OL6: the lost theory of government
The best way to understand government is to assume everything you know about it is nonsense. Or so at least I claimed in Chapter 5. Let’s demonstrate it by solving the problem from scratch.
Growing up in the modern Western world, you learned that in all pre-modern, non-Western societies, everyone—even the smartest and most knowledgeable—put their faith in theories of government now known to be nonsensical. The divine right of kings. The apostolic succession of the Pope. The Marxist evolution of history. Etc.
Why did such nonsense prosper? It outcompeted its non-nonsensical competitors. When can nonsense outcompete truth? When political power is on its side. Call it power distortion.
And why, dear open-minded progressive, do you think your theory of government, which you did not invent yourself but received in the usual way, is anything but yet another artifact of power distortion, adapted to retain your rulers in their comfortable seats?
Probably because there is a categorical difference between modern liberal democracy and the assorted monarchies, empires, dictatorships, theocracies, etc., which practiced the black art of official mind control. The priests of Amun tolerated no dissent. They flayed the heretic, the back-talker, the smartmouth, and stretched his still-living flesh to crack and writhe in the hot African wind, till the hyena or the crocodile came along to finish him. But now they are all pushing up the asphodels, and Google hasn’t even thought about deleting my blog.
You think of freedom of thought as a universal antibiotic, a sure cure for power distortion. It certainly allows me to post my seditious blasphemies—for now.
But as a progressive, your beliefs are the beliefs of the great, the good and the fashionable. And as we’ve seen over the past few chapters, power can corrupt the mind in two ways: by coercion, or by seduction. The Whig, the liberal, the radical, the dissenter, the progressive, protests the former with great umbrage—especially when his ox is being gored. Over the past four centuries, he has ridden the latter to power. He is Boromir. He has worn the Ring and worked it. And it, of course, has worked him.
Today’s late Whiggery, gray and huge and soft, lounges louche on its throne, fastened tight to the great plinth of public opinion that it hacked from the rock of history with its own forked and twisted tongue. The mass mind, educated to perfection, is sure. It has two alternatives: the Boromir-thing, or Hitler. And who wants Hitler? Resistance is more than useless. It is ridiculous. The Whig cackles, and knocks back another magnum of Mumm’s.
And a few small rats wear out our incisors on the stone. In this chapter we’ll learn the real principles of government, which have spent the last four centuries sunk under a Serbonian bog of meretricious liberalism. (“The funk… of forty thousand years.”) We’ll have to wait until Chapter 7 to see what government is today.
The two, of course, have nothing to do with each other. Nor is this likely to change soon. Nor can you do anything about it. So why bother? Why think about government?
The only defense I can offer is Václav Havel’s idea of “living in truth.” As a fellow cog in the global public supermind, you are bombarded constantly and from every direction with the progressive theory of government, with which all humans who are not ignorant, evil or both must agree by definition, and which makes about as much sense as the Holy Trinity. If you are ready to be the nail that sticks up and is hammered down, you can be a “conservative,” which ties up a few of the loose ends, and unties others. It also makes you a social pariah, unless most of your neighbors are named “Earl.”
This shit is stressful. Most of us already have stressful lives. Do we need it? We don’t. The nice thing about understanding government is that it gives you an off button for the endless political yammering. While it may replace this with a bit of despair as regards the future, the future is a long way off. And not entirely without hope, but that’s another post.
In any case: government.
First, let’s establish what a government is. A government is a sovereign corporation. It is sovereign because its control over its territory is not subject to any higher authority. It is a corporation because it has a single institutional identity. All governments in history fit this definition, unless their sovereignty is compromised by some stronger power. In this case, that power is the true sovereign, and your analysis should be aimed at it.
Second, what makes a government good or bad? The easiest way to think about this problem is to think subjectively. Assuming you have no power over the government’s decisions, under what kind of government would you prefer to live? Given two governments A and B, what would make you move or want to move from A to B, or vice versa?
The key is that we are evaluating a government based on what it does, not what it is. As Deng Xiaoping—probably the greatest statesman of the 20th century—put it: “Who cares if the cat is black or white, so long as it catches mice?”
The subjective approach asks whether the government catches mice. It does not ask who the government’s personnel are, or how they are selected, or how they are managed. Perhaps they are all Dinka warriors from the middle of nowhere, Sudan, chosen for their impassive visages as they execute the brutal Dinka ritual of auto-hemicastration with no implement but their own fire-hardened fingernails. If they govern well, so much the better.
Your subjective desires for government may be different from mine. They probably are. In a world of good governments, subjective preferences would reduce to the trivial and cosmetic. If I am in the market for fast food and I see a Burger King next to a McDonalds, I will go with the King. Why? Does it matter?
Fast food is an fine metaphor for government. You’d think managing a sovereign corporation is probably more complicated and difficult than operating a fast-food chain. Heck, operating a nonsovereign US state is probably harder than flipping burgers. And if B is harder than A, you’d think anyone who can pull off B would ace A.
But if I saw a McDonalds next to a Calmeat, Mickey would be my man. Of course, there is no Calmeat. We do not live in a world where the State of California sees fit to operate restaurants, fast or otherwise. There is no state burger. Even as an open-minded progressive, however, I’m afraid you will have to concede that if there was a Calmeat, it would either be either horrible or horribly overpriced, and probably both.
Why? It will become obvious, if it isn’t already. But what it tells us—if this isn’t already obvious—is that we don’t live in a world of good government. California is better-governed than nine-tenths of the Earth’s surface. And there is no way its government could flip a decent burger. As Mark Twain put it:
Omar Khayam, the poet-prophet of Persia, writing more than eight hundred years ago, has said:
“In the four parts of the earth are many that are able to write learned books, many that are able to lead armies, and many also that are able to govern kingdoms and empires; but few there be that can keep a hotel.”
Twain’s quote does not strike me as authentic—but I quail at the notion of Calstay. In any case: not only do we not live in a world of good government, we live in a world of disastrously bad government. If the 20th century does not go down in history as the golden age of awful government, it is only because the future holds some fresher hell for us.
So we are not concerned with the subtleties of good government. We are not interested in excellent government. It would be nice, but we would be satisfied with mere competence—perhaps with whatever enables McDonald’s to survive in a world that contains not only BK but also In-N-Out, even though its burgers taste like boiled cardboard. Our goal is the mere basics.
Here are the basics: a government should be secure, effective, and responsible. None of this is rocket science. The only secret is that there is no secret.
Let’s define and analyze these qualities individually, assuming the others in each. When we explain how to make a government responsible, we’ll assume it is secure and effective. When we explain how to make it secure, we’ll assume it is effective and responsible. Etc.
Let’s start with effectiveness. Effectiveness is the ability to accomplish what you’re trying to do. Under what design is a government most effective?
We can think of effectiveness as a measure of good management. A well-managed enterprise hires the right people, spends the right amount of money on them, and makes sure they do the right things. How do we achieve effective management?
We know one simple way: find the right person, and put him or her in charge. This single, frail being, our administrator, holds final decision-making authority—the Roman imperium—over budget, policy, and personnel.
In the military world, this is called unity of command. In the (nonsovereign) corporate world—and in the nonprofit world that opposes it—this individual is the CEO. Even that most anarchic of human endeavors, the open-source project, tends to follow the administrator design.
Why does individual administration work? When said individual is a douche, it doesn’t. There is no reliable formula for good management. But there are many reliable formulas for bad management. A better question is: why does management by committee not work?
Divided control of any human enterprise tends to fail because of a phenomenon generally known, around the office, as politics. Politics always emerges when management breaks down. An individual manager, with undivided control of some enterprise, can only succeed by making the enterprise succeed. Replace one manager with two—the unorthodox administrative design known as “two-in-a-box,” a disaster I personally have experienced—and either has a new way to succeed: making the other fail. The more cooks, the worse the broth.
In every human endeavor outside government itself, undivided administration is well-known to produce optimal results. If Peet’s could beat Starbucks, Southwest JetBlue, or In-N-Out Mickey D’s, by adopting a “separation of powers” or a “constitution” or some other architecture of leadership by consensus, one of them would certainly have tried it.
Contemplate, dear friends, the great heap of rococo procedural ornamentations that have replaced the simple principle of personal decision in the modern Western government. Montesquieuean separation of powers is the least of it. Outside the military, in which the principle of command still functions to some extent, it is simply impossible to find anyone with unified responsibility for getting anything done. And even military officers, while they have some vestiges of imperium—rapidly being sucked away by the judicial system—seldom control anything like their own budgets, and have zero power over personnel.
So: the modern aversion to individual management cannot be motivated by effectiveness. Undivided administration is more effective, period. We can only explain the penchant for collective decision-making as a function of responsibility or security. It is hard to see how it has anything to do with security. It must be a matter of responsibility.
But, in a system where no individual can be connected reliably with any success or failure, where is the responsibility? As none other than Woodrow Wilson put it, in 1885:
It is quite safe to say that were it possible to call together again the members of that wonderful Convention [of 1787] to view the work of their hands in the light of the century that has tested it, they would be the first to admit that the only fruit of dividing power had been to make it irresponsible.
Wilson himself, of course, had a great deal of undivided power. Nor did he use it responsibly. When we think of sovereign executives, we tend to think of bad examples. We think of Hitler, not of Frederick the Great. We don’t think of Sultan Qaboos or Lee Kuan Yew or Hans-Adam II. If you think this is a coincidence, think again. But perhaps a thought-experiment will help.
Washington, especially since it governs not only the United States but also most of the world, is just too huge to serve as a good thought-experiment for government. It’s easier and more fun to think in terms of California, if California could somehow be a sovereign state. Assuming security and responsibility, how could we produce effective government in California?
The answer: find the world’s best CEO, and give him undivided control over budget, policy and personnel. I don’t think there is any debate about it. The world’s best CEO is Steve Jobs.1
Which would you rather live in: California as it is today, or Applefornia? Which would you rather carry: the iPhone, or the Calphone? I rest my case.
So let’s segue into responsibility. Assuming a government is responsible and secure, we know how to make it effective: hire Steve. But how do we make it responsible?
Steve, after all, is a turbulent fellow. He is moody at best. He could easily go around the bend. And he is already a notorious megalomaniac, a tendency that total imperium over the Golden State—including its new military forces, whose heads are shaved, whose garb is white linen, and whose skill in synchronized martial-arts demonstrations is unmatched even on the Korean peninsula—can hardly ameliorate.
A responsible, effective government has three basic parts. One is the front-end: all the people who report to Steve. Two is the middle: Steve himself. Three is the back-end: the people Steve is responsible to.
Apple itself, like all public corporations in the modern system, has a two-level back-end: a board of directors, elected (in theory) by a body of shareholders. There is no reason to copy the details of this system. Corporate governance in the US today is nothing to write home about. It is the principles that matter.
Call the back-end the controllers. The controllers have one job: deciding whether or not Steve is managing responsibly. If not, they need to fire Steve and hire a new Steve. (Marc Andreessen, perhaps.)
This design requires a substantial number of reasonably cogent controllers, whose collective opinion is likely to be trustworthy, and who share a single concept of responsibility.
What happens if the controllers disagree on what “responsible” government means? We are back to politics. Factions and interest groups form. Each has a different idea of how Steve should run California. A coalition of a majority can organize and threaten him: do this, do that, or it’s out with Steve and in with Marc. Logrolling allows the coalition to micromanage: more funding for the threatened Mojave alligator mouse! And so on. That classic failure mode, parliamentary government, reappears.
Call a controller model with a single shared concept of responsibility coherent. How, with an impossibly fuzzy word like “responsibility,” can we round up a large number of intelligent individuals who share a common definition? The task seems impossible. And our whole design relies on this coherent back-end.
Actually, there’s one way to do it. We can define responsibility in financial terms. If we think of California as a profitable corporation, a capital asset whose purpose is to maximize its production of cash, we have a definition of responsibility which is not only precise and unambiguous, but indeed quantitative.
Moreover, this definition solves a second problem: how do we select the controllers? If our controllers are the parties to whom the profits are actually paid, and their voting power is proportional to the fractions they receive, they have not only a shared definition of responsibility, but an incentive to apply that definition in practice.
We have, of course, reinvented the joint-stock company. There is no need to argue over whether this design works. It does. The relevant question is: in the context of government, does this financial definition of responsibility actually match the goal we started out with?
In other words: will an effectively managed government (remember, we are assuming security and effectiveness), which is responsible only in the sense that it tries to maximize its profits in the infinite term, actually provide the good customer service that is our goal? Will it catch mice for us? Or will it flay us, and hang us out to dry, etc.?
As a progressive, you consider undivided government (“dictatorship”) the root of all evil. It is impossible to enumerate the full list of reasons behind this belief. It’s like asking you why you prefer a romantic candlelight dinner for two at a simple, yet elegant, French restaurant, to being dragged alive behind an 18-wheeler at highway speed until there is nothing on the rope but a flap of bloody skin. When we add the abominable and astonishing suggestion that said government should actually turn a profit, we reach maximum horror. But if we are not willing to question even our deepest beliefs, our minds are hardly open.
First, it helps to remember that profitability is hardly antithetical to good customer service. Again, try the restaurant analogy. If all restaurants were nonprofits, do you think we would have better food, or worse? How does a nonprofit restaurant differ from Calmeat, which has no institutional incentive to keep its diners coming back? Perhaps if the restaurant is a small cooperative run by people who really love food, it will continue to shine. California is not a small anything, and my own interactions with its employees have revealed no such passion.
Second, I suspect that your deepest fear about undivided government is that it will in some way prove sadistic. It will torment and abuse its residents for no reason at all. Perhaps, for example, Steve will decide to massacre the Jews. Why not? It’s been done before!
Think about this for a minute. Steve is responsible to his controllers, who evaluate his performance based on his stewardship of one asset: California. The value of California is the sum of the value of its shares. If one goes up or down, so does the other.
Which is worth more? California, or California infested by Jew-eating crocodiles? Which can be made to produce more revenue? The former, clearly. Jews pay taxes. Crocodile dung doesn’t. And from the perspective of either Steve or the Jews, what is the difference between crocodiles and stormtroopers? At least the former will work for free.
Perhaps this is skipping ahead slightly, but one way to understand why Stevifornia will not be sadistic and aggressive is to explain why the Third Reich and the Soviet Union were. Sadism was not profitable for Hitler or Stalin—not that they cared, all that much. But they cared a little. Money meant power, and Hitler and Stalin certainly cared for power.
The sadistic side of these states is best understood as part of their security model. Hitler and Stalin were not gods. They could not shoot lightning bolts or resist bullets. They rose to and stayed in power by ruthless intimidation, up to and certainly including murder. Stalin didn’t kill all those Old Bolsheviks because they had bad breath or had made passes at his wife. In the 20th century’s “totalitarian” states, murder foreign and domestic was an essential strut in the Leader’s security design. We will not be reproducing this element. But I digress.
Third, as a progressive, you think of government as a charitable institution. You think of its purpose as doing good works. And indeed, today’s governments do many good works. They also do many things that are not good works but purport to be, but that is beside the point. Let’s assume that all its good works are good indeed.
Clearly, good works are not compatible with turning a profit. It is easy to see how California improves its bottom line by refraining from the massacre of Jews. It is hard to see how it improves its bottom line by feeding the poor, healing the lame, and teaching the blind to see. And indeed, it doesn’t.
So we can separate California’s expenses into two classes: those essential or profitable for California as a business; and those that are unnecessary and wasteful, such as feeding the poor, etc., etc. Let them starve! Who likes poor people, anyway? And as for the blind, bumping into lampposts will help them build character. Everyone needs character.
I am not Steve Jobs (I would be very ill-suited to the management of California), and I have not done the math. But my suspicion is that eliminating these pointless expenses alone—without any other management improvements—would turn California, now drowning in the red, into a hellacious, gold-spewing cash machine. We’re talking dividends up the wazoo. Stevifornia will make Gazprom look like a pump-n-dump penny stock.
And suddenly, a solution suggests itself.
What we’ve done, with our separation of expenses, is to divide California’s spending into two classes: essential and discretionary. There is another name for a discretionary payment: a dividend. By spending money to heal the lame, California is in effect paying its profits to the lame. It is just doing it in a very fiscally funky manner.
Thus, we can think of California’s spending on good works as profits which are disbursed to an entity responsible for good works. Call it Calgood. If, instead of spending $30 billion per year on good works, California shifts all its good works and good-workers to Calgood, issues Calgood shares that pay dividends of $30 billion per year, and says goodbye, we have the best of both worlds. California is now a lean, mean, cash-printing machine, and the blind can see, the lame can walk, etc., etc.
Furthermore, Calgood’s shares are, like any shares, negotiable. They are just financial instruments. If Calgood’s investment managers decide it makes financial sense to sell California and buy Google or Gazprom or GE, they can go right ahead.
So without harming the poor, the lame, or the blind at all, we have completely separated California from its charitable activities. The whole idea of government as a doer of good works is thoroughly phony. Charity is good and government is necessary, but there is no essential connection between them.
Of course, in real life, the idea of Calgood is slightly creepy. You’d probably want a few hundred special-purpose charities, which would be much more nimble than big, lumbering Calgood. Of course they would be much, much more nimble than California. Which is kind of the point.
We could go even farther than this. We could issue these charitable shares not to organizations that produce services, but to the actual individuals who consume these services. Why buy canes for the blind? Give the blind money. They can buy their own freakin’ canes. If there is anyone who would rather have $100 worth of free services than $100, he’s a retard.
Some people are, of course, retards. Excuse me. They suffer from mental disabilities. And one of the many, many things that California, State of Love, does, is to hover over them with its soft, downy wings. Needless to say, Stevifornia will not have soft, downy wings. It will be hard and shiny, with a lot of brushed aluminum. So what will it do with its retards?
My suspicion is that Stevifornia will do something like this. It will classify all humans on its land surface into three categories: guests, residents, and dependents. Guests are just visiting, and will be sent home if they cause any trouble. Residents are ordinary, grownup people who live in California, pay taxes, are responsible for their own behavior, etc. And dependents are persons large or small, young or old, who are not responsible but need to be cared for anyway.
The basic principle of dependency is that a dependent is a ward. He or she surrenders his or her personal independence to some guardian authority. The guardian holds imperium over the dependent, i.e., controls the dependent’s behavior. In turn the guardian is responsible for the care and feeding of the dependent, and is liable for any torts the dependent commits. As you can see, this design is not my invention.
At present, a large number of Californians are wards of the state itself. Some of them are incompetent, some are dangerous, some are both. Under the same principle as Calgood, these dependents can be spun off into external organizations, along with revenue streams that cover their costs.
Criminals are a special case of dependent. Most criminals are mentally competent, but no more an asset to California than Jew-eating crocodiles. A sensible way to house criminals is to attach them as wards to their revenue streams, but let the criminal himself choose a guardian and switch if he is dissatisfied. I suspect that most criminals would prefer a very different kind of facility than those in which they are housed at present. I also suspect that there are much more efficient ways to make criminal labor pay its own keep.
And I suspect that in Stevifornia, there would be very little crime. In fact, if I were Steve—which of course I’m not—I might well shoot for the goal of providing free crime insurance to my residents. Imagine if you could live in a city where crime was so rare that the government could guarantee restitution for all victims. Imagine what real estate would cost in this city. Imagine how much money its owners would make. Then imagine that Calgood has a third of the shares. It won’t just heal the lame, it will give them bionic wings. But I digress.
So we move on to our third essential: security. (Note that this is Arnold Kling’s objection to the above design, which I’ve given the cute name of neocameralism.)
Security is the art of ensuring that your decision process cannot be compromised by any force, domestic or foreign. Steve, for instance, is entirely indifferent to the opinions of Stevifornians, except inasmuch as those opinions affect his quarterly numbers. This is the ideal “type 3” state: you think what you want, and Steve does what he wants. The government neither controls public opinion, nor is controlled by it.
If nothing quite like a neocameralist government has ever existed in history, the reason is not hard to figure out. How do you secure an intricate decision mechanism like the above? What happens if the controllers decide to fire Steve, and Steve doesn’t want to go? How does Steve remain in power if a million Stevifornians storm the presidential palace, and the guards side with the crowd and turn their guns around?
Fortunately, we do not have to design a solution that will protect Charles X (no relation to Malcolm) from the machinations of the treacherous Marmont. The neocameralist state never existed before the 21st century. It never could have existed. The technology wasn’t there.
Secure neocameralism depends on a cryptographic decision and command chain (CDCC). Once the world has cryptographically secure government, it will wonder how it ever lived without it.
In the world of today, the security of all governments is dependent on mere personal loyalty. The US Army could take over Washington tomorrow, if it wanted to. It certainly cannot be compelled to obey the President, the Supreme Court, the Congress, or anyone else. It so happens that the US military has a strong tradition of loyalty—a tradition that was tested, for example, in the case of the Bonus Army. Would today’s Army fire on an American mob? Especially a mob that shared its political orientation? Hopefully we will not find out.
The only reason that we accept this appalling and dangerous state of affairs is that we don’t know there’s an alternative. But there is, actually—in the form of permissive action links. This is an old Cold War design that implements the command side of a CDCC, for nuclear weapons only. (The control codes are in the President’s pocket.)
In a full CDCC government, the sovereign decision and command chain is secured from end to end by military-grade cryptography. All government weapons—not just nukes, but everything right down to small arms—are inoperable without code authorization. In any civil conflict, loyal units will find that their weapons work. Disloyal units will have to improvise. The result is predictable, as results should be.
Cryptographic command of the military has a critical effect on political dynamics: it makes public opinion irrelevant. Today, even the most militaristic of military despotisms has to invest considerable effort in persuading, cajoling or compelling the public to support it, because the army is inevitably drawn from that public. Witness Marmont, who decided his chances were better with Orléans than Artois.
This is the final blow in the elimination of politics. Men enter politics because they have a lust for power. Good men as well as bad men lust for power, and sometimes it does happen that good men lust for power, seize it, and use it to do good things. But it is more the exception than the rule. And the lust for power is an eminently practical one—if no power is available, no one will bother to scheme for it.
Take Apple, for example. Mac users, such as myself, are tied to its vagaries. For example, the battery for the MacBook Pro is shite. It’s disposable. I believe it may actually be made of toilet paper, chewing gum, and old paper clips. I go through two a year, and I hardly use them.
How do I cope with this appalling injustice? I deal. Why do I deal? Because even if I went on to the right forums and whipped up a screaming mob and persuaded them to march around and around and around 1 Infinite Loop, chanting slogans and burning old batteries, I know that it would have absolutely no impact on Steve’s handling of the problem. (Which I suppose he doesn’t think is a problem at all.) In fact, it would probably make him more stubborn.
There is simply no way for anyone outside Apple to influence Apple’s decision process by the use of force. Apple is not sovereign. It does not have a white-robed black-belt army. It relies on the security forces of Uncle Sam, or at least Cupertino. But the problem is solved, anyway. And I consider this a good thing.
Cryptography applies to the back-end as well: the decision side. If the controllers vote to refuse to renew Steve’s key, and anoint Marc instead, Steve will no longer have command of the army. He won’t even have command of his office door. He will have to call security to let him out of the building. (If you doubt that this is technically feasible, it is.)
Once we realize that 21st-century technology is needed to implement the neocameralist design, we understand why good old cameralism, Frederick the Great style, was the best that previous centuries could do. What Whigs call absolute monarchy (and non-Whigs just call monarchy) collapsed the controllers and the administrator into a single royal person, solving the decision problem quite neatly—and introducing a nasty biological variable into the responsibility mix. And on the command side it relied on loyalty, which was not always there.
Was royalism a perfect system? It was not. But if we imagine a world in which the revolutions and civil wars of the last four centuries had never happened, it is hard not to imagine that world as happier, wealthier, freer, more civilized, and more pleasant. At least if you’re an unregenerate Jacobite like me.