Tuesday, May 29, 2007

Microkernel government

Here are some considerations that went into the design of Juristopia. The overall goal of Juristopia is to improve the most important functions of government (especially defense and the abatement of public nuisances) while preventing the corruption, oppression, war, genocide, and other abuses that have so often come with police powers and taxation. Those evils have been particularly prone to occur when those powers are bundled into a locus of sovereignty, a la the personal totalitarianism of the Justinian Code, Bodin, and Hobbes or the parliamentary totalitarianism of Bagehot. These traditions of legal procedure, assuming political relationships are a matter of delegation rather than of property, have given us almost all of the worst in Western history: the Caesars, the Tsars, Napoleon, the Kaisers, the communist dictators, Mussolini, Franco, and Hitler among others -- based on the profoundly false and destructive assumption, derived from the legal procedure of the Roman Empire, that there must be "one person" who is "responsible" for all politics and law -- a person or (for Bagehot) small organization sitting at the top of a vast pyramid of principal-agent, usually boss-employee, relationships.

Although it discards totalitarian political structure and legal procedure, our proposed form of government is based on historically proven legal mechanisms. With the clarity of legal procedure it avoids the vague nonsense that often passes for political philosophy. Much of the political structure of Juristopia is based on highly evolved common law mechanisms such as property and contract, but these are used in the same basic manner as in the common law, rather than as misleading analogies or mere labels. Nor is there any hand-waving about how "the market" will magically produce efficient law. An efficient market is an outcome, not a cause, of good law. Some kinds of legal competition are good and some are pathological: Juristopia encourages the former and strongly discourages the latter.

Juristopia borrows from a number of historical political and legal systems, ranging from the modern constitutional administrative state to the monarcho-franchise system of late medieval England. Important design themes of Juristopia include separation and decentralization of powers. Its central design theme is the replacement of most delegation relationships (i.e. principle-agent or boss-employee) in politics with property and peer contract relationships. In its radical unbundling of power it satisfies many of the ideals of green libertarianism, minarchism and even anarcho-capitalism, yet it retains and even enhances the valuable functionality of modern government, such as environmental protection and defense. It strongly encourages the healthy legal competition that occurs before any dispute has arisen, such as choice of law and choice of forum, but severely discourages the unhealthy forum shopping that occurs after an injury has already taken place. It thus provides both default jurisdictions and the ability of parties to opt out of them, for future disputes between themselves, by contract.

The main innovations of Juristopia are (1) to revive a kind of jurisdiction that was once a strong bulwark against totalitarianism, namely jurisdictions held as property rather than by delegation, and (2) to combine these with the recently more common bulwarks against totalitarianism, especially separation of powers and a Bill of Rights defining limits on the exercise of franchise powers against individuals.

The result might be called, to borrow some computer science lingo, a microkernel government. A microkernel in a computer is a small program that contains only the most essential functions needed to manage a large number of other programs called "servers" that are conceptually independent of each other. Similarly, our governmental microkernel provides only the miniminum procedural functionality needed to handle franchises, a wide variety of which in turn handle the substantive law. Many of the advantages listed by Wikipedia for microkernel programs point up analogous advantages of microkernel government:

* "security: it is more secure as more operations are done in user mode than in kernel mode" -- similarly, the failure of one governmental function is much less likely to spill over into other functions or to cause general political failure. And since a small franchise requires fewer powers than a large government, any abuses of its powers are likely to be far less severe.

* "reliability: a simpler kernel design and functionality typically results in a more reliable operating system" -- similarly, the compartmentalization produced by property rather than delegation relationships increases the reliability of each function.

* "flexibility: new features can be added and unnecessary ones can be removed. This makes it suitable for both large and small systems" -- similarly, to add or remove a governmental function requires merely defining and funding a new franchise, or for a jury to defund it.

* "portability: most of the processor-specific code is in the microkernel, which makes it easier to port the kernel to a new platform" -- similarly, our new form of government can be applied in a wide variety of contexts -- as a new commercial jurisdiction, as the basis for an online game, or to augment or replace a traditional government.

As a final note, although decentralization is a key feature, this decentralization is generally one of subject matter rather than territory. Territory is far less important in Juristopia than it is in most of today's governments. Its federalism and localism takes the form of unbundling jurisdiction by subject matter rather than by territory. It is expected that franchises will operate over large territories and even globally. The physical boundaries of its jurisdiction will depend on its function and won't necessarily bear any relationship to the boundaries of other franchises. Indeed, Juristopia goes much farther than most forms of government in unbundling territory from government, thereby reducing the legal transaction costs of operating in multiple territories and increasing the healthy forms of legal competition.

(BTW, please let me know which spelling you prefer: :-)


Anonymous said...

I have, since I opened my eyes politically,economically, and philosophically, been truly wanting to theorize new forms of decentralized governments. It's sad that so many people in the US think our form of government is and has to be the best. I hope you continue to expand on this.

Mencius Moldbug said...

I like the microkernel analogy, although I am writing this from an OS that in theory is a microkernel, but in practice is monolithic. Hm.

I find Juristopia to be very much in the Western tradition of benevolently governed utopias, going back of course to Plato.

My approach is much stupider: a country (preferably a city) is a piece of property, just like a ranch. The ranch is owned by someone, a person or corporation, who controls what happens on it.

As Adam Smith told us, it's not from the benevolence of the butcher or the baker that we expect our meat or bread. Nor today is it from the benevolence of Toyota, JetBlue, or Starbucks that we get our cars, flights, or coffee.

Somehow people have failed to apply this lesson to government. They don't expect their cars to be manufactured by a charity. They know what the result of this is: the Trabant. They probably even understand that if the President of Trabant was elected by the East German people, the result would be even worse. But somehow, they still insist on government as charity.

As we speak, a quarter of the world's cranes are in Dubai, which is a government run by an absolute, unelected ruler whose goal is simply to profit off his subjects, who have no right to vote. Moreover, Dubai is a hideous hellhole in the worst place on earth. Yet it sustains a large and seemingly quite contented expat community, which is growing like a weed. Hmm.

All of Western civilization, from about 500 to about 1800, was built by monarchies: privately-owned countries. Then some bright intellectuals had the idea of Government for the People. The result has been one tyranny after another.

Tyranny is actually a form of civil war. Asymmetric civil war to be exact. Tyranny happens when the title of a nation-sized piece of land is unclear and insecure. People will do almost anything to get their hands on it.

The relationship between benevolent government and tyranny is very simple. A benevolent government is one that leaves money on the table. It does not charge its customers for the cost of capital. It does not try to maximize revenue.

Therefore, it attracts attacks from those who wish to divert this revenue - or who merely seek power, as all men do. When there is foregone revenue, they can in effect promise it to their supporters as a payment to be gained once power is seized. This is how socialism works.

Secure government is not benevolent or malevolent, any more than Starbucks is benevolent or malevolent. It runs under the rule of law not because it loves the People, but because law is the most efficient way to maintain order and prosperity on a patch of land.

You can model any non-revenue-maximizing state as a revenue-maximizing state that makes dividend payments to those it would otherwise tax. It is then a Pareto optimization to make those shares real, and make them negotiable. Thus, benevolence is always an illusion.

Why then is it so popular? Simply because it is a way of manipulating people. The central problem of government is maintaining security. As Louis XVI found, internal security is a pretty big part of this. His grandfather, the Sun King, was actually quite an innovator of official benevolence. The idea got a little out of hand, but it was basically sound: convince your subjects that you love them, that you are their father. It is only a short step from here to democracy.

So your view of sovereignty is precisely backward. Property ownership is impersonal. A landowner doesn't care what his tenants do in their spare time.
Paternal benevolence is the relationship of master to slave, and it is this that undermines all these ideas of the benevolent state.

I think my efficient state might have a very similar "operating system" to your benevolent state. But yours would decay over time and mine wouldn't, because mine has a built-in incentive not to decay. Whereas yours has a built-in incentive to do just that.

Efficient government under a simple one-level property model (sovereignty), in other words, is actually balanced - unlike the Montesquieu design which gave us what we have now, which maintains the appearance but not the reality of balance. An efficient sovereign has every motivation to respond to security challenges by precisely the level of response needed, and no more. A benevolent government is likely to compromise, and buy off the attacker with the revenues it is supposedly protecting. The consequences of this are clear.

Mencius Moldbug said...

BTW, I prefer "Juristopia."

Anonymous said...


(1) Adam Smith lived under a system of substantive common law -- property rights, contract, and so on -- that allowed the market conditions he described to occur. It's absurd to propose an arbitrary lawmaker _and_ assume such conditions at the same time. Many other governments, including some owned by shareholders which Smith criticized, did not have good laws and thereby had terrible economies.

(2) Dubai is a special case of a trading center that must continually attract new people -- thus maintain low exit costs. The vast majority of the globe is quite different -- it is full of fixtures and social networks on which wide-area monopolies of force impose very high exit costs. You don't impose any principle to keep governments the size of cities, nor any solution to keep the main fixtures of the world from being extorted.

(3) You don't address the central role of legal procedure and that its rules and incentives are profoundly different from those of the market: see my new post here.

Thanks for the spelling recommendation -- I'm coming to agree with that.

Mencius Moldbug said...


I like your new post and am pretty much in agreement with it. I may go farther in some respects.

Introducing a Ring of Fnarg was exactly my response to your (1). In real life, sovereignty is the top level. A government does not have a bigger flea upon its back to bite it, so to speak. It has to define its own law. The question was: suppose this problem was trivial? It is not trivial, but a half-answer may still be useful.

I am still mystified by your exit-cost argument. Tons of normal contractual relationships have high natural exit costs. Think of what a pain it is to exit even a simple rental apartment. Furthermore, Fnargland has every incentive to offer new citizens a liberal, negatively-defined rule of law with unenumerated rights, and it has no incentive to smash its integration with the global economy by withdrawing that law. Dubai, Hong Kong and Singapore have homes and businesses as well - they are not mere airport rest stops.

The global-monopoly argument (why won't cities merge?) is far more interesting. This was the reason my first example used a global monopoly. If you imagine an expanding world in which new land could be created, the cost of capital becomes the cost of fabrication (perhaps L5 colonies?) rather than the cost of buying the ability to 0wn the Laffer curve.

I suspect that in the end, at least until we make it out of the solar system, it is impossible to escape global revenue maximization. But this is okay, by my book, because we really do "owe it to ourselves" - as long as taxes are actually paid to shareholders, rather than being wasted in unproductive state activity, the efficiency cost is rather minimal. Any tax burden that seriously reduces efficiency should also move you away from the Laffer peak.

So I do actually suspect that all the cities in the world might end up working together as a cartel. But if at least the initial distribution of shares in those cities is equitable, this is by no means the end of the world.

And it is certainly much better, in my book, than creating a vacuum of power that provides the energy for violent movements who want to capture it. This is the point of my purist formalist model - to create a force that counterbalances the growth of the state, and is motivated by more than benevolence. I am still not sure I've really gotten this across.

Anonymous said...


It's far, far cheaper to leave a rental apartment for one across town than to move from one state to another (leave one's friends, and often relatives, behind, and try to convince your SO to give up her job and friends and relatives and come with you), much less move to a new country with different languages and laws.

And often other countries won't let you in. Furthermore, historically more than a few countries have had Iron Curtains making it difficult to get out (late medieval China and the Soviet Union being hardly the only examples of this, not to mention conditions like slavery and serfdom that involved legal impositions of high exit costs on subsets of the populace). This strategy of oppression works best when labor is fixed to the land, but has also been imposed on merchants (medieval China) and others when leaders can satisfy their preferences without a thriving foreign trade, as they often can.

Kim Jung Il is a good current example -- a sovereign and totalitarian hereditary monarch in all but name -- able to extract any degree of profits he wishes from an economy the law of which his father and himself have controlled for many decades. Kim Jung Il is just the kind of total owner-of-all you propose. How, then, do you explain the vast superiority of South over North Korea, in terms of economic wealth, liberty, or almost any other criteria other than satisfying the preferences of a Great Leader?

Only a small subset of our populace can afford to spend their lives as global travellers. Some larger subset "visit" and "leave" remotely by conducting remote trades. It is by specializing in attracting this tiny minorty of globetrotters and the somewhat larger minority of remote traders that incentivize the tiny countries of Dubai, Hong Kong, and Singapore with their unusually good policies. It's ex ante legal competition and it's a great thing (Bermuda in insurance law and Delaware in corporate law are two more good examples).

But that very minority of globetrotters has few local connections and can easily leave. You cite Singapore and Hong Kong and ignore Kim Jung Il in North Korea. The very same totalitarian structures next door to Dubai, in Saudi Arabia, in the old Shah's Iran, and so on have a much lower percentage of mobile globetrotters in their population, and this has led to very different results, sometimes partially relieved by riches from vast supplies of Western-craved oil. It's no coincidence that you pick very special purpose countries and ignore the normal and hisorical dominant ones, as your argument falls apart precipitously as one expands one's vision.

Generally speaking, the best way to maximize exit costs and minimize liberty is to promote sovereignty and its kissing cousin, totalitarianism.

BTW, your argument against my jury-limited franchises is also a an argument against your own idea of government-for-profit. There is nothing in my proposal that suggests franchises can't be for-profit ventures, and I suspect many of them will be. The big differences are that I have juries who can cap the rates franchises desire to charge, and I have legal choice to allow some kinds of opting-out and legal competiton, but you have no such mechanisms. You may argue that juries given your propaganda would be ineffective, that juries will just give franchises what they want, but that outcome is no different than a for-profit government charging whatever taxes it likes. Indeed to the extent people can contractually opt-out (choice of law and choice of forum clauses) the competition and low exit costs in my proposal are far superior to yours. I have two big protections against excess taxation and legal oppression -- juries and legal choice -- and you have none.

Mencius Moldbug said...


Kim Jong Il is in fact an excellent example.

Because his control over his own state is, as far as outsiders can tell, terribly insecure. He is surrounded by a world that considers his rule illegitimate and is no doubt very happy to work with any insider ready to risk Kim's wrath and plot a coup. And in fact there seem to have been quite a few attentats in the last decade or so.

Kim is the anti-Fnargl. Everything in his regime is subordinated to the goal of maintaining internal security. If he relaxed for a second, I'm sure he would end up like Ceausescu.

Does anyone really think this man believes in socialism? He believes in himself. He believes in Swedish hookers and Dom Perignon.

KJI would love to liberalize and set up controlled capitalism as the PRC has. It would certainly earn him a lot more luxury goods. The reason he doesn't is almost certainly that he's afraid to. There have actually been some attempts to set up special economic zones in NK, but these piss off the Chinese for obvious reasons.

The real tragedy is that, due to its fanatical commitment to human rights, the West or South Korea can't simply buy North Korea from its rulers. I'm sure KJI would be delighted as hell to spend the rest of his life as a billionaire ex-King of Korea, mixing it up with the other playboys in Portofino. But in real life, he has only two options: tyranny or death. It is quite understandable that he chooses the first.

As for migration, barriers to entry are the inevitable result of modern democracy. Economic migration, among people who were certainly not globetrotters, has actually been very common in the past - aren't we living in a country founded and populated primarily by economic migrants?

I suspect that if the US opened its borders freely - not that it should - we would get something like a billion people. No, not all globetrotting capitalists.

But the globetrotting capitalists are actually a pretty damned important part of any prosperous economy. As I'm sure you know.

I have no legal protections against oppression and overtaxation, because I am not under the illusion that there is any such thing as self-enforcing law. Instead I substitute economic protections, which are self-enforcing.

Remember that economic growth is exponential. The exponent is small, but anything that screws with it - overtaxation or oppression - is going to be extremely unprofitable.

If you look at past monarchies, every instance of bad government I can think of follows the pattern above. Unprofitable behavior is almost always related to external or internal security issues. Rulers whose security situation is stable - for example, the small German states in the Holy Roman Empire - exhibit none of these problems.

For example, I think it's pretty clear that the travel and trading bans of medieval China and Japan were basically security-related, not an attempt to raise exit costs.

Anonymous said...


Sorry, I didn't buy leftist propaganda during the Cold War and I'm not buying it now. That "but for U.S. pressure" argument has long been been used by leftists to try to explain why every one of the dozens of communists countries miserably failed their people -- why Fidel Castro's Cuba is mired in poverty, why Stalin and Mao and Pol Pot killed tends of millions, and so on. The leadership record of these countries is quite stable compared to that of the wealthier and far more libertarian democracies. Even if we ignore the high instability of "ownership" inherent in frequent elections, there were more successful assasinations in the West. For much of the last century communists countries were quite secure, usually strongly allied with each other agaginst the West -- most allied with the Soviet Union which was invulnerable to U.S. threats behind its nuclear umbrella. The reality for most of the Cold War was that the West threatened communism no more than communism threatened the West. If insecurity were the main cause of oppression, Western Europe should have descended into an orgy of genocide during the Cold War.

And now that Kim Jung Il has nuclear weapons, your theory predicts that North Korea will soon, quite unlike any communist country before it, magically transform into a libertarian paradise that will put South Korea to shame. The reality is that these despotisms, in some cases (as in North Korea) practically equivalent to monarchies, went way out of their way to impose high exit costs, because oppression satisfied the preferences of the leadership. I recommend some reading material on the Berlin Wall and on the Iron Curtain generally. These weren't called "prison states" for nothing.

I'm not of course accusing you of being a red in libertarian drag. But you are going way out of your way to try to conveniently label all the many examples of highly oppressive and destructive monarchs or near-monarchs as "insecure." A ruling family that has been in power since the 1950s, that successfully passed the torch from father to son, has always been so extremely insecure so as to starve its people while the same culture to the South turns into wealthy capitalists? Baloney. Their ability to oppress their peoples to such an extreme degree comes from their very security. Historically Kim's family has had a far more stable rule than any ruler of Hong Kong (which in any case hardly has had "clear ownership" and thus quite violates your theory) or (a better example for you) Singapore. Thailand, Taiwan, Japan, South Korea, and any number of other Asian countries are far less politically stable and far more prosperous than North Korea.

The main difference between Singapore and North Korea, as I pointed out before, is what the government specializes in. You've conveniently picked the three current ones with very small polities that specialize in a way only about three small countries on the planet could -- a very small political niche -- as world trade hubs, for which maintaining low exit costs is, quite unusually for a government, incentivized rather than disincentivized for the controller(s) of police powers.

I'd be far more impressed if you came up with an example of a libertarian monarchy in an oil-rich or agriculture-dominated country. You can't do it. Heck, I'd be happy to see you come up with an example of libertarian monarchy in an industrial country.

Of course I am not basing microkernel government on anything like "self-enforcing law." It's a rather bad reflection on you that you would dismiss such a well-research design, with its many carefully designed checks and balances, in such a cavalier fashion. Preventing the rise of "self-enforcing law" is what the real political structures you are ignoring prevent -- separation of powers, franchises, and so forth Real mechanisms, not magical hand-waving about Adam Smith and "the market" and how communist rulers would magically turn into libertarians if only they'd pay dividends to Party members.

I'm glad you've read Wealth of Nations as far as the famous unbenevolent butcher of Chapter 1, but you'd do well to keep reading. The good Professor Smith was under no illusion that for-profit governments act like companies in a free market. Here is what Smith, who unlike us moderns was able to observe many privately owned governments in his day, had to say about privately owned governments:

"[The small Danish colonies of St. Thomas and Santa Cruz] were under the government of an exclusive company, which... had not only the power of oppressing [the colonists], but a great temptation to do so. The government of an exclusive company of merchants, is, perhaps, the worst of all governments for any country whatsoever." (pg. 722 in the Bantam Classic 2003 edition).

There is much else in Chapter 7 along these lines. Smith also has some very good observations about the cheapness of land and remoteness from central government power as enabling liberty. His observations parallel my comments about exit costs and liberty. I'd love to see you do a blog post on Chapter 7 of Wealth of Nations and try to rebut Smith's arguments, unless (I hope) you are convinced by them instead.

Mencius Moldbug said...


I am not sure how I could be clearer, but perhaps the problem is that I used the phrase "internal security" without defining it.

Internal security is security against residents of a country who don't control the state and want to.

No quantity of nuclear bombs will provide internal security to Kim Jong Il. In fact, I am sorry I mentioned the fact that his internal enemies could summon external support. Because, as you say, it is too reminiscent of Kim's own propaganda.

If you read a little about NK (the One Free Korea blog is good) you will see how insanely paranoid Kim's internal security model is. Or at least it strikes us as insane. I suspect, however, that it is in fact quite rational.

Communist countries were quite secure against foreign invaders. Communist rulers were not at all secure against other members of their own party, because the party had no formal model of succession as did monarchies. Why do you think Stalin purged the Old Bolsheviks, and the other intelligentsia? I have recommended it before, but I think Sebag-Montefiore's Court of the Red Tsar is invaluable reading for anyone who wants to understand the Stalinist phenomenon. All of Stalin's direct reports were terrified of him, and all spent a substantial amount of their time trying to figure out how to get rid of him. It was a stable system in a sick kind of way.

What has happened in the PRC is highly illustrative. There is no comparison at all between the post-Mao leaders, even Deng, and Mao. This is because the PRC has developed a highly formalized succession model, vaguely similar to that of the medieval church.

Hu Jintao is not a dictator - he has a scheduled retirement coming up. Jiang Zemin actually tried to retain some power after he was supposed to retire. He failed. China is in the early stages of a transition from lawlessness to law, and the benefits are obvious.

Today, the Chinese Communist Party is stronger than ever. And China is as free as it has been since the Party captured it. These points strike me as inarguable.

What is your explanation of this transition? Do you have a different view?

For all the reasons Rothbard etc. has cited, I am not a fan of Adam Smith. I used his quote because it was well-known, not because I wanted to cite him as the Bible.

The big picture is that every system of government in the world in Adam Smith's time was far less oppressive - by the definition of Smith's own time - than every government in the world today.

I have not read Smith, and I'm not sure what the Danish merchants were doing that pissed him off, or why. Rothbard has done a good enough job of convincing me that Smith's ability to untangle cause and effect was distinctly limited. However, he was a man of his time and if he could see DC today, he would take back everything bad he ever said about Louis XIV.

And this is the outcome of two centuries of Montesquieuian self-balancing designs. Can you really be so surprised that I question the entire premise of this line of development?

Of course the experiment is not controlled. But neither were Smith's observations. This is why I prefer to work, like Mises, deductively from first principles.

First principles tell me that these so-called checks-and-balances designs are not in fact balanced, so I am unsurprised to learn from my history books that they have.

Anonymous said...


"China is in the early stages of a transition from lawlessness to law, and the benefits are obvious."

I'm glad we both recognize the importance of rule of law rather than arbitrary rule. And as your China example suggests the most important parts of the law are not the substantive but the procedural, including but hardly limited to the laws of succession to political powers.

"First principles tell me that these so-called checks-and-balances designs are not in fact balanced, so I am unsurprised to learn from my history books that they have."

This strikes me as the crux of the problem: you're being severely misled by the gross overextension of some simple set of first principles. Intersubjective reality is not compressible to such simple forms without severe loss of crucial bits. The praxeological axioms work great for most macroeconomics but are not much good for anything else, and certainly not for involuntary or otherwise very-high-exist-cost transactions. If you don't believe me try reconciling the praxeological axioms with the blog post I just wrote on legal procedure, which you said you agree with.

Indeed, any simple set of first principles in intersubjective matters must always be highly oversimplified. Even the corporations you laud have checks and balances, some of them rather similar to governmental ones. As an antidote to ideology I strongly recommend a good inductive dose of the history of institutions instead.

For starts I recommend a some posts of mine explaining checks and balances in corporations:

Separation of Duties
The Principle of Least Authority

Then I recommend the Wikipedia article on "Subject Matter Jurisdiction", which will help shed light on my writings on jurisdiction as property. (Subject matter, not territory, form the most important "boundary lines" of jurisdictions).

Then I recommend Controlling the State by Scott Gordon: all about the history of checks and balances in corporate-like governmental structures, i.e. republics, from ancient Athens to Venice (which had an extremely interesting model) to the U.S.

Mencius Moldbug said...


The existence of force does not negate the effectiveness of logic. Nothing does and nothing can.

Reason does not have to be compressed. And it applies just as well within a framework of law - which, as you note, is always imposed by force - as it does outside it.

More on this on my blog, in a bit.

The corporate influences on constitutional history are clear. But they strike me, frankly, as of the "cargo cult" variety.

Various people have tried to design effective systems of government by copying private social structures which clearly worked. Rousseau's "social contract," which of course is not a contract at all, is an even simpler example. It is not clear if the designers who chose corporate titles, such as "president," really understood why corporations work. And it's pretty clear that the result has not been a success.

At least I think it's pretty clear. I am really not sure how a libertarian can believe that the art of government is a solved problem in any way, shape or form. History to me is a depressing list of solutions that looked like they might work but didn't, punctuated by occasional flashes of extremely incomplete and unstable success.

Venice and Athens are interesting - as failures. You might have noticed that the Serenissima is no longer with us. It is edifying to look at these patterns of failure, but you don't build a 747 by strapping wings to your back and leaping from a belfry.

This is why the modern corporate model - which I have seen functioning, as intended, in practice - interests me. Because it works. It constitutes government-sized organizations which fulfill exactly their intended purpose.

It also delegates security, internal and external, to the state. And it itself, in its amoral desires, is restrained by the state. So the answer is not trivial. But as a place to start from, it strikes me as incomparably underestimated and vastly more interesting than the long, dreary record of failed republican designs.

I really, really, really recommend Jouvenel's On Power and Burnham's The Machiavellians. If nothing else, any skeptical look at the small-r republican tradition from a stance outside it is worth its weight in gold.

Anonymous said...

MM, you may have misunderstood what I meant by compression. All deductive systems, whether those of Euclid or Mises, have a very small Kolmogorov complexity -- it's just the complexity of their axioms. If you don't understand this go read up on Kolmogorov complexity. It is the existence of far greater complexity in the real world that negates the ability of simple logic to describe the intersubjective world (economics, politics, etc.) The complex world cannot be compressed with negligible lossiness into any simple axiomatic system.

The use of axiomatic deduction to create "theories of everything", outside physics, is a pernicious species of totalitarian thought that has led and will lead to totalitarian politics. The entire complex world is compressed into a hypersimplistic system and most of one's ability to know the social world is lost as a result. Far worse still when it such thought systems are forced on others. Hegelians, Marxists, and many other mental pathologies have gone down this route, and now we may have the modern abuse of praxeology to add on.

The development of medieval and modern republican governments out of the corporate form is no mere matter of "cargo cult" or the like. Medieval city-states and boroughs were usually legal corporations that used the corporate law of the time as their default law. And there were many intermediate forms between the purely commercial and the purely governmental.

You persist in the modern habit of putting "government" in one box and "private" in a completely separate box, but in the Middle Ages and Renaissance there were all sorts of real intermediate forms. They were heavily borrowing real law and corporate controls from each other, not mere distant imitations.

As for Montisquean separation of powers, they worked quite well until the 20th century, and during the 20th century they worked far better than their main alternative, communism. I agree that by themselves they are not the solution. That is why I have added jurisdictional unbundling and legal choice. But I hardly agree with your extreme notion of tossing every single hiotorically valuable mechanism to control power out the window in favor of some vague notion tht "the market" will solve all problems of involuntary power. Really, I have not heard anything so naive since the idea that "the dictatorship of the proletariat" will lead to "the withering of the state," and I'm getting quite impatient with you persisting with such utter silliness.

Mencius Moldbug said...


I still see scientism, or I guess mathematism, in your attempt to import mathematical concepts into human affairs. You can't reasonably refute reason.

For example, Mises' definition of rational action as revealed preference is (like any definition) irrefutable. It is a mental tool.

Of course it can be used to oversimplify the world - but so can equally true mathematical truths. Your use of Kolmogorov complexity in human affairs reminds me of the ways people used to invoke Einstein or Heisenberg.

Mises' "axioms" are not really axioms, any more than a "cable modem" is a modem. This kind of metaphor was fashionable in his time and so he used it. Today I'd like to think he'd be the first to deplore it. Rhetoric is not math.

You are on firmer ground, though, I think, in taking umbrage at my caricature of corporate forms in government as mere imitations. Certainly our modern separation of government and corporation evolved from a wide variety of organisms which were not easily categorizable in today's terms.

But the result was as I described: terminology that was originally legal and formalist, as in the case of municipal corporations, was left as a kind of evolutionary appendix within democratic structures. The success and survival of the true joint-stock corporation was not anyone's plan: it was the result of the fact that this, unlike its competitors, was a praxeologically stable structure.

Montesquieuean checks and balances certainly had failed by 1861. Arguably they had failed by 1828. Jefferson in 1825 saw very clearly how it would go down.

As for tossing things on the garbage heap, you appear to have relegated the entire tradition of monarchist/legitimist thought to that pile. Your entire playbook seems to come from authors who are republicans, democratists, or both. Of course in a democratic republic, and one which controls its own educational and academic system, this is entirely to be expected. But given, again, the level of government failure in our society, I think it's worth looking farther. I am not endorsing Filmer, Hobbes or even Maistre, but I suspect a glance at future history would have surprised them less than it would have surprised Rousseau, Locke or Mill.

I don't at all think "the market" will solve the problem of involuntary power. I think it will create an involuntary power to balance the involuntary power of those who would otherwise gain by expanding the state.

But it is not fair of me to try and make this case in your comments section - I should explain more coherently on my own blog what I'm talking about.

Anonymous said...

MM, I don't recommend playing the authority game. Since I'm the author of the leading paper on the monarcho-franchise legal structure of medieval and Renaissance England, you at least grossly exagerate when you suggest that I've studied only democratic or republican authorities. The parties and judges quoted in the Selden case reports were quite thoroughly monarchical. It is you who seem to have almost no familiarty with non-totalitarian monarchical forms (such as monarcho-franchise and king-in-Parliament).

I'm also quite familiar with the totalitarian monarchical tradition of the Justinian Code, Bodin, and Hobbes, of which your philosophy seems to be but a faint reflection. I am furthermore familiar with Olson's theory of the stationary bandit (a great explanation of monarchy and the "economics" of the use of force which I highly recommend). I'm also familiar with Hoppe's time-preference theory of monarchy which you seem to be citing -- and which by itself predicts that North Korea should be more stable and libertarian than any democracy. (Unless, of course, we engage in your circular reasoning: theory predicts that oppression comes from insecurity; Kim Jung Il oppresses his people, therefore he must be insecure, therefore his dynasty is not a counter-example to the theory).

Mencius Moldbug said...


Who, me? I'm an authority on nothing. I have no credentials or credibility.

All I'm saying is that you appear to subscribe to the Whig theory of history. You write about the monarchical period, but you always take the side opposed to the monarch. To a Whig by definition everything before republicanism is a predecessor of republicanism, and this indeed seems to be your take.

It is exciting and interesting to see the barons of medieval England depicted as proto-cypherpunks. I agree with you that we have much to learn from this system of law which was so effectively decentralized. But fundamentally, these guys were not libertarians - they were a bunch of vicious thugs. Kings and barons alike. The decentralized legal structures of medieval England were a product of military reality, and they disappeared when that reality changed.

I would like to see a return to decentralized law. Simply because I think it works better. But we cannot turn time back to the era of knights, archers and pikemen, and I think that if you're interested in designing systems that work in the modern world you have to be very careful that you're not inadvertently depending on this variable which no longer exists.

I am indeed citing Olson and Hoppe. You seem to disagree with them, especially the latter. But you don't seem willing to explain why. When you cite North Korea, I give you a pointer to what I think is going on in North Korea. You don't seem interested in refuting this characterization, either through inductive or deductive means. The same for my characterization of the PRC. And you also don't seem interested in engaging with the Continental antirepublican tradition of writers like Jouvenel and Kuehnelt-Leddihn, or their 19th-century predecessors such as Maistre and Taine.

Obviously in all of this you are in the majority. Most people would probably find your views extreme and unusual, but they are mainstream next to mine. Virtually no one defends the positions or writers I'm trying to defend.

I'd like to think that would make my perspective extremely nonthreatening, compared to the task of arguing with a socialist. There is no way I can paint your views as marginal or heap demotic ridicule on them. Au contraire! I would just like to think that, since you have already decided to think for yourself about these matters rather than accepting the conventional wisdom, my even more unconventional views would stimulate curiosity rather than outrage.

Mencius Moldbug said...

BTW, in case you're interested in the internal stability of the North Korean regime, this query is a good start.

I also recommend Chang and Halliday's book on Mao, and Sebag-Montefiore on Stalin.

The reality of totalitarianism is much more interesting than the stereotypes. Kim is not Stalin is not Mao, and neither of them bears much resemblance to George III, who in turn was not much like Louis XIV. I think the temptation to lump them all together in the museum of non-republican regimes is very much to be resisted.

Anonymous said...

"...you appear to subscribe to the Whig theory of history... You write about the monarchical period, but you always take the side opposed to the monarch."

It was quite popular in the 20th century -- especially among the same people who ignored the crimes of communism -- to be snide about the Whig theory. But there is much of value in it. It is still the most mature and realistic libertarian theory that exists. Like much else it has been buried by the Progressive revolution, so I highly recommend reading what the Whigs themselves said (e.g. Trenchard and Gordon in Cato's Letters) rather than the idiotic slanders you might have read about them in 20th century Progressive tomes.

The main thing wrong with it, on which you and I have a lot of common ground, is the way it overly strengthened Parliament. Like most ideologies it discarded too much of tremendous value -- especially it discarded the mature Anglo-Norman model of political property rights in favor of a republican version of the Romanist delegation model. By this route it evetually morphed from distrusting power into an apotheosis of government in the guise of "the people." (And that, BTW, besides the factors Smith cited, is the main reason Progressivism took eventually took over).

Thus unlike the Whigs I am quite a fan of monarcho-francise England and to some degree (despite my criticisms and warnings) of some kinds of colonial government. Our big differences are

(1) I am not willing in an ideological rush to throw out the Whig babies with their dirty bathwater. The role of highly evolved corporate controls like separation of powers and the principle of least authority, for example, remain crucial.

(2) that I am much more of a fan of the forgotten franchise part and you are a fan of what the Romanists recognized, namely the sole monarch. Your view of it seems to be the Romanist stereotype of "feudalism", that medieval government really was just a military hierarchy. This is quite far from the case, especially in England that was very secure from invasion after the Conquest. While the ownership often did have origins -- several hundred years back -- in the Conquest, relations after that time were governed almost entirely by property law -- a largely peer-to-peer structure which is almost the exact opposite of a master-servant military hierarchy. But again this is an idea lost to the modern world and I don't expect folks to pick up on it quickly, alas. I can only just urge readers to again Wikipedia "subject matter jurisdiction" and then Google "jurisdiction as property" and read carefully. There is over 900 years of Romanist brainwashing (about 500 years in the English-speaking world) to overcome in order to understand this stuff.

All modern political scientists. including all libetarian and "anarcho-capitalist" ones, suffer under Romanist fallacies, so one cannot take any of them too seriously. Olson is probably the one I agree most with, but largely just on basic theory since like all political science the rest is quite incomplete and speculative. I don't disagree with Hoppe in principle, but one thing he seems to have missed (and you definitely missed) is that the same argument applies to political property rights decentralized by subject matter rather than by territory. That is the key insight you are missing.

Furthermore, these time preferences start out extremely weak for a new regime. They are even weak compared to other forces for at least the first half-century of monarchy as the Kim Il Jung family demonstrates. Even if monarchy is great in theory how do you get from here to there? You don't have to believe in the Singularity to recognize that our cultural norms are changing much faster than in the Middle Ages and that our children cannot be guarunteed the same political rights we wish to devise to them. We no longer live in a world of long time horizons, particularly in politics.

Mencius Moldbug said...


Don't get me wrong - I am entirely uninfluenced by the Progressives. Or I'd like to think I am, anyway. I would love to turn the clock back to Cobden and Bright, Cato's Letters, etc.

My problem with Whiggery is very simple and practical: the Whigs lost. Moreover, they lost not as the result of a single military or political battle, but as the result of a long and continuous struggle that played out in the same way in all Western countries.

I find it laudable, but incredibly unwise, that libertarians want to refight this battle without any real theory of what happened and why. Talk about the triumph of hope over experience.

At the very least, Burnham's "Machiavellians" - Pareto, Michels, Mosca, and Sorel (who I would replace with Jouvenel) are absolutely critical to this debate.

I think you're mistaken in equating the Machiavellian view with the Romanist view. Machiavelli himself was certainly no Romanist, and in fact, Burnham starts The Machiavellians by delivering a savage reaming to none other than Dante, whose De Monarchia is certainly in the Romanist tradition.

Unless you can find a way to connect the Romanist-absolutist and Progressive traditions - I suppose one would have to go through Hegel for this, although how you get from Filmer to Hegel is a mystery - I think it's imprudent to describe anyone today (let alone me!) as a Romanist. There is such a thing as parallel evolution. A bat is not just a bird that flies at night.

The real gift of the Machiavellians was that they looked at the world as it was, not as it should be. Not that they didn't think it could be improved, but that when doing so, one has to "start from here."

Because what happens if the structure of power is not aligned with the structure of law? You alluded to this, I think, in your post on force. What happens is that the law is destroyed. Law does not enforce itself.

Therefore, in order to advance from the informal to the formal, we have to look at the structure of power as it is today, and formalize that. This structure may be suboptimal - in fact, it is suboptimal. But formalization is about turning power into property, not about finding some magic ring that can defeat power without being abused.

I think this is the forest you're missing for the trees when you look at the medieval franchise system. Franchise law existed because it was a recognition of political reality. The barons were powerful. They enforced their own law, whether the king wanted to let them or not. This is why seisin is such a big issue in this period. The Romanists were doing exactly what I recommend against - they were using an invented system of law as a weapon to try to change the real distribution of power. The only reason this appears to have worked is that the real distribution of power was, in fact, changing, and Romanist theories of absolute monarchy just surfed this wave.

The situation now is very different. The US is a corporation that owns a continent. I think this is terrible. I strongly disapprove of it, and I view it as nothing less than the profits of crime.

But it was also a crime for William I to invade England and redistribute its land to his thugs, who became the barons whose rights you're so keen on. Franchise law, by ratifying the result of this crime, made everyone's life better.

The US does not have absolute power in the parts of North America that it owns. The powers it holds in seisin are limited. And so must be its property right if power is to be converted into property. But formalization is a fragile process, and any attempt to combine it with an actual redistribution of power poisons the reaction.

Once power becomes property, it can be managed precisely. It can be divided, not just geographically but also, as you suggest, horizontally. I have no objection to this. In fact, I don't even have the expertise to quarrel with the divisions you propose. They sound very sensible to me.

But the problem now is that we have a situation where the US de jure is one thing, and the US de facto is something entirely different. Jure and facto must align, and it is always easier to adjust the former.

The glory of law, the reason this design pattern keeps reemerging, is that once you get formal power and actual power to be the same thing, they tend to stay that way. The power of law by itself, the Schelling-point effect of well-known rules that everyone can agree on, is not strong. It is a Van der Waals bond, not a covalent one. It is enough to keep the pencil balanced on its point. It is not sufficient to get it there.

Thus my LBO-type scheme. The whole art of formalization is to gore as few oxen as possible. People still have to be convinced that it's a good idea, but they don't have to be forced to sacrifice.

I strongly believe that no matter how atrocious the distribution of real power in society, even if it involves actual slavery, law is always better than non-law. Slavery is not good, but legal slavery is better than illegal slavery. (Think how much better off both North and South, white and black would have been, for example, if the North had bought the South's slaves rather than fighting for their freedom - okay, they weren't fighting for freedom, they were fighting for the Union, but that's a whole separate argument.)

Mencius Moldbug said...

BTW, I should mention that the idea of law as a Schelling point is due to Daniel Nagy. Not that I hadn't thought of it before, but he mentioned it to me before I mentioned it to him, so I need to credit him if even in this sly and underhanded way :-) I imagine others have had the same thought as well.

Anonymous said...

"Unless you can find a way to connect the Romanist-absolutist and Progressive traditions..."

Allow me to clarify my terminology. I've been using "Romanist" in the broad sense of anybody who believes in sovereign power (in contrast to severe federalism, franchises, taking the law into one's own hands, and so on). Here's a breakdown of subcategories of Romanists:

A Byzantist believes that all sovereignty -- the zenith of all executive, legislative, and judicial power -- lies in a supreme leader (in Byzantium, that was Justinian and the other emperors).

A Cromwellian (I've been using the term "Bagehotist") believes that sovereignty lies in an elected body. Jefferson and Franklin were also close to believing this.

Both kinds of Romanists in their pure form believe in arbitrary legislative powers. In contrast the old Anglo-Norman Parliament restricted its statutes to certain categories: generally grants of property (including political property, a.k.a. franchises) and treaties. In Juristopia the Board of Directors may only grant political property rights -- it may neither set taxes nor engage in other kinds of legislation (I neglected treaties). Also in contrast is the corporate system (like the U.S.) where statutes are by-laws that may not conflict with the articles of incorporation. Also in contrast are natural law restrictions on legislation, a la Locke.

There are probably many Romanist connections to Progressivism, but the most obvious to me is its deprecation of both charter rights and traditional natural rights in favor of a Cromwellian assertion of arbitrary rights to legislate in the "economic" sphere. Thus Footnote Four.