A Libertarian Response to “Against Szabo’s Law”
Two days ago Vlad Zamfir published an essay arguing that existing governance mechanisms in crypto are dangerous and that we should move toward a system that makes it easier to intervene in the operation of crypto networks.
I see a strong correspondence between Vlad’s objections to existing crypto governance and common leftist objections to libertarianism.
This is not intended as a full response to Vlad’s essay — I recommend Vitalik’s reply for that. My goal is to examine only the parts of Vlad’s essay that exhibit this interesting correspondence.
Overview of Vlad’s Argument
Vlad describes “crypto law” as the protocol for handling disputes in crypto governance. He suggests that the three most important crypto laws today are:
#1: Don’t break the protocol.
#2: Keep crypto law legal.
#3: Do not implement changes to the blockchain protocol unless the changes are required for the purpose of technical maintenance.
Vlad wants to replace these three laws with:
Crypto Law is responsible for managing disputes in blockchain governance, and making sure that they are resolved via legal processes that don’t break the protocol.
Vlad essentially wants to do away with Law #3, which he calls “Szabo’s law.” Vlad prefers a system that places a higher priority on conforming to legal systems outside of crypto and using political mechanisms to intervene in the operation of crypto networks.
Szabo’s Law and Non-Intervention
Libertarians believe that traditional governments should be stripped of their power to intervene in people’s affairs (with a few exceptions).
Similarly, many crypto-libertarians believe crypto-governance processes should lack the power to intervene in the operation of crypto protocols (with a few exceptions).
I’ll refer to Law #3 as “Non-Intervention” instead of “Szabo’s law” to better reflect that it’s a straightforward application of libertarian theory to crypto and not just some ad hoc theory that Nick Szabo came up with. I’ll refer to Law #2 as “Legality”.
Arguments Against Non-Intervention vs. Arguments Against Libertarianism
All quotes are from Vlad’s essay.
Not only does the legalization of Szabo’s law determine governance outcomes (always in favor of not intervening with the execution of software), it minimizes the space for political and legal conversations that question whether those outcomes are desirable.
This critique could be made almost identically against libertarianism.
Libertarianism seeks to minimize the decisions that can be influenced by the political process in the same way that Non-Intervention seeks to minimize the space of possible crypto governance decisions.
Both libertarians and crypto-libertarians have theories for why minimizing the power of people who might want to intervene will lead to better outcomes. The gist of both theories is that this power will be captured by those who will use it in ways that benefit themselves at the expense of the overall system.
It also makes sense if Szabo is so radically jaded that he believes that crypto law and politics cannot be worth the effort, no matter what form the crypto legal system might take.
Imagine the same critique made against libertarianism: “Libertarianism only makes sense if you’re so jaded that you believe that giving the government lots of power can’t be worth the effort, no matter what form this government might take.”
The libertarian might reply: “There’s a chance that someone will come up with a way to give government a lot of power and have it work out better than libertarianism would, but my reading of history suggests we haven’t figured out how to do that yet. That’s why I’m a libertarian. Why don’t you offer up a concrete proposal and then we can evaluate how well it’s likely to work?”
Vlad wants us to abandon Non-Intervention because he sees risks from autonomous crypto systems and sees more interventionist systems as leading to better outcomes. It’s hard to evaluate his proposal given that he doesn’t go into detail about either of these things.
Why is it worth giving up the predictability and non-gameability of Non-Intervention to prevent these abstract risks?
Why should we believe that the risks brought about by giving governments influence over crypto protocols are any less concerning?
Szabo’s law is not anti-political. It is a law that is aimed at shutting down political debate in order to guarantee Nick’s preferred political ends.
I regard this kind of anti-social behavior to be bad-faith participation in blockchain governance.
Non-Intervention is indeed political, but it seems no more “bad faith” than libertarianism.
A leftist might accuse a libertarian of trying to “shut down political debate” because most of the issues that the leftist wants to use politics to decide are issues that the libertarian wants to keep outside of the political realm.
Does Vlad think libertarianism is inherently in bad faith because of this? If not, then Non-Intervention isn’t inherently in bad faith for an analogous reason.
Next, Vlad claims that Non-Intervention is both “insecure” and “aggressive”:
“We don’t deal with disputes that aren’t related to tech maintenance” is insecure in crypto law’s ability to legitimately manage the disputes that might arise in blockchain governance.
This seems like a word game. Here’s a similar example:
“Libertarianism holds that government is bad at solving big societal problems.”
Is it legitimate to rephrase this as “libertarianism is insecure in government’s ability to solve big societal problems” so that we can tar libertarianism with the word “insecure”?
If so, how about “leftism is insecure in society’s ability to function well under a system of minimal government”?
“There’s nothing we can do for you” is an aggressive posture, when someone has a legitimate dispute.
This suggests that Vlad would also call libertarianism “aggressive.”
I don’t see how this is aggressive. I can see how someone could call it cold or uncaring. Aggression is an active thing.
And why is crypto law insecure and aggressive?
Now we’ve gone from saying that Non-Intervention is “insecure in <something>” to just “insecure.”
Can I also shorten “leftism is insecure in society’s ability to function well under a system of minimal government” to “leftism is insecure”? If not, what’s the difference?
We can use the same technique to associate almost any adjective with any philosophy.
Nick knows that autonomous software isn’t always going to be legal or politically popular, and he is determined to use crypto law to shut down any legal and political coordination that would undermine his mission. This antisocial behavior makes me question…
Libertarianism is also an attempt to “shut down” any legal and political coordination to tax individuals and spend the resulting money on occupational licensing. Is it antisocial that libertarians advocate for a society where government doesn’t have this power?
Does Vlad think that the very act of limiting the future power of a system is antisocial? If not, what makes normal libertarianism OK but crypto-libertarianism antisocial?
Democracy is broken in part because of the asymmetry between special interests and rationally apathetic voters. Patri Friedman gives a great two minute description of this here. Structurally, crypto governance seems like it should suffer from similar problems. This provides some justification for the restriction of decision making power in crypto governance.
Szabo’s “mission” involves preventing such a broken system from being used by bad actors to make everyone collectively worse off. From a libertarian perspective this is the opposite of antisocial.
Non-libertarians of course have a different view, but at this point Vlad’s critique seems to require as a premise that libertarianism is lame/antisocial.
We need to abandon Szabo’s law to adopt a more open and secure legal posture. One that acknowledges rather than shrugs off its responsibility to carefully manage disputes.
Imagine the same argument applied to libertarianism: “We need to abandon libertarianism in favor of a government system that acknowledges and doesn’t shrug off its responsibility to solve society’s problems.”
To make the point of these libertarianism comparisons explicit: ask yourself whether these arguments and rhetoric present a solid case against libertarianism. If not, then because the connection between Non-Intervention and libertarianism is so tight this is evidence that these aren’t solid arguments against Non-Intervention.
We cannot forsee the nature of all of the blockchain governance disputes that will arise in the future, and need to retain the ability to remain flexible enough to adapt to changing circumstances, we cannot afford to blindly pledge our fates to a future with autonomous software.
The same argument applied to libertarianism: “We can’t foresee the problems that our society will face in the future, so we must retain the ability to solve problems that require a powerful government. We can’t afford to blindly pledge our fates to market mechanisms and the emergent properties of individual choices.”
We should admit that we collectively have an obligation to manage the disputes that will arise from the operation of global public blockchains to the best of our crypto legal ability, so that as many people as possible can enjoy the benefits of global public blockchains.
Applying the same rhetoric to libertarianism: “We collectively have an obligation to manage our societal problems to the best of our governmental ability — which requires giving our government adequate resources and power — so that as many people as possible can enjoy the benefits of living under our government.”
A Potentially Better Argument Against Non-Intervention
Vlad’s current approach might persuade people who already dislike libertarianism that Non-Intervention is bad, but crypto-libertarians won’t find it convincing.
A more plausible way to convince crypto-libertarians to abandon Non-Intervention is to show that governance of decentralized systems is different enough from governance of centralized systems that using essentially the same approach for both doesn’t make sense.
One obvious difference is that crypto governance is decentralized (at least in some cases), and forking makes exit much easier in crypto systems than in traditional systems. Ease of exit means there’s plausibly less motivation to limit the decision making power of crypto systems. I’ve written about this here.
Summary
Non-Intervention is essentially libertarian theory applied to crypto: an attempt to minimize decision making power based on the idea that if left in place it would be used to do more harm that good.
Many of Vlad’s arguments against Non-Intervention look like unconvincing arguments against libertarianism. Translating these arguments into the equivalent anti-libertarian arguments makes the structure of these arguments easier to understand.
It may still be possible to convince crypto-libertarians to abandon Non-Intervention by focusing on the differences between decentralized governance and centralized governance.
Follow me on twitter.
