Start with what is true, because the rest of this chapter only earns the right to be read if this part is said properly.
Bitcoin is the most important monetary invention of the last hundred years. In January 2009 it demonstrated, for the first time, that a currency could exist with a fixed supply, no issuer, no company, no board and no permission — and that it could keep running while every institution that predicted its death went on to have a worse decade than it did. Seventeen years of uninterrupted operation is not luck. It is engineering.
It also answered the first of this book's three questions completely. Can more of it be created, and by whom? No, and by no one. Twenty-one million, enforced by consensus rather than policy, with a schedule that nobody has been able to renegotiate. Whatever else is said in this chapter, that achievement stands, and everything that follows is built on it — including the coin this book is about.
But Bitcoin answered one question, and the other two were left where they were.
Transparency was the point The ledger is public. Every transaction is visible, every address is visible, every amount is visible, and the whole history is retained forever.
This was not an oversight. It is how Bitcoin verifies without trusting anyone: if everyone can check the books, no one has to be trusted to keep them. The design is elegant, and it worked.
The cost was built in at the same time. Transparency means the ledger is a permanent, public, queryable record of who paid whom. Addresses are pseudonymous, not anonymous — a distinction that has been commercially exploited for a decade by an industry that exists specifically to attach identities to them. Chain-analysis firms sell clustering, heuristics and attribution to exchanges, governments and anyone else with a budget. Once an address is linked to a person, everything that address ever did is linked to that person, permanently, including the transactions that happened before the link was known.
There is no mechanism to opt out. There is no version of Bitcoin where the history is not there. The record is the security model, and the record is the exposure.
A coin is not always a coin Once history is public, history can be graded, and graded history becomes a price.
Coins that passed through a hack, a sanctioned address, or a market the authorities dislike are marked. Exchanges run screening on deposits and sometimes freeze or refuse them. In 2018, the United States sanctions list named Bitcoin addresses for the first time, which turned an analytical category into a legal one: an address you transacted with could place you, unknowingly, on the wrong side of a list.
The word for the property being lost is fungibility — the condition in which any unit is interchangeable with any other, so that a holder never has to ask where their money has been. Gold has it. Cash mostly has it. Bitcoin increasingly does not.
Andreas Antonopoulos, who spent his career explaining Bitcoin to the people who would use it, put the stake plainly: tainted coins are destructive, and if you break fungibility and privacy, you break the currency. That is not a critic of Bitcoin saying it. It is one of its most influential teachers.
The part that is harder to argue with Everything above concerns the ledger. There is a second layer, and it concerns who produces the blocks.
Mining is the process that orders transactions, and a miner chooses which transactions to include. That is not a flaw in Bitcoin; it is what mining is. The question is whether anyone has ever used that discretion to exclude a payment, and whether the structure makes that easy or hard.
The record says: yes, and easier than the popular account suggests.
In May 2021, Marathon mined block 682,170 and left out transactions associated with sanctioned addresses, on purpose. After a backlash it reversed the policy four weeks later. In October 2023, the pool behind roughly fourteen percent of blocks, F2Pool, skipped sanctioned payments; its co-founder confirmed the filter publicly. In December 2024, nine F2Pool blocks left out sanctioned payments again, and other pools mined them later.
Nothing was stolen. The payments eventually settled. But a filter was applied at the point of production, by a private company, using information that every miner already has — and it was applied because the company decided to apply it.
What a filter needs is visibility, which Bitcoin provides by design, and a small number of willing pools. The second condition is closer than most people think. Three pools have mined a majority of all blocks in recent measurements — Foundry, AntPool and F2Pool. Around forty percent of hashrate has run through AntPool and its proxy pools. More than eighty percent of Bitcoin mining chips have come from a single manufacturer, Bitmain.
And the off switches have been pulled before, from outside the protocol entirely. In 2017, Bitmain's firmware contained a function that could remotely stop Antminers; it was removed after disclosure. In 2021, a mining ban in China cut Bitcoin's hashrate by about half in five weeks. In January 2022, an internet shutdown in Kazakhstan took out thirteen percent of hashrate. In 2024, the Texas grid asked miners to power down during a demand event and hashrate fell twenty-five percent in four days.
The network survived all four. It is remarkably resilient. But in each case, a government, a grid operator or a chip manufacturer decided how much of it ran, and for how long.
Why the base layer stayed small The capacity question was argued in public for years, and the argument was settled on purpose.
Through 2015 and 2016 the network debated raising the block size limit — the cap on how much transaction data a block may carry. Bitcoin kept the cap, activated SegWit in 2017, and pushed throughput to layers built above the chain. Part of the community disagreed hard enough to fork, and Bitcoin Cash exists because of it.
The decision was defensible on its own terms. Small blocks keep validation cheap, which keeps node operation within reach of ordinary people, which is what stops the network from becoming a service run by a handful of institutions. That is the same argument this book will make about chain size in a different context, and it deserves to be stated fairly here.
But the consequence is unavoidable. A base layer that settles a handful of transactions per second cannot be the layer where people pay each other. Payments move up to Lightning, Lightning requires channels, channels require liquidity and an online node, and for most users the practical path back to the base layer runs through a custodian. The design kept the node small and moved the payment experience into exactly the intermediaries the design was meant to remove.
That is not a contradiction unique to Bitcoin. It is the shape of the trade-off: you can have cheap validation or cheap payments on the base layer, and Bitcoin chose validation. The interesting question — the one this book keeps coming back to — is whether a different chain design can have both, which is what cut-through and a four-gigabyte ledger will be asked to demonstrate later.
What it became None of this makes Bitcoin a failure. It makes it a success that drifted.
The drift is visible in what people do with it. Bitcoin was written as peer-to-peer electronic cash, and the phrase is in the title of the paper. What exists today is closer to a settlement asset and a speculative instrument: a store of value held through custodians, wrapped into funds, traded on venues, and increasingly correlated with the same equity markets it was designed to be independent of.
Consider the ordinary facts of ownership. Most holders do not self-custody. Most exposure runs through an exchange, a fund, or a brokerage account. The people who own Bitcoin in the way its author intended — keys on their own device, no intermediary — are a minority of a minority.
This is not hypocrisy. It is gravity. A scarce asset with a rising price attracts capital, and capital arrives through institutions, and institutions reintroduce exactly the intermediaries the design was built to remove. The asset did not fail. It was absorbed.
The price is real, and so is the achievement. Both things are true at once, which is why the argument has to be made carefully rather than loudly.
What this leaves open Come back to the three questions.
Store of value. Answered. Twenty-one million, no issuer. Whatever happens next, this part is done, and everything after Bitcoin has to match it rather than improve on it.
Medium of exchange. Partly answered. It settles, but capacity is a handful of transactions per second, fees move with demand, and the base layer was never designed for a coffee. The answer to this one has been pushed to layers above, which work, and which reintroduce custodians and channels.
Sovereignty. Answered at the issuance layer, less so at the production layer. No one can stop you holding a private key. Someone can decline to mine your transaction, and a handful of pools have. Someone can also read every payment you have ever made, which is its own kind of control — the kind that does not need to block anything to change your behavior.
So the design Bitcoin left unfinished is specific. It is not speed, and it is not supply. It is the combination of privacy and fungibility, at the base layer, without asking the user to opt in and without adding a second system that can fail on its own.
That last clause matters more than it sounds, and it is where the next chapter goes. Because the obvious way to add privacy to a transparent chain is to build a shielded pool beside it — a second mechanism, with its own cryptography, its own assumptions, and its own bugs. That approach has produced most of the privacy in this industry. It has also produced every major failure in this category since 2019.
There is another way to do it, and it starts by not having anything to hide in the first place.