Cryptocurrencies are antifragile

It’s common to separate things into two categories: those that are fragile and those that are robust. Regular glass is for example fragile as it breaks easily, often by accident, and bullet-proof glass is robust as it can withstand a significant amount of force before it breaks.

But there’s a third category that we often don’t think about. It’s called antifragile and it goes beyond robustness as it grows even stronger when stressed. I think the phrase “what doesn’t kill you makes you stronger” explains antifragility quite well.

At first it might sound impossible; how could something grow stronger from pressure? It would be like if we had armor that was self-healing but after getting shot it also learned from the bullet and adapted to resist the next shot better. Clearly that’s only possible in science-fiction.

But that’s exactly how our immune system works. After defeating a virus that might make us very sick, our body learns to recognize the virus so it can attack and kill it much earlier. This is how we might develop immunity against COVID-19 and it’s the basic principle that make vaccines work.

Taleb describes a myriad of things that are antifragile; for instance financial long options, street fights, love and religion. As I’ll argue in this chapter cryptocurrencies also belong to this group.

The longer it lives the longer it’s expected to live

Time is one of the things that break down the fragile and make the antifragile stronger. This is exemplified by the Lindy Effect:

The old is expected to stay longer than the young in proportion to their age.

In other words the longer something lives, the longer it can be expected to live. The Lindy Effect is only relevant to non-perishable goods; it applies to things like inventions or money, but doesn’t apply to humans or food.

It’s a heuristic, not a natural law, but the implications generally make sense. For instance that Bitcoin will survive another decade, that a newly created cryptocurrency promising to solve all of Bitcoin’s problems will fail and that gold will outlive them all by a few millennia.

The largest bug bounty in the world

In the software development world a bug bounty is a reward for someone who finds a vulnerability and reports it in a responsible manner. This is vastly preferable to them exploiting it or selling it on the black market, in particular if it’s a serious one. For example it’s easy to imagine the problem caused by a bug that allowed an attacker to view all private messages in Facebook, so Facebook might be willing to pay a hacker a lot of money if they help them patch this security issue.

While a decentralized project like Bitcoin doesn’t have an official bug bounty, Bitcoin itself functions like a very large bug bounty. Because if you find a way to break Bitcoin, and transfer coins to yourself, you could steal billions of dollars worth of Bitcoin and nobody would know who did it.

This is why I think the Lindy Effect applies to cryptocurrencies; with each day that pass without a fatal bug being found in a cryptocurrency, the probability that it will survive another day increases. With such a large sum of money to incentivize people to search for bugs, the fact that no fatal bugs have been found yet inspire confidence.

Antifragility loves small errors

What kills me makes others stronger

Another property of the antifragile is that they like small errors, because after recovering after an error they grow stronger. Therefore they want many small errors, but not errors so big the recovery time becomes too long.

This is how our muscles grow stronger. When we do strength training we break down our muscles by causing micro tears, and with rest they heal and make the muscles stronger. But we must be careful to not overdo it otherwise we’ll break down, only becoming weaker.

And it also holds true for the immune system. Some types of vaccines even give you a very small dose of the virus—just enough to activate your immune system and teach it how to kill the virus—but not so large to make you so sick you couldn’t function as normal. And of course you don’t want to die in the process.

Cryptocurrencies work the same way. When a weakness is discovered developers will introduce a fix, eliminating the weakness. As long as the cryptocurrency doesn’t completely die, weaknesses will only strengthen it, making it harder to disrupt.

Splitting ensures antifragility

While you can argue that an individual cryptocurrency is antifragile, antifragility also applies to the cryptocurrency concept itself. People like to point at the thousands of cryptocurrencies as some sort of drawback, but it’s actually the opposite. Cryptocurrencies being created and then dying off en mass is good as each time one fails we might learn something new and use that to improve the survivors.

Forking a cryptocurrency is a fundamental antifragility force. This means forking the ledger, where a cryptocurrency splits into several but all the addresses and coins are still there, or forking the code but restarting from an empty ledger.

This process allows a community with different ideals to explore their own path, it allows us to explore different ways to solve a problem and it even allows a cryptocurrency to recover from a catastrophic bug. Forking is the ultimate fail-safe.

Concrete examples

Let’s look at some concrete examples of how antifragility might come into play for a cryptocurrency:

1. Developers disagree on the best way forward

   The coin splits in two and both paths are explored.

   For example in Bitcoin there was a big argument of how best to scale. Some wanted to explore on-chain scaling to find the limits and others wanted to move away from the blockchain to “second layer solutions” and depend on them to scale.

   This lead to the BTC/BCH split where a part of the community left Bitcoin (BTC) to create Bitcoin Cash (BCH) that aims to prioritize on-chain scaling.

   You may wonder, which approach is the correct one? That’s the best part: it doesn’t matter as the cryptocurrency idea will live on regardless. The big issue is if both approaches are a dead-end.

2. Reference client abandons core values

   What would happen if the developers of a cryptocurrency decided to print coins from thin air and make themselves rich? Or make some other change that would destroy the cryptocurrency?

   Then they would be replaced.

   For example Monero replaced—or kicked out—the old developer team early on in it’s history when the developers pushed changes that the community disagreed with.

   A more recent example is the BCH/ABC split. Since Bitcoin Cash split from Bitcoin in 2017, Bitcoin ABC (Bitcoin Adjustable Blocksize Cap) has been the reference client that has dictated every change that went into the protocol. This ended in 2020 when ABC wanted to reroute 8% of the block reward to an address under their control. The change had serious centralization concerns, and in the end the miners and the BCH community rejected them and they split away to their own minority fork.

   Not only did Bitcoin Cash replace the reference client, they replaced it with a handful of cooperating developer teams, reducing the damage a rogue developer could cause.

3. Marketplaces freeze your assets

   Decentralized marketplaces that allow for non-custodial trades are created. This means the marketplaces never control your funds and all trades are made directly with the other person, making it impossible for anyone to freeze your funds.

4. Governments declare cryptocurrencies illegal

   Governments have declared drugs and prostitution illegal for centuries, but they still manage to thrive. Making them legal would paradoxically make them easier to control, and I think it’s the same with cryptocurrencies.

   If governments banned cryptocurrencies people would still trade them in person and move to decentralized marketplaces, which are inherently harder to disrupt. We would move away from centralized payment services and move towards a peer-to-peer economy, like how Satoshi envisioned.

   For example Monero supporters worry that governments will declare cryptocurrencies with good privacy features illegal (a reasonable thing to worry about). This will soon be a smaller issue as work on a Bitcoin-Monero atomic swap project is underway, which would allow people to trade Bitcoin and Monero without any third-party.

5. A 51% attack

   One of the worst things that can happen to a cryptocurrency is a 51% attack, but even such an event can strengthen it.

   For example an attacked chain could change the mining algorithm, making all existing mining rigs worthless. This would make it impervious to the same type of attack and if an attacker wants to make another attempt brand new mining rigs would have to be purchased or created.

   As another example during the BCH/BSV split the BSV (Bitcoin Satoshi’s Vision) miners threatened to attack the BCH chain, and to not allow any transactions to be confirmed. This would essentially kill the chain, but this attack was never carried out and protection was added to the BCH software that made it more difficult to carry out such an attack.

Antifragility brings confidence

There’s a worry that governments will kill cryptocurrencies by making them illegal; that PayPal will co-opt Bitcoin and turn it into a corporate coin; that Facebook or governments will create a centralized digital currency that out-competes real cryptocurrencies; that developers will go rogue; or that a fatal bug will bring it all down.

I’ve gotta be honest, they don’t worry me too much. I see them as temporary setbacks that are harmful in the short-term, but will in the end make cryptocurrencies even stronger.

Therefore I’m convinced that as long as internet exists, so will cryptocurrencies.