In the Ayn Rand subreddit, a Reddit user posted a link to an article allegedly refuting anarcho-capitalism. They wrote:

Ayn Rand was pro-state, when that state was limited to protecting individual rights. But she never wrote an in-depth treatise specifically addressing anarchism.

Many allegedly pro-liberty people today think that a state--by definition--cannot be limited to protecting rights. This essay explains in detail why these people are wrong, and why only a proper state can do well at protecting liberty.

My first reply is a criticism of Rand’s thoughts on the matter as I understand them. I also respond to the article itself further down. You can see the resulting discussion on Reddit.

---

Rand wrote a poor criticism of anarcho-capitalism in a chapter called ‘The Nature of Government’ in her book The Virtue of Selfishness. It starts:

In unthinking protest against this trend [into lawlessness and tyranny], some people are raising the question of whether government as such is evil by nature and whether anarchy is the ideal social system.

Many anarcho-capitalists/libertarians have given this a great deal of thought – they are far from “unthinking”. She should have steel-manned their position instead of dismissing it out of hand.

Anarchy, as a political concept, is a naive floating abstraction: for all the reasons discussed above, a society without an organized government would be at the mercy of the first criminal who came along and who would precipitate it into the chaos of gang warfare. But the possibility of human immorality is not the only objection to anarchy: even a society whose every member were fully rational and faultlessly moral, could not function in a state of anarchy; it is the need of objective laws and of an arbiter for honest disagreements among men that necessitates the establishment of a government.

These objections are essentially on the same level as those raised by people new to anarcho-capitalism, such as ‘Who would build roads without government?’ In other words: they’re an implicit confession that they haven’t given the issue much thought, since these questions can be and have been answered.

Why couldn’t private arbitration companies provide objective laws for arbitration of honest disagreements? Rand attempts an explanation:

One illustration will be sufficient: suppose Mr. Smith, a customer of [arbitration service] A, suspects that his next-door neighbor, Mr. Jones, a customer of [arbitration service] B, has robbed him; a squad of Police A proceeds to Mr. Jones’ house and is met at the door by a squad of Police B, who declare that they do not accept the validity of Mr. Smith’s complaint and do not recognize the authority of [arbitration service] A. What happens then? You take it from there.

She implies that they could never resolve their conflict – or worse, that they would be in a perpetual state of war – because they don’t have a shared jurisdiction, an underlying legal framework. This is a mistake. I have identified this mistake as an instance of what Karl Popper called ‘the myth of the framework’ – more on that in this article of mine.

Many libertarians, notably David Friedman, have explained (for example here and here) how situations like the one Rand describes could be resolved peaceably. For example, what Popper calls the ‘logic of the situation’ in other contexts enforces arbitration agreements in the absence of government: arbitration services are repeat players who want to make money and have a reputation of trustworthiness and reliability.

Also, the criticism that arbitration services would be in a perpetual state of war actually applies to the present reality of governments being at war with each other. War is expensive, so, as private and profit-seeking companies, arbitration services would avoid war as much as possible, whereas governments can tax the bejesus out of their citizens and/or print money to finance even the longest and bloodiest of wars. In other words, the risk of war is much greater when governments hold coercive legal monopolies over different territories than when voluntary arbitration services exist. A common libertarian saying goes: any criticism of libertarianism is really a criticism of the status quo. This case is no different.

Additionally, a common libertarian objection to Rand’s stance is that countries are already in a state of anarchy with each other. They don’t submit to a shared world government, and yet international diplomacy already works (and for the cases it doesn’t, I preemptively refer again to how governments have a much greater incentive to be bad actors than voluntarily funded corporations could have). Her scenario does happen in real life when crimes are committed across borders and law-enforcement agencies from different countries collaborate despite a lack of shared jurisdiction. Solutions to this common problem already exist.

So the distinction is arbitrary: why should governments be able to act as sovereigns and engage in diplomacy, but their subjects shouldn’t? It takes objective knowledge to devise rules for peaceable interaction and conflict resolution, sure, but why would the creation of such knowledge require a special group, government, and not just many rounds of conjecture and criticism by anyone with good ideas? Why would the source of that knowledge matter?

Objectivity is a consequence of such values and virtues as consistency, predictability, dependability, rationality, conscientiousness, and honesty, and private arbitration services can embody these values and virtues just as much as government, if not better. Anyone advocating the view that only government can create objective knowledge (legal or otherwise) better make a convincing epistemological argument why such knowledge can only spring from government and not any other group of people or even individuals.

Rand’s argument takes on a form similar to that of gun control: as others have pointed out before me, those in favor of gun control don’t actually want no guns. They just want government to have all the guns. Rand’s take is essentially the same logic applied to diplomacy instead of guns: she wants diplomacy to be a privilege reserved for governments only, who (weirdly, primitively, anachronistically) enforce laws based on territorial control rather than reason.

Lastly, how would an objectivist society not eventually turn libertarian anyway? In an objectivist society, government cannot aggress upon its citizens. That means citizens would be free to start their own private arbitration services and use them instead of the government’s legal services. In other words, the government would have competition, resulting in just the types of situations Rand considers to be (but really aren’t) reductions to absurdity. In the realm of politics, how is objectivism not already a libertarian philosophy?

Rand said never to compromise on basic principles, but limited government is just such a compromise. You have to go all the way: libertarian. The view that people couldn’t live without government is cynical and infantilizing.

I see Yaron Brook cowrote the book you mention. That doesn’t bode well. He was in an episode of Lex Fridman’s podcast, along with libertarian Michael Malice. As I recall, Malice addressed most (all?) of Brook’s criticisms, but Brook was not able to address nearly all of Malice’s. But did that change Brook’s mind? No. Shouldn’t it have?

---

I also wrote a criticism of the article itself. For some reason, Reddit won’t let me post it, but here it is:

---

First of all, thanks for taking the ancap view seriously and taking the time to address it. Here are some thoughts:

You claim that the present state of coercive legal monopolies also known as governments is the result of a free-market failure. Even if that were true (I doubt it since there has never been a free market), the (Popperian) ancap argument is not that markets never fail but that error correction is easier in free markets. Following Karl Popper, we shouldn’t judge institutions by their prophetic ability to guarantee preconceived outcomes, but by their ability to correct errors. (For clarity, Popper was not an ancap, but I think given his epistemology he should have been.)

So people can choose to have a defense agency that permits the mutilation of babies’ genitals [...]

Like circumcision under present governments?

[I]t is the initiation of force, regardless of intent, that is destructive to mankind and is to be avoided.

I agree. Even force that is sanctioned by the victim is still immoral.

By what standard are we to judge what makes a good governmental agency? By what is best for human life and happiness. This is presumably the reason any honest and decent AnCap would advocate for his system over fascism or socialism: Because he holds his system to be better for human life in general–and specifically, his own.

Well, I’m an ancap, and far be it from me to tell people what is best for their life and happiness. I instead refer again to Popper’s yardstick: which system is better at correcting errors? For example, a market optimizing people’s happiness might turn out to be a mistake, and it should be able to correct that mistake. Fascism and socialism, at their core, are wrong because they do not allow for error correction (at least not without bloodshed). I’m not even sure that fascists and communists would disagree that life and happiness are worthwhile goals.

An objectivist government may not actively thwart error correction like fascism and socialism do, but the focus on specific goals (life and happiness) over the methodological goal of having the ability to correct errors without bloodshed is still a notable parallel. Like, you criticize coercive governments for using force “for people’s own good”, but a government that enforces a legal monopoly to optimize life and happiness sounds like the same mistake. And what happens if one of the government’s subjects disagrees with the government about what maximizes his happiness? Or maybe he’s not interested in maximizing his happiness, maybe he’d like to live his life differently, what happens then? Does he get to disagree?

[A]n Objectivist Ideal Government (OIG) would collect taxes from no one.

I could be wrong, but I thought an OIG would still collect voluntary taxes. You later write than an OIG “does not even collect compulsory taxes” (emphasis mine), so maybe that’s what you mean, too.

I’d be interested to know how an OIG would finance itself if people chose not to pay it.

If a gross over-punishment is carried out for a small initiation of force by an individual, the excess punishment effectively becomes an initiation of force.

Spot on. I’ve had similar thoughts around how excess punishment would be arbitrary, and one of the key tasks, if not the key task, of law is to prevent and address the arbitrary in human interaction. If the law itself becomes arbitrary, then all bets are off.

Apart from optimizing human life and happiness, you list as requirements for the ideal government (again, I’d focus on error correction rather than trying to set up an ideal, but that aside):

1. It follows proper standards for the protection of life and property.
2. It implements proportional punishment (more generally, non-arbitrary laws).
3. It has presumption of innocence. (Not a lawyer but it’s notable that US law currently only has presumption of innocence in criminal law, where guilt has to be proven beyond a reasonable doubt – but in civil law, it’s merely the preponderance of evidence that matters, ie the evidentiary standard is merely: ‘more likely than not’.)
4. It protects children, whom you describe as “not yet fully independent, rational agents”). (Adults are rarely fully rational either, by the way, nor do I think that one’s rationality is ever completely developed in the sense that there could be a level beyond which rationality could never be improved.)

You then say that an OIG would meet these requirements, so any deviation from an OIG would be a threat to human life and thus an initiation of force. That’s a logical mistake. Any other government that meets these requirements while otherwise deviating from OIGs (eg by not requiring a monopoly over some territory) would be fine, too. So the rest of that section doesn’t apply. Worse, you then set up a situation where the primary offense of a crime is not that it violated someone’s rights but that it defied the authority of government.

Next, you give two examples that result in injustice due to compromises. I agree compromises should be avoided, especially compromises on basic principles, as Rand argues. The problem with those specific examples is that they are like a challenge to ancaps: the implicit argument is that there are problems that ancaps haven’t solved yet. (Essentially the same underlying logic as when people ask how roads would work in an ancap society.) I suspect that what you call private governmental agencies (PGAs) would be very good at finding actual solutions over compromises since they want to be known as agencies that can deliver satisfactory solutions. They would have predefined agreements with each other about what to do in a given situation, and would refine and improve these policies over time (governments already do this when they engage in diplomacy and have things like extradition agreements). Again, rather than seek specific solutions ahead of time, ask which system would be better equipped to correct errors, ie come up with solutions to problems. Creativity, collaboration, solving problems are all traits that PGAs would have an economic incentive to be good at.

You then describe the dynamic between an overwhelmingly powerful PGA and an underdog PGA, where the powerful one would essentially have an incentive to at least bully the underdog into submission or even start an all-out war to wipe out competition from the underdog. I think you underestimate the resulting reputational damage and the cost and unpredictability of wars. You also don’t seem to consider that the underdog could partner with other PGAs to protect itself. Once again, governments already do this.

On the differences between retaliatory force and production: you write that force, including defensive/retaliatory force, doesn’t create value. You say it either destroys value or protects value from other force. Please explain why people pay for personal protection. Like, how is such protection not a value?

Clients can attempt to fire the company, but the company has a greater coercive power than the client, so a corrupt company can blockade or imprison the individual and prevent him from hiring anyone else.

There’s nothing worse such a company could do if it has any interest in retaining any clients in the future.

[…] PGAs would check each other in the same way the Crips check the Bloods.

No, because PGAs would have voluntary employees, whereas criminal gangs essentially enslave their members. They are not the same.

Lastly, you set up a nightmare scenario that might occur in a free market of PGAs. I again point to Popper’s criterion. In addition, the scenario sounds more like a criticism of the status quo, where good countries (the West) have to defend against bad countries (Russia, China, North Korea, etc.).

Since the article mentions specifics, I’ve tried to address those specifics while also presenting some underlying principles (Popper’s criterion, epistemology, etc.). I suspect focusing on the underlying principles will make for a more productive discussion than going back and forth on specifics. You said in your other comment that your article already addresses several of my arguments. I think it addressed some but not all. In particular, I’d appreciate thoughts on my application of Popper’s myth of the framework.