2 departmental legislature begins
a process which either re-writes its budget to conform to the L1 budget legislature, or it instructs its L1
legislature to object and to force the issue back up to the L2 budget legislature.

Meanwhile, those departmental budgets approved at the L1 level go down to the L0 level for detailed
analysis. The L0 people may discover discrepancies, large or small, which bubble back up to L1 for
small fixes, or all the way up to L2 for large fixes.

This is not far from the budget process today, with the committees playing the L1 role and the subcommittees
playing the L2 role.

I would like to introduce timings. I think today, that there are very loose rules about when bills are reported
out. And, there is a lot of procedural stuff about 1st, 2nd, and 3rd readings of a bill. I need to learn more about
details, but my goal is to prevent "bundling", "hostage taking", and "poison-pills".

Is there an executive role in the budget? No. No need for one. Its the same as the president today, he sends emissaries
to Capitol Hill to lobby for his proposals. In my design, he controls the cabinet secretaries who lobby for him.



So, does the president control the executive branch agencies directly? Well, no. He can veto the cabinet secretary,
but that secretary implements the policy his pyramid of legislatures has produced. Again, is this any different than
the current situation, other than that the president can fire any secretary to get his way? Recall Nixon and the
Saturday Night Massacre, where he fired the Atty. General, the Assistant Atty. General, etc. until he got someone
who carry out his order. (The someone was that shit Bork.) In my version, that can't happen. If the Greens elect
an Environment Secretary, the president has to deal with that.

What then is the real role of a president?

I think, as I said before, that he controls the spotlight of attention for legislation.
More on this below in discussion of foreign relations and military.

END OF PART 1 OF RESPONSE (too big to post as one)

[ Next Thread | Previous Thread | Next Message | Previous Message ]

Replies:

  • Re: RE: updated version of NK-govt - PART 2 -- arendt, Tue 2003-08-19 20:01:24
  • Re: RE: updated version of NK-govt - PART 2 -- arendt (posting for redeye), Tue 2003-08-19 20:02:46
    Post a message:
    This forum requires an account to post.
    [ Create Account ]
    [ Login ]
    [ Contact Forum Admin ]

    Forum timezone: GMT-8
    VF Version: 3.00b, ConfDB:
    Before posting please read our privacy policy.
    VoyForums(tm) is a Free Service from Voyager Info-Systems.
    Copyright © 1998-2019 Voyager Info-Systems. All Rights Reserved.