Philosophy Dunbar's State

JagerIV

Well-known member
Constitution Of Dunbar
Purpose of Project

My goal is to lay out and explore the constitution and functioning of a Global State set up to run on human scales. This is the source of the title Dunbar, in reference to Dunbar's number. I will lay out some principles and goals guiding this project, and then a Constitution laying out an attempt to implement those principles.

Both the principles and Constitution are intended to be updated as discussion hopefully progresses. May this be an interesting learning experience.

Principles

1) Personal: Where possible, it is preferable for something to be done personally by a sovereign individual, rather than an impersonal bureaucrat.

2) Subsidiarity: Ideally, an issue is handled as locally as possible.

3) By Dunbar's numbers and other supporting empirical work I've seen, 5 can be close, 15 can be friends, 50 acquaintances, and you can know 150. Thus, where close cooperation is necessary groupings of 5-15 should be maintained where possible, and a group should not go above 50-150 individuals is it can be helped. 500 may be a max for a single grouping if no personal relations can be maintained.

4) Promote Humanity: The States goal is to promote a fundamentally physical human society, and thus should er on a human rather than in human organization if at all possible. Human Scale, human needs.

5) Labor Theory of Property: builds on Lockian philosophy of property where ownership is tied in some way to mixing your labor with the property.

Government NameLandPeopleTax Owed Income/LandBoarder Walk TimeMax Boarder (km)Elector Per RepresentativeTotal Pop Per ElectorMin Number of Reps People's AssemblyMin PopMax Number of Preps Per AssemblyMax Population
1ManorFamily7%/7%1 Hour57713914503,150
2CountyClan6%/6%1 Day501391181,63837534,125
3RegionTribe5%/5%1 Week200181,6382236,036300491,400
4Statenation4%/4%50 Days1,0002236,03625900,9002258.1 million
5CommonwealthEthnic3%/3%1 Year5,00025900,9002724 million150135 million
6LandRace2%/2%7 Year25,0002724 million28681 million751.8 billion
7FederationHumanity1%/1%49 Years100,00028681 millionAnyAnyAnyAny
Constitution of Dunbar
Preamble

We the people, in order to secure for ourselves, our children, and the human race liberty and family, property and community, elect to form and pledge our loyalty, property, and lives to the formation, growth, and life of the Dunbar Imperium, and the subsidiary governments that form it.

The Individual

The individual is endowed with rights and obligations, depending upon their status as a citizen or subject.

A citizen is a human of age and mental competency who has recited the oaths of citizenships.

Rights:

a) A citizen has a right to own land.

b)

Obligations:

a) A citizen is responsible to pay a 28% tax on all income, paid to the level of government the income is earned on.

b) A citizen is responsible to pay a 28% tax on all land owned, based upon the value of the land.

The Family

Governments


Manor: A manor is the lowest level of Government. Its boarder must be walkable within 1 hour for the initial drawing of the boarder, to be no more than 5 km long. To the degree the boarder is not walkable, the equivalent of the boarder length must be walked. Human powered boats may count for walking when practible.

All who wish to run for a Manorial government must also complete a walk of the boarder within 1 hour. Such a walk is qualifying for 7 years.


To be continued. . .
 
Last edited:

JagerIV

Well-known member
...Context, please?

... Shoot. I meant to get the whole thing posted over the weekend, but finking with the spreadsheet took all the time. Was hoping to have time tonight to finish it, didn't really mean to post it yet.

Shoot.

The short of it is I was playing around with the idea of a State organized out on principles of Dunbar's number:

https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Fa3f4915c-8a8b-40cf-8300-7b5af350b5c0_620x285.png


For representatives, each layer was set so the number of representatives on the next higher level were within a Dunbar's number of the level bellow, allowing a chain of personal connections to be maintained:

5 family per head of household, about 10 heads of house per local representative, 50 local reps per state rep (the biggest jump) and then more or less a 5-1 ratio for the rest, with some fudging to get cleaner numbers. Locality maybe should be a little bigger, and a State a little bit smaller. Such number fudging took much of my time.

Such a system however does produce a very tall government: the costs vs benefits of such a tall government were one of the things I though would be interesting to think about.

The other way Dunbars number is applied is to limit the size of any assembly, and thus government, to 150, since by Dunbar's that's about the limit of what you can get as a cohesive group. So, you have a lot of assemblies, but each one is kept small enough that every representative can potentially matter, since at say a 100 every vote is 1%, preserving agency of the assemblymen.

So, the federal government in such a scheme instead of 535 elected representatives in two houses, you would instead have 1,716 spread across 1 Federal assembly, 3-4 Union assemblies, and 10-15 commonwealth assemblies.

As part of a measure to try and limit the concentration of power, I was experimenting with the idea of a government employee cap per level. This is built on Dunbars too.

1 exec has 5 direct underlings, who can have 15 in their team, who can have 50 mid managers, who each can operate a company of up to 150, which assumedly have their own internal hierarchy. So, the largest organization a single executive can command is 5x15x50x150=562,500 line personnel, plus the management of about 1,000 individuals.

Each Government then has 3 executives, because that's another idea I was playing around with, for the 1.6 million employees per government cap limit reached at the commonwealth level.
 

JagerIV

Well-known member
One thing I've thought over is how much voting might need to go on in this system: if you have the full 10 layers, that's 10 representatives, and if you vote for the executive too, that's 10-30 executives to vote for too. So, 20-40 people to vote for. Very hard to get enough info to make informed decisions across that entire stack.

The simple fix would be for the vote to just go to the next level up: the people at large vote for their clan rep, likely the head of the household. So, 5 choose 1 who will represent them.

Then, from among their number, say 10 Clan representatives elect one of their own as locality representive. And onwards up the chain.

This is very different of course from a traditional election: rather than votes per say, its just about getting enough endorsements from your peers. Doesn't produce any explicit losers, which might be nice. If the Clan Assembly has 100 members, if 10 votes produces a locality representative then 10 people will be selected to move up, and everyone will have voted for a "winner".

Since people are curmudgeonly, you'd probably have to do some form of jury style coercion: from the morning of election day, all the electors are locked in a room until everyone has voted. Then maybe if after some time period some people have still not voted, you draw lots.

So, no losers per say, and maybe fairly legitimately non-partisan: I'm not quite sure how political parties would interact with a series of 5-15 man votes. I'm sure there's a way, but I'm not immidiately seeing it.

This seems quite superior, but I'm just theorizing things, and I know stuff always sounds better in one's head.
 

LordsFire

Internet Wizard
This is an interesting theory, but it runs into some of the usual problems of abstraction.

Particularly, how it ignores the relationships that people have outside of this hierarchy of representation, and how it completely diffuses any kind of responsibility due to the multiple layers of separation.
 

JagerIV

Well-known member
This is an interesting theory, but it runs into some of the usual problems of abstraction.

Particularly, how it ignores the relationships that people have outside of this hierarchy of representation, and how it completely diffuses any kind of responsibility due to the multiple layers of separation.

Hm, doesn't any theory of government generally abstract out the much of the relationships outside the hierarchy? I'm not sure this abstraction doesn't do so particularly, besides it being an idea in the process of baking, instead of something with a full book of explainers.

On diffusion of responsibility, I'm not sure such a system is diffused in such a way to limit responsibility. If anything it would seem likely to increase responsibility in some manners, as the power is more direct and narrower in focus:

Lets say you had a land dispute in the West owned by the Fed: getting a response on that is difficult: of the 500 or so Federal representatives, you have 10 representatives in that area, 2% of the total, putting a low cap on how much individual interest the Federal government overall can focus on it. And even if you get a federal ear, the layers of federal bureaucracy makes it all fairly unaccountable.

Meanwhile, if the Land is owned by a Commonwealth, you probably have one dedicated just to the West, so its within the interest of a 100% of the assembly, and even if you only get say 10 reps to care, in a smaller assembly of 100 that's 10% rather than 2%.

Of course, if all power does get concentrated into the highest level of the hierarchy, then the practical increase in representativeness goes away, and its a tool of avoiding responsibility. Thus the various heavy handed measures to try and keep power from all concentrating into the highest levels, namely the tax cap, dispersion of funds, and employee caps.

The big goals here are

1) Representative: the state effectively represents the people
2) Responsibility: As much as possible, responsibilities are clear, even if overlapping
3) Dispersed Power: Power is widely distributed
4) Cohesion: society and government are structured to lead to cohesion
5) Human Scale: As much as possible, keep things in human scale, rather than giant incomprehensible conglomerates.
 

JagerIV

Well-known member
It sounds like you're trying to recreate the effects of our Federal system, back when the constitutional limits on Federal powers were actually at least somewhat followed.

Hm, maybe, but that sort of seems like suggesting every Republic is an attempt to re-create the Roman Republic.

Its a federal system, but the mechanism here seems rather different.

Edit: Some continuing thoughts/reforms

If each representative is not chosen by a traditional at large election, but instead by receiving enough recommendations to be promoted up the chain, then it makes sense to keep each level at the level of personal familiarity, especially since as you say the rep will have relationships beyond the political hierarchy. By the Dunbar number, this suggests about 5-15 per rep. This gives enough so some personal connection can be maintained among the direct reps, and provides enough flexibility to form functional governments over a wide variety of population layouts, while still guaranteeing fairly local government at, well, the local levels.

The max, min and averages of such a system are calculated below, showing the maximum, minimum, and "average" number of people represented by each rep, and the scales of their potential states, with the minimum state having 50 reps, while the maximum having 150 representatives.


NameMinimum Represented Pop (5 per)Average Represented Pop (10 per)Max Represented Pop (15)Minimum Government (50 Reps, minimum represented)Average Government (100 Reps, average represented)Maximum Government (150 Reps, max represented)
Clan510152501,0002,250
County251002251,25010,00033,750
Society1251,0003,3756,250100,000506,250
State62510,00050,62531,2501 million7.6 million
Commonwealth3,125100,000759,375156,25010 million114 million
Confederation15,6251 million11.4 million781,250100 million1.7 billion
Union78,12510 million170 million3.9 million1 billion25.5 billion
Federation390,625100 million2.5 billion19.5 million10 billion375 billion
Protectorate1.9 million1 billion38.4 billion95 million100 billion5.7 trillion
Imperium9.77 Million10 billion576.6 billion488 million1 trillion86.5 trillion

Names have also been adjusted, to reflect a non-territorial/territorial pairing scheme, where the different levels in an alternating order are non-territorial vs territorial.

For example, a Clan would not be bound to a particular territory: As long as you can make it to the Clan election/meetings, you can vote as a member. A county, however, refers to a definite territorial location. Territorial representatives likewise would probably have residency requirements. Thus, for example, a father who worked overseas 8 months of the year may be able to still vote with his family for clan representation, and may vote for his County representative, but might not be eligible to sit as the actual county representive, depending on how strict residency requirements are.
 
Last edited:

JagerIV

Well-known member
NameTax Rate IncomeTax Rate LandPer Capita Tax, $70,000 Average IncomePer Capita Tax, $60,000 Average Land ValueUS Estimated "Average" representedReps US (300 million)Governments(average assembly) US Average Income Per Rep
Clan10%$7,0006.645 million600k (75)$46k
County9%$5,4006.6*13.5=893.3 million24k (140)$480k
Society8%$5,60089*6.6=600500k5k (100)$3.4m
State7%$4,200600*13=8k40k280 (140)$44.8m
Commonwealth6%$4,2008k*6=48k6.6k66 (100)$202m
Confederation5%$3,00048k*5=240k1,32010 (132)$720m
Union4%$2,800240k*5=1.2m2644 (50)$3.4b
Federation3%$1,8001.2*5=6m531 (53)$10.8b
Protectorate2%$1,4006*10=60m5(partial)$84b
Imperium1%$60060*5=300m1(partial)$180b
Totals30%25%$21,000$15,00075.7 reps per Imperial

For a decentralized system to stay decentralized requires safeguards to preserve it, since the tendency of power is to centralize.

One of the methods proposed here is the method of allocating Tax revenue.

Tax is constitutionally set on two taxes, a tax on Income, and a Tax on land value. The different levels are then due a set rate on those taxes, as listed above. These add up to a set 30% tax on income, and a 25% tax on land value. If you want local government, it has to be paid for.

However, this is not practically as high as it might appear, due to how the funds are distributed. For example, Clan income rate is 10%. Lets say you had a family of 2 children, 2 parents, and the 2 grandparents. 3 of the Adults work still for good jobs, say $60k a year, so the full family makes $180k a year income, with 20k investments for a nice round $200k. Then say they own an acre of land collectively worth $50,000.

With this much income and land, at these tax rates they pay $60k in income taxes, $12,500 in land tax. Total Tax burden 36.25%. Fairly high, but not necessarily ruinous either. European tax rate.

However, being a clan rep only requires 5 votes, including the individual. The 6 members of the family can thus elect one of the men to be the clan representative. Now, say his clan overall has 400 members, about 60 families to make up the clan assembly. Say the clan members match US income average of $70,000 per capita. Overall then, the assembly takes in $2.8 million per year with its 10% rate.

However, instead of spending it as a pot, the clan instead is required to distribute the money to the representatives proportional to the number they represent. Since this family has 6 members, their clan representative thus receives $7,000*6=$42,000 to spend as the representative sees fit.

If he simply wishes to offset his taxes, he can treat it as a tax refund, reducing his tax burden to $30,500, a mere 15% tax burden.

It further encourages general buy in if the assembly votes to provide anything: the Clan can vote to pass a measure through the normal 50% vote measure, such as to establish a local clinic or school, or upgrade the Clan house, but to fund anything requires pledges from specific representatives. You can pass something on 50%+1 votes, but then you have to pay for it out of 50%+1 of the money.

And with all the representatives having the option to just pay for it themselves, that further puts a great deal of pressure on the government companies to perform: the Clan as a collective could contribute $6,000 each to run a small school for the local children. However, if the school is terrible and the parents can instead take their $14,000 per child and send them to a private school, they can do that and take their money out of the clan school. Or decide the Clan is too small a location, and instead pledge $1,000 per person to the county school system one level up...

Many Choices, preserving accountability.
 

JagerIV

Well-known member
Territorial thoughts

LevelAllowed Pop Range
(min-average-max)
Max SizeMin number Per Texas (695k km^2)Per pop (min - average -max)
County1k -10k - 30k100 km^26.9k29k - 2.9k - 966
State31k - 1m - 7.6m10,000 km^270935k - 29 - 3.8
Confederation781k - 100m - 1.7b1 million km^20.69537 - % - %
Federation19.5m - 10b - 375b100 million km^2.00695 1 - % - %
Imperium448m - 1t - 86t10 billion km^2.0000695% - % - %

Final throwing things at the wall, particularly new thoughts. I've toyed with the idea of territorial and non-territorial layers. Above are the territorial layers, and I got thinking, what if there was some limit here too, to force things to be on a "human" level.

The county thus is approximately scaled to what's walkable in an hour: people walk approximately 5 km/h, so from a central location everything in a roughly 10x10 square would be within reasonable walking distance, and thus something of the maximum of a natural human community. I tried to come up with "human" scales for the rest, but in the end settled on a mere 100x increase between each layer: this matches the rough 100x increase population between layers, and I think it produces usefully scaled increments.

If nothing else, it puts a limit on the scale of human greed: a county government by the above system can claim a 100 km^2 of territory on its own if it can merely get 1,000 people to settle there, but not more.

This also should make it more acceptable for governments to shrink as they grow, a necessary step under this scheme: County pop is capped at roughly 30k, and practically more like 10k is a practical pop. 10k over 100 km is a population density of 100 people/km^2, which is quite low for any sort of built up area. Manhattan is roughly 30k/km^2, suggesting under this scheme it would be 2-4 Counties.

Then again, the non-territorial clan layer can provide a bit of relief there: one may live and work in New York, but then every year or so return to one's home town to vote at their clan hall. Thus, the population residing in Manhattan gets counted for political purposes in their home town. The local Clan probably likes this, since it brings high earner NY incomes into the clan's pop, raising the average payout, The NY gets the relative power of being a relatively big fish in a small pond, and the NY county maximizes its revenue by keeping as much of that high value NY real estate within their taxing authority, and also helps keep the county at a size where functional services can still be provided.

On the other end of things, such territorial limits also potential leaves large areas of unincorporated land, leaving open the option to those who wish to found their own government:

map-population-density-texas.png


The minimum population size for a county is 1,250. Spread over a maximum sized county, that's a population of 12.5 pop/km^2, about 5 pop/m^2. This suggests the areas in deep green might not have enough people to form counties. If people are cluttered in 5,000 pop counties, then even if each is max size (and many wouldn't be), the current Texas Population would only support 5,800 counties coving at most 580,000 km^2, leaving around 20% of the state unincorporated at the county level.

If the States cluster around where there's people for about 1 million per State, 30 States would only cover about 40% of the State. Its only at the Confederation level that an area as large as Texas can be fully incorporated without issue.

This limit then creates an interesting situation: you might have clans in an area that's not densely populated enough to form counties, or isolated counties that aren't close enough to form a State, ruled over by comparatively distant and limited Confederate government.

The idea of a durable frontier is facinating. Puzzling over how that could operate is interesting. Rules as written may be that you don't get high level representation, but a much reduced tax burden. So, if you move to somewhere too sparsely population to have a county or State government, you don't have the chain of layers to vote a Confederate Representation. However, your also missing those two layers of taxation, so on land your only taxed by Confederation up, so instead of 25% of land value, your taxed 9%.

Income the chain of voting, but also responsibility, is also broken: you might have your clan organization, but with no higher level, that suggests your tax obligation might end at your clan. At least if your income is paid to the clan: If you work for some large company that pays taxes at the Commonwealth level, maybe they withhold the full 30%, take their 6% cut and send on the other payments to higher levels, but then you don't have lower levels they distribute to. Major changes in incentives if they are allowed to keep the 18% tax rate, or have to return it. Or maybe a compromise position: if they can't pay your Clan/Society tax, since the taxpayer is not presently part of one, the Commonwealth can hold onto such money in trust until such payments are possible. Using it as collateral, in the usual bank way...

How this system handles non voters in general in an interesting mental puzzle. For another time.
 

JagerIV

Well-known member
@LordsFire I'm debating between ripping out the existing first post and replacing it with a mock Constitution showing the current "consensus" for this imagined government, updating it as I think of things and update things.

Or, I might write up the constitution as I've figured it by that point in a new post, so progress can be seen as it goes.

I'm currently leaning to the first: means anyone new can see the overall system being considered, and the first post right now doesn't really provide much useful independent information anyways.

Uh, what you know, running across something along the lines thought here, though obviously radically different, considering similar issues.

 
Last edited:

JagerIV

Well-known member
NameBoarder Walk TimeMax Boarder (km^2)Travel per Day to reach Max (km/day)Rough land Area at Max Scale km^2Lower Levels Max SizePop Per RepTotal Pop per RepMinimum Pop RepresentativesMinimum Pop StateMax Pop State (150 Pop Representatives)
1: Manor1 Hour551.56077422941,050
2: County1 Day4040100651391353,18513,650
3: Region1 Week20028.52.5k25181,6382845k245.7k
4: State49 days1,00020.462.5k252236k21756k5.4m
5: Commonwealth1 Year5,00013.691.56m2525900k1412.6m135m
6: Union7 Years25,0009.7839m252724m7168m3.6b
7: Federation49 Years100,0005.6UnlimitedUnlimited28681m10?102b?

Experimenting with fewer rows: makes things cleaner, but also makes each level more complex, and forces a higher representative to represented ratio. Not sure if its really a good trade.

Edit: Finding good results with the magic number 7. 7 layers means just about everyone should be able to remember all the layers they are a part of, certainly the 5 that should be impactful on an individual.

7 creates the clear middle level, 4- the State, giving it special importance.

I was also able to figure out a good way to make the scale of the states "human" with the idea of the league, what's walkable in a day.

Thus, the size of a government is constrained by if the boarder would be walkable within a certain time: setting the boarder requires walking it during an initial declaration, and then running for that level of government would also require walking that boarder: requires traveling across much of the state, and puts a soft (ish) fitness test on people running. Walking 25,000 km requires a dedication to the Union, high level of fitness, and effectively the dedication to a 7 ish year campaigning across the frontiers of the Union.

I've considered minimal territorial requirements, but that seems like it would be generally unnecessary: People in the sate government have every intention of maximizing the size of the state possible: it maximizes tax revenue!

However, the requirement to walk the boarder to the degree possible would also likely encourage that a government's boarders follow some natural contours. As well as the creation of roads to get clear boundaries, especially for lower levels.

Overall, I'm pretty happy with 7: 10 layers I think is too many, 5 layers however force some award scale jumps. This preserves truly local governments, and large scale overarching governments.

Probably better ways to make the arguments here, but toying around have produced a system that seems to give useful results.

Also happier with the names, though Manor, county, and Union I'm still a bit iffy on: Manor and county feel a bit too feudal for such a project, though maybe a splash of the traditional is useful.
 
Last edited:

Users who are viewing this thread

Top