A Complete Alternative to Bourgeois Democracy

The System of Systems

A schism election is a term of art that I have given to a process for choosing an outcome. The basic way that it works is that you first vote for one of two categories, let's say "left wing" or "right wing." On than same ballot, you vote within the first category and chose one of two categories again. If the election has three rounds you might vote within that kind category yet again. Everyone votes in all categories though not all votes wind up being counted. Only the majority in the first round gets its votes counted in the second round, and only the majority in the second and first rounds get counted in the third round (if there is a third round). A schism election produces a schism, that is, instead of representing the median voter it represents the strongest minority block within a voting district. It lets you choose the majority within the majority, rather than having to compromise all the time. The whole purpose of a schism election system is to allow multiple voting districts across a wide territory to experiment with various systems of government, and to do this experimentation in a structured and peaceful way that prevents a breakdown into civil war and authoritarianism. By its nature schism elections must have be a regular occurrence, say every 10 years, or every 20, so that if the people are unhappy with one system of governance they can throw it out and get another. Regular democracy produces compromise, schism elections produce experimentation.

A schism election CAN produce an oppressive government. Let us say that a right-wing government comes to power and it bans abortion. The women where it occurred then flee the area and it winds up with a shortage of women. To get women to move back it has to give them a trade off to make it worth their while, perhaps it also bans premarital sex, or perhaps it strengthens maternity care, child care, and other financial benefits. The right to leave is an absolute right, it is the one right that a system with schism elections must always demand. Exit forces every system in a federal system with schism elections to offer a trade-offs to entice the population to deal with whatever regulatory structure it creates. This act of voting with one's feet creates an actual limit on regulatory overreach and tyranny. And of course nothing stops the system from also having basic rights enforced by a Supreme Court at the Federal level, though it works better if the rights are absolutely minimal and specific.

Bourgeois democracy is the only system most people have ever lived under, so they tend to treat it as the default setting of politics rather than one design among many. It is one design. It has a specific ownership structure: every adult in the territory is a shareholder with one share, the government is the company, and the election is the shareholder meeting. Bourgeois democracy is basically customer owned government. There's nothing sacred about that arrangement, no reason to think it is appropriate for everyone, and if we treat governance as an engineering problem, we can build something else, something which is also democratic, but something that allows for much more latitude. Here is a complete system that combines three ideas I've written about or mentioned separately: schism elections, shareholder republics, and a heavily modified version of review democracy, aka loyalty scorecard democracy.

Start with the schism election. The United States has 3,100 counties, and each one holds its own schism election to choose which governance company will run it. The county is the right unit because the state is too coarse. Do this at the state level and you trap millions of people in a system they didn't choose. Do it at the county level and you get thousands of small experiments running side by side. The county level is better than the state because it is more fine-grained and allows for a wider expression of preferences. It doesn't deny millions of people representation of the doing it at the state level would. What is a schism election? Give me a minute and I will tell you.

A corporation, for lack of a better term, runs a county. The company that runs the county is a business, and it can be organized in different ways. The schism election sorts those ways into two categories, left wing and right wing, and the categories are defined by who owns the company. Left wing governance companies are owned by everyone in the county, either as customers or as workers. Customer ownership means every person over 18 gets a vote in the company's internal elections. Customer ownership is basically just regular democracy because it's ownership structure is customer owned, and therefore voter owned, since every voter can participate in its internal elections. That is ordinary bourgeois democracy, and it's worth seeing it for what it is: government as a customer-owned cooperative.

In contrast, worker ownership means only the people in the county who are workers get to vote in internal elections. That is communism, and it changes the economics and what economic relationships are legally allowed. Surplus revenue from the governance company is returned to the voters as a dividend, business owners are shut out of the franchise, and the county gains the power to socialize businesses.

Right wing governance companies come in three types. A religious governance company is run by a church, and its internal voting is whatever the church's own governing structure says it is. A restricted franchise company only lets landowners vote, or only business owners, depending on how the firm registered itself on the ballot. A corporate governance company is exactly what it sounds like, with shareholders voting their shares, including class A and class B shares if the charter says so. Only shareholders have a say.

In all cases the internal charter of the governance corporation determines how its internal elections are handled. Religious corporations are the only ones that can have a hereditary monarchical organizational structure for their internal governance.

Keep the two levels of voting separate, because they are easy to confuse. The schism election is the first vote, where the whole county decides what kind of governance company it wants. The second vote is inside the company, where its owners decide how it is managed. A county that votes for a restricted franchise firm is not thereby giving every voter a say in the firm's operations. It is handing the county to a company whose internal elections belong to landowners, or business owners. Voters choose the ownership structure, and then the ownership structure decides who governs.

The schism ballot itself works in stages. Voters choose left or right. Then they choose a subcategory: bourgeois democracy or communism on the left, or religious, corporate, or restricted franchise on the right. Within the winning subcategory, registered governance companies compete for the contract, and the winner takes the county.

Now for the check on all of this, which is where review democracy comes in. My earlier version of it worked by drafting a few hundred citizens by lottery into a jury, having them rate every bill their politicians voted on, and turning the ratings into fidelity scores that decided who could run for higher office and who got barred from re-election. That works for a legislature. It doesn't fit here, and the differences are big enough that this should be treated as a different mechanism.

First, the jury isn't a few hundred people. Every registered voter in the county who wants to take part can, which can mean thousands of participants. Second, this review isn't a primary and doesn't set term limits. Its job is narrower: it is an iterative process for deciding whether the incumbent governance company deserves to be fired, and for informing the schism election that follows if it does. Third, the process is overseen by permanent civil servants, and those civil servants work for the state government, not the county. The county cannot capture its own referee.

The process runs in stages. It begins with a mock election, a simulation of the schism election. The result is only a ranking, and it identifies who would come in first, second, and third. Those three become the candidates for scrutiny. The jury then reviews the actions and results of each of them, looking at what they have done in the other counties where they have governed. A governance company that wants your county has to show you what it did to somebody else's.

After that the jury is asked whether it wants to keep going. If it votes yes, the fourth, fifth, and sixth place companies get the same examination. If it votes no, the process moves on. Then the jury reviews the incumbent, meaning all of its actions and all of its results during its term. Then comes the only vote that matters at this stage: fire the incumbent, or keep it.

If the jury keeps the incumbent, the process ends and the company continues. There is no schism election. If the jury fires it, the real schism election happens, the tally is counted, the results are announced, and the new governance company takes office. The whole cycle repeats every ten years.

The ten-year term is what makes governance vandalism punishable. A governance company knows it will be reviewed on its results, and it knows that a successor will inherit whatever it leaves behind. So it is illegal to saddle a future county with problems and debt. Load a county with liabilities, gut its infrastructure, or borrow against the next decade to make this one look good, and you have committed governance vandalism, and the corporate executives who did it go to prison. Prison is the point. A fine is a cost of doing business, but a cell is not.

The entire process is supervised at the federal and state level. Issues related to race and immigration are decided only at the federal level, so no county can settle them for itself, and race based parties are not allowed to run in schism elections. Counties get wide latitude over how they are governed, and those subjects are not part of that latitude.

Put the pieces together and you have a full alternative. The schism election lets each county choose what kind of institution it wants running it, from a customer-owned democracy to a church to a corporation. The shareholder republic supplies the structure, with government as a company that has owners, executives, and a charter. Review democracy supplies accountability, with a county-wide jury that judges by track record and can fire a company that has failed. You say you want a government that answers to the people it governs. This is one way to build it, and it doesn't require anyone to pretend that a single system is right for 3,100 different counties.

The Encyclopedists

They say history is written by the victor, they also say that those who do not study history are doomed to repeat it. If something is this important you would think there would be a reference for checking what happened. If something this important is so corruptible you would think there would be a record for holding the writers to account.

The ultimate check on all of this is a book, or really a shelf of them. Every schism jury gets a full set from an Encyclopedia of Success and Failure. A jury deciding whether a governance company deserves to be fired needs something to measure it against, and right now voters have almost nothing. They have campaign ads, a few cable news segments, and whatever someone ranted about at Thanksgiving. In the beginning of the schism system they will have an encyclopedia listing all the successes and failures of the government. After awhile they will have both the encyclopedia and the results of various experiments. There should be multiple competing encyclopedias, at least one produced by each State in The Union, all vying for legitimacy, with voters having access to a digital version of all of them. Each encyclopedia is used to train a large language model so that information can be delivered in question and answer form with countless examples presented order to give a large sample size for every question. A voter sitting on one of these juries can query one or more of the encyclopedias in order to find out anything they want to know. They can compare the performance of any governance company to other examples so they know how far it has to go.

The encyclopedia is organized by subject, like any other encyclopedia, and it records every action a government has taken that can be documented. Examples: forestry management, housing policy, monetary policy, zoning, policing, sanitation, environmental regulation, foreign intervention, all of it. Under R you would find Regime Change, and that entry would cover every declassified instance, how it was carried out, who carried it out, what it cost, and what came after. The description of events is printed in black. That part has to be factually accurate, and the permanent state civil servants who referee the schism process are the ones who police it. Academics in each state are required to participate and being an academic in any field means also being an encyclopedist.

The judgments are printed in color. Success and failure are value judgments, not facts, and the encyclopedia doesn't pretend otherwise. A liberal and a conservative can look at the same set of events and honestly disagree about whether it worked, because they disagree about what the government was supposed to be doing in the first place. So the left writes its verdict in blue and the right writes its verdict in red. If both sides agree, the verdict prints in purple, and if they disagree you get a blue verdict and a red verdict sitting under the same block of black text.

The first benefit is that it separates the facts from the values, which almost no political discussion manages to do. Right now the two sides argue about both at once, so nobody ever finds out which disagreements are about what happened and which are about what should have happened. With the encyclopedia you can see the seam. A juror looking at a page of black text with a blue verdict and a red verdict knows immediately that the fight is about values, and a juror looking at a page of purple knows that this is one of the rare things both camps will admit to.

The second benefit is memory. Governments forget things on purpose, and voters forget them by accident. A governance company that wants your county has to show you what it did to somebody else's, and the encyclopedia is the place where the jury checks the claim. A company that says it balanced the books in the last county it ran can be compared with the entry on municipal finance, and a company that says it reformed housing can be held to the housing entry. Promises become checkable, which raises the cost of lying. It won't stop lying, but a lie now has to survive a jury with a reference book. Each of the sprawling encyclopedias is probably kept in digital form but every 10 years a snapshot is taken and printed on archival media so that there is a hard record of the actions and statements taken by everyone in the past. This acts as a vast fact checking source and prevents the winner from having an exclusive monopoly on history.

The third benefit is that the verdicts are written down in advance and in ink. A party can't quietly redefine success after the fact, because its old judgment is printed right there in its own color. If the left called a policy a success in 1985 and calls the same policy a betrayal in 2015, the jury can see both. The same goes for the right, It also gives the two camps an incentive to think harder about their verdicts, since they are going to be quoted back to them for decades. And it gives everyone in the system a sense of historical memory which tends to break them of the arrogant American habit of thinking you know everything and disrespecting other ways of doing things.

The fourth benefit is that it keeps the experimentation from being wasted. The whole point of schism elections is that thousands of counties try different systems, and an experiment nobody records is just a story. Over a few cycles the encyclopedia turns those experiments into a data set, with the same actions tried under a religious firm, a worker-owned firm, and a corporate firm, all of it indexed by subject. That's how a country learns things, and at the moment we learn very little because we run one experiment at a time and argue about the results, and every generation history is forgotten so that mistakes get repeated. For example: long before the Soviet Union and Maoist China tried collectivized agriculture the Jamestown colony in America tried it and starved as a result. Again, you would think something as important as history would have a reference for all its successes and failures. The job of the encyclopedias is to maintain that record, the reason every state produces one is to keep things honest through competition. The reason all voters have access to all of the encyclopedias of each state is to use prestige and reputation to pressure the writers to stay honest. Encyclopedias are generally a collaborative effort between two political factions so as to give both of them greater legitimacy by contrasting their actions with an opponent. The adversarial text color system is a feature and not a bug. 

In Conclusion 

Putting it all together you have a complete system for allowing people to experiment with various forms of government and documenting the consequences of those experimentations. After a few centuries of this the experiments may no longer be necessary and the future political system, perhaps a monarchy, can simply point to a centuries worth of data to justify each and every decision it makes. Instead of saying that it has a divine right to rule it simply says that it has 3,100 experiments spaced across 200 years of history as its guide. It can say, "no we're not doing that and here are 25 examples for why it failed." It doesn't have to assert some blind obedience to God or higher power. It doesn't have to continuously prove itself. It simply has a mountain of data to justify every decision it makes and it might even have an AI you can argue with all day and all night that justifies everything the system is doing. Hell, the AI might become the monarch, instantiating a digital governance with its vast history and knowledge of human reactions. And since it's cognitive bandwidth is unlimited it might direct the actions of every department independently of any human management structure. This system of systems becomes the training ground for something potentially much better than any human governance. And since an AI never dies you have solved the succession problem once and for all.