1
Hal Finney read the email twice.
Friday afternoon, October 31, 2008. Santa Barbara, California. His home office was cluttered as usual. Two monitors, technical books stacked beside the keyboard, a half-cold cup of coffee. Through the window, the dry Southern California sun lit up the eucalyptus leaves. A peaceful afternoon. Lehman Brothers had filed for bankruptcy six weeks earlier, but the crisis had left no trace on this neighborhood's landscape.
Finney was a programmer. More precisely, a programmer for privacy. Since the early 1990s, he had worked as a core developer on PGP (Pretty Good Privacy), an email encryption program, and was a well-known name in the cryptography community. In 2004, he had even built his own system called RPoW (Reusable Proofs of Work)—an attempt to convert computational power into "cost," thereby endowing digital tokens with value. It worked, but it required a central server. If the server died, the system died with it. That was the wall cypherpunks had been running into for over a decade.
That afternoon, at that very hour, Wall Street remained in turmoil. In Washington, Treasury Secretary Paulson was anxiously monitoring the effects of the $700 billion Troubled Asset Relief Program (TARP), signed into law less than a month earlier. CNN was airing interviews with the newly unemployed, and people carrying cardboard boxes still streamed across trading floors. The financial system was collapsing.
On Finney's monitor, it wasn't the news but his email client that was open. A new post had appeared on the Cryptography Mailing List. The sender: Satoshi Nakamoto. A name he had never heard. The subject line: "Bitcoin P2P e-cash paper."
"I've been working on a new electronic cash system that's fully peer-to-peer, with no trusted third party."
A fully peer-to-peer electronic cash system with no trusted third party whatsoever. Finney stopped at the first sentence. "Trusted third party"—he knew exactly what that phrase meant in the cypherpunk lexicon. Banks. Governments. Central servers. Every digital currency project attempted over the past twenty years had ultimately failed because it could not eliminate this "trusted third party."
The attached document was nine pages long. Short for an academic paper. The title: "Bitcoin: A Peer-to-Peer Electronic Cash System." Eight references. The prose was restrained—no marketing, no hyperbole, no tone of revolutionary proclamation. It read as dry as an engineering specification.
2
The mailing list's response was lukewarm.
The first reply came two days later from James A. Donald. "We very, very much need such a system. But the way I understand your proposal, it does not seem to scale to the required size." Polite skepticism. Other responses struck a similar temperature. Some pointed to the energy consumption problem; others argued that the solution to the Byzantine Generals Problem was incomplete.
The cypherpunk community was fatigued. It had already weathered several failures.
There had been David Chaum's DigiCash. Founded in 1989, it was the first serious digital currency company. Chaum was a visionary who laid the theoretical foundation for anonymous electronic cash with his "blind signature" technology, but DigiCash depended on a central server. If the company running the server went bankrupt—and in 1998 it actually did—the currency vanished with it.
What about Wei Dai's b-money? Proposed in 1998, it introduced the concept of a distributed ledger and currency issuance through computational power. The idea was pioneering, but the consensus algorithm was never implemented in code, so it remained theoretical.
Nick Szabo's Bit Gold bore the closest resemblance to Bitcoin's design. The vision was to create digital scarcity akin to gold through proof of work, but it never fully solved the double-spending problem.
Double spending. The fundamental dilemma of digital currency. Digital data can be copied. If I can send one unit of my digital currency to A and simultaneously send it to B—if the same money can be spent twice—then it is not currency but a file. The only existing solution was singular: place a central authority that monitors every transaction. Banks serve exactly that role. When money leaves my account, the bank deducts the balance so that the same funds cannot be spent again.
Yet Satoshi's nine pages claimed to solve this problem without a bank. How?
All participants share the same ledger. When a new transaction occurs, participants in the network verify whether it is valid. The method of verification is computation—computers are tasked with solving a difficult mathematical puzzle, and the first participant to solve it bundles the transactions into a "block" and appends it to the existing ledger. In the process, new bitcoins are issued as a reward. To tamper with the ledger, one would need to command more than half of the entire network's computing power, and because that cost exceeds any gain from manipulation, it becomes economically irrational.
Adam Back's Hashcash proof-of-work mechanism, developed in 1997. Wei Dai's distributed ledger concept. Scott Stornetta and Stuart Haber's timestamp chain. Ralph Merkle's Merkle tree. Satoshi's nine pages combined all of these into one. The individual components already existed; what no one had managed was assembling them into a single working system.
Most on the mailing list failed to recognize this distinction. The prevailing reaction was "these are all ideas that have failed before." The very fact that the components already existed worked against it. Nothing looked new. But just because every Lego brick is known does not mean the structure built from them is predictable. Satoshi's innovation lay not in the invention of parts, but in the design of their combination.
Hal Finney saw it. He read the attached nine pages to the end and volunteered to test the code himself.
3
It was Halloween, by coincidence, when the nine-page white paper appeared. But the timing was more than coincidental.
Six weeks earlier, Lehman Brothers had gone bankrupt. Treasury Secretary Paulson had pressured Congress into enacting the $700 billion TARP, but its effects remained uncertain. The Federal Reserve had extended an $85 billion emergency loan to AIG. While taxpayer money rescued banks, four million households were losing their homes. Wall Street bankers collected their bonuses. Outrage simmered.
The crisis of 2008 was a failure of overlapping layers. A failure of gatekeepers—Moody's and S&P had stamped AAA on garbage. A failure of models—the Gaussian copula's correlation assumptions were divorced from reality. And a failure of the system itself—no one knew what the underlying assets of CDOs actually were, nor did anyone care to find out.
Satoshi's white paper was a direct response to all three failures.
If the gatekeepers had failed to guard the gate—what if the gatekeepers themselves could be eliminated? If no one could trace the underlying assets of a CDO—what if every transaction were transparently recorded on a public ledger? If the entire system had depended on Moody's as a single point of failure—what if the architecture were distributed, with no single point of failure at all?
The white paper's opening sentence—"a system for electronic transactions without relying on a trusted third party"—must be read in this context. Banks, credit rating agencies, regulators—these intermediaries had stood at the center of capital allocation for six hundred years. The Medici bill of exchange functioned on trust in the Medici name. The Bank of England's pound functioned on the guarantee of the Crown. The AAA ratings of 2008 functioned on trust in Moody's. And when that trust was betrayed, the entire system collapsed.
Satoshi was not proposing to eliminate trust. He was proposing to change its foundation—a declaration that trust would be built not on human judgment and reputation, but on mathematics and cryptography.
Consider the Byzantine Generals Problem—a thought experiment posed by computer scientist Leslie Lamport in 1982. Generals of the Byzantine Empire have surrounded an enemy city. Each general is encamped on a separate hilltop, able to communicate only through messengers. To attack, all must strike simultaneously; if only some attack, they lose. The problem is that some of the generals may be traitors. They might send false messages or refuse to follow the agreed plan. Messengers could be bribed by the enemy; messages could be forged.
The problem was an allegory for a fundamental challenge in computer networks. How do participants who do not know each other reach consensus without a central command? Throughout the history of finance, the answer had always been the same—establish a central authority. The Medici Bank's headquarters controlled each branch. The Bank of England served as the financial system's guarantor of last resort. A savings bank's credit review committee decided whether loans would be approved. With a central command, the Byzantine Generals Problem disappears. But when the central command becomes corrupt—like Portinari, like Moody's—the entire system is imperiled.
Satoshi's solution was to achieve consensus without a central command. Proof of work. To send a message, one must pay the cost of computation, and the majority of honest participants recognize the chain with the most computation invested as the "true" one. For a traitor to fabricate a false ledger, they would need to outcompute the rest of the network combined, and because that cost exceeds the benefit, it becomes economically irrational.
For the first time, the judgment that humans had always monopolized—"Is this transaction valid?"—was being transferred to code.
Of course, most of the cryptographers on the mailing list did not read the moment that way. To them, it was merely another technical proposal.
4
Saturday, January 3, 2009. Time unknown.
Satoshi Nakamoto generated Bitcoin's first block. The Genesis Block. Block number 0.
Every blockchain has a starting point. The first link in the chain. Because this link cannot reference a previous block—no previous block exists—it must be written directly into the code (hardcoded). Satoshi embedded a single hexadecimal string in the data field of this first block:
The Times 03/Jan/2009 Chancellor on brink of second bailout for banks
It was the front-page headline of the British daily The Times that day. An article about Chancellor of the Exchequer Alistair Darling considering a second bailout for the banks. This single line of text served two functions.
One was technical. A timestamp. A device proving that this block could not have been created before January 3, 2009. If it had been backdated, the creator could not have known that day's newspaper headline in advance.
The other was political. "Chancellor on brink of second bailout for banks"—at the very moment taxpayer money was flowing back into banks, the first block of a monetary system that needed no banks was born. An arrangement difficult to dismiss as coincidence.
The Genesis Block contained a mining reward of 50 BTC. However, due to a structural quirk in the code—these 50 BTC are not registered in the global UTXO (Unspent Transaction Output) set—they can never be spent. A digital monument. Bitcoin's first fifty coins were buried that way.
Over the following days, Satoshi generated blocks alone. Block 1, Block 2, Block 3... It was an era when mining was possible on the CPU of an ordinary desktop computer. The difficulty level was set at its minimum of 1. Satoshi was the only one on the network. Somewhere in a room, a single computer was producing a block every ten minutes. There was a time when that was all Bitcoin was.
The gap between the Genesis Block and Block 1 is its own small mystery. Under normal conditions it should have been roughly ten minutes, but in reality it took six days. Why those six days were empty is unknown. Satoshi may have been performing a final review of the code. The wait may have been deliberate. Perhaps the Genesis Block was generated multiple times until one felt right. This six-day gap is among the most trivial yet most resonant of the many mysteries surrounding Satoshi.
5
On January 10, 2009, Hal Finney posted two words on Twitter.
"Running bitcoin"
It was the first recorded instance of a human being—other than Satoshi—running the Bitcoin software. From his home office in Santa Barbara, Finney installed and launched the Bitcoin client v0.1 on his computer. A Windows GUI, a crude gray dialog box. The CPU fan began to spin. In 2009, an ordinary laptop CPU could mine Bitcoin.
Two days later, on January 12, Satoshi sent Finney 10 BTC. Recorded in Block 170, this transaction was the first transfer between two individuals on the Bitcoin network. The screen would have displayed "Status: 0/unconfirmed." A few minutes later, once the transaction was included in a block, the status would have changed. Confirmed. Ten BTC had moved from Satoshi's address to Finney's.
There was no bank in this transaction. No review committee, no credit rating, no paperwork. Two computers, one protocol, and a cryptographic signature. That was all.
The difference in speed was dramatic. In a savings bank's credit review committee, five people debated for over two hours before deciding the flow of 68 billion won (roughly $50 million at contemporary exchange rates). A Medici bill of exchange took twenty-five days to travel from Florence to Bruges. A Bank of England bond subscription required twelve days. The time it took Satoshi to send Finney 10 BTC was approximately ten minutes—the time until the next block was generated and the transaction confirmed. A speed of capital movement that had taken six hundred years to compress from twenty-five days to two hours was, in a single leap, reduced to ten minutes.
Of course, the value of 10 BTC was close to zero. It was no comparison to 68 billion won. But what mattered was not the amount—it was the structure. No human had approved the transaction. That was unprecedented in the history of finance.
Finney was excited. He exchanged emails with Satoshi, reporting bugs and suggesting improvements. When cynicism dominated the mailing list, Finney was virtually the only person who responded seriously to Bitcoin's potential. In one of his early emails to Satoshi, he wrote that if Bitcoin succeeded, each coin could someday be worth millions of dollars.
In early 2009, one bitcoin was worth zero. More precisely, no price existed. There were no buyers. On October 5 of that year, a user called "New Liberty Standard" posted the first exchange rate. It was calculated based on electricity costs. 1,309.03 BTC per dollar. One bitcoin was worth roughly 0.08 cents.
6
On May 18, 2010, Laszlo Hanyecz, a programmer in Florida, posted on the Bitcoin forum (Bitcointalk).
"I'll pay 10,000 bitcoins for a couple of pizzas... like maybe 2 large ones."
Four days later, on May 22, a user named Jeremy Sturdivant ordered two Papa John's pizzas, had them delivered to Hanyecz, and received 10,000 BTC. The value at the time: approximately $41. About $20 per pizza. It was the first recorded instance of Bitcoin being exchanged for a physical good.
This transaction matters not because of how the pizza tasted. It matters because the final barrier Bitcoin needed to clear to become "money"—the consensus that someone would accept it in exchange for goods—was established for the first time. Economists call this "the function as a medium of exchange." The Medici bill of exchange could move capital across borders on a single sheet of paper because the counterparty trusted that the paper could be redeemed for gold florins. Hanyecz's pizza transaction was the first moment that the same kind of trust, however modest, even if only worth two pizzas, was established for Bitcoin.
10,000 BTC. This number would be invoked every time Bitcoin's price rose thereafter. The cryptocurrency community celebrates May 22 as "Bitcoin Pizza Day." Did Hanyecz make a mistake? Probably not. In 2010, Bitcoin was an experiment. It was a digital token traded among a small group of enthusiasts on a forum, and the act of buying pizza with it was itself part of the experiment. Hanyecz himself said in later interviews that he had no regrets. If someone had not been the first to spend Bitcoin, it would have remained forever "something that could be spent but never was."
The Bitcoin ecosystem of 2010 was small and quiet. Users numbered in the low thousands at best. There was barely anything resembling an exchange. Mt. Gox, which opened in July 2010 and became the fastest-growing early Bitcoin exchange, had originally been "Magic: The Gathering Online eXchange"—a trading card marketplace. A card-game exchange had reinvented itself as a digital currency exchange. That single fact is enough to gauge where Bitcoin stood in 2010.
Mining was still possible with the CPUs and GPUs of ordinary computers. Hardly anyone bothered to calculate whether mining revenue exceeded electricity costs. Technical discussions dominated the Bitcoin forum; talk of price was secondary. A world where the very concept of price had not yet been established.
7
While Bitcoin grew slowly, Satoshi Nakamoto was quietly disappearing.
From 2009 through mid-2010, Satoshi was actively engaged. He answered technical questions on the forum, revised code, and exchanged emails with other developers. His writing style was consistent—precise, courteous, never betraying emotion. Records indicate he used British English ("colour," "analyse"), though whether that reflected his actual nationality or was a deliberate misdirection, no one knows.
On December 12, 2010, Satoshi posted his final message on the Bitcoin forum. It was technical in nature, with no special farewell.
In April 2011, Satoshi sent an email to Gavin Andresen, a core Bitcoin developer. "I've moved on to other things." After that, the email address went silent, and the forum account ceased all activity.
Just before vanishing, Satoshi had expressed concern over one incident. WikiLeaks had announced it would accept Bitcoin donations. In late 2010, under pressure from the U.S. government, Visa and Mastercard had blocked payments to WikiLeaks, and WikiLeaks was considering Bitcoin as an alternative. Satoshi warned that this would be "kicking the hornet's nest." His judgment was that attracting government attention while Bitcoin was still immature would be dangerous.
And then he left. Without any explanation.
Satoshi's disappearance completed Bitcoin. The statement is paradoxical, but precise. Had the founder remained, that person would have become yet another central authority. Satoshi's pronouncements would move markets. Satoshi's decisions would steer the protocol's direction. Pressuring Satoshi would mean controlling the system. By disappearing, Satoshi made Bitcoin a system without a leader. No one could sue Satoshi, arrest Satoshi, or lobby Satoshi.
Addresses believed to belong to Satoshi's wallet hold approximately one million BTC. This estimate comes from the "Patoshi Pattern" analysis published in 2013 by blockchain analyst Sergio Demian Lerner. These coins have never been moved. At current valuations in 2026, the sum amounts to tens of billions of dollars at contemporary exchange rates, locked in a digital vault, visible to everyone on the blockchain yet movable by no one. Whether Satoshi is alive or dead, one person or many—there is no way to know.
When Giovanni de' Medici died, his son Cosimo inherited the bank. When the Bank of England's William Paterson departed, the governorship passed to the next in line. In institutions built by humans, there was always a successor, and with the successor came continuity of power. Satoshi left no successor. What he left behind was code. And code needs no successor—the rules are already written within it.
8
Hal Finney, too, was running out of time.
In August 2009, Finney was diagnosed with amyotrophic lateral sclerosis (ALS). A disease in which motor neurons progressively die. The same illness Stephen Hawking endured. There was no cure. Even after the diagnosis, Finney continued contributing to Bitcoin's development. When typing became difficult, he switched to an eye-tracking device. He wrote code by directing a cursor with the movement of his eyes.
On August 28, 2014, Hal Finney passed away. He was fifty-eight.
Among the last things he wrote was a post on the Bitcoin forum. In a message dated March 19, 2013, Finney described his condition with quiet composure. His entire body was paralyzed and he could no longer move, but with the eye-tracking device he could still write, and he was still programming. He wrote that he remained optimistic about Bitcoin's future—each letter entered by the movement of his eyes.
Finney's remains were cryopreserved by the Alcor Life Extension Foundation, pending a future technology that might cure ALS. The symmetry is fitting: the first person to recognize Bitcoin's potential now rests in the custody of a technology that does not yet exist. The ability to see value in something that does not yet exist—Finney was that kind of person.
The identity of Satoshi remains unknown to this day. Many have asked whether Finney himself was Satoshi. Finney denied it, and his family maintains the same position. Whether he was Satoshi may not matter. What matters is that someone existed who read Satoshi's nine pages and judged, "This can work." When the other cryptographers on the mailing list expressed skepticism, Finney installed the software and ran it. He responded not with theory, but with action.
Throughout the history of capital allocation, judgment had always belonged to the human domain. A Medici branch manager gauged a merchant's creditworthiness by looking at his face. The Bank of England's directors debated the safety of government bonds. A savings bank loan officer argued over pre-sale rate assumptions. The tools of judgment changed—bills of exchange, government bonds, BIS ratios, the Black-Scholes formula—but the agent of judgment was always human.
Bitcoin inverted this structure. "Is this transaction valid?" is decided not by a human but by the network. The rules are written in code, and code knows no exceptions. There is no compassion, no lobbying, no consortium pressure that can alter the outcome. When a credit review committee chairman said "Let's approve it conditionally," his judgment was shaped by factors beyond the numbers—the relationship with the consortium, the pressure of sales targets, the watchful eyes of superiors. In Bitcoin's protocol, none of that exists.
Whether this is liberation or deprivation is a question that remains unanswered.
In the credit review committee of Chapter 3, the chairman's "conditional approval" contained something that could not be reduced to numbers. An impression of the construction company CEO's track record. A feel for the mood of the real estate market. The psychological pressure of an impending inspection by the Financial Supervisory Service, Korea's financial regulator. Human judgment always carried this "something beyond the numbers." Sometimes it became wisdom. Sometimes it became bias.
Bitcoin's protocol has nothing beyond the numbers. If the rules are met, the transaction is approved; if not, it is rejected. No gray zone is permitted in between, and nothing like "conditional approval" is possible. This coldness makes the system robust, but human context is excluded.
The nine-page white paper proposed something larger than a digital currency. It was the beginning of an experiment to transfer to code the authority of judgment that humans had monopolized for centuries.
9
But what Satoshi solved was only a fraction of finance.
Bitcoin solved remittance. The act of A transmitting value to B. Without a bank account, without borders, without the constraint of business hours. The chain that began with the Genesis Block has not stopped once since. Every ten minutes since January 3, 2009, a new block has been generated, and as of 2026, that record continues unbroken.
But finance is not remittance.
Finance is lending. The act of loaning money to someone, collecting interest, and liquidating collateral if they cannot repay. Finance is insurance. The act of distributing in advance the cost of an event that has not yet occurred. Finance is investment, the trading of risk, the pricing of an uncertain future. When the Medici Bank issued a bill of exchange, it was not simple remittance but the creation of credit. When the Bank of England underwrote government bonds, it was lending to a nation. What the credit review committee examined in Chapter 3 was a PF loan—project financing.
Bitcoin could do none of these things. Lending? Impossible. There was no mechanism for setting interest rates, no capability for automatically liquidating collateral. Bitcoin was a currency, not a bank.
The essential function of a savings bank is "to receive money, make a judgment, and lend it out." It takes deposits, reviews loan applications, approves them, collects interest, and liquidates collateral when loans go bad. In this entire process, the only part Bitcoin could replace was "sending money." The simplest act of all.
Then what about the rest? Could "judging and lending" be implemented in code?
In the decade after Bitcoin, countless people attempted variations of this question. If remittance without a bank was possible, could lending without a bank also be possible? Insurance without a bank. An exchange without a bank. If everything a bank does could be implemented in code—could there be a bank without a bank?
In 2015, a Russian-Canadian programmer named Vitalik Buterin launched Ethereum. If Bitcoin was "programmable money," Ethereum was "programmable everything." On Ethereum, one could deploy smart contracts—code that executes automatically when predefined conditions are met. Write a rule in code that says "if collateral falls below a certain ratio, liquidate automatically," deploy it on the blockchain, and that rule runs twenty-four hours a day without rest. No chairman's approval required, no business-hour constraints.
In the summer of 2020, this possibility exploded into reality. While the world was locked down under COVID restrictions, an entirely new financial system was being constructed on the Ethereum blockchain. Lending, trading, insurance, derivatives—the functions of finance that banks had built over centuries were being replicated in a few lines of code. Some called it "DeFi Summer."