The Cypherpunks and the Dream of Private Money

By the early 1990s, the mathematics for private digital communication existed. Public-key cryptography had been published in the late 1970s. David Chaum had shown how to make digital cash that did not reveal who paid. What did not yet exist was a community determined to put these tools into ordinary people’s hands, and willing to argue with governments about it. That community formed in the San Francisco Bay Area in 1992.

Who the cypherpunks were

The founding group included Eric Hughes, a mathematician; Timothy C. May, a physicist who had retired early from the chip maker Intel; and John Gilmore, an early employee of Sun Microsystems and a co-founder of the Electronic Frontier Foundation. They began meeting in 1992 to talk about cryptography and its political consequences. The writer and hacker Jude Milhon, attending one of the early meetings, is credited with joking that they were cypherpunks, a play on cyberpunk fiction and the cipher. The name stuck.

The group soon set up an e-mail list, hosted on a machine run by Gilmore. Anyone could join. Within a few years it had hundreds of subscribers and carried a constant stream of messages: technical proposals, political arguments, announcements of new software and, inevitably, quarrels. Many of the people who would shape internet privacy and digital cash over the next two decades passed through it.

A workshop in public

The list worked as a kind of open workshop. Someone would post a design for an anonymous message system or a payment protocol; others would look for weaknesses, sometimes within hours; the author would revise it or abandon it. Source code was shared, criticised and improved. Physical meetings continued in the Bay Area and spread to other cities. For a newcomer, reading the archive was an education in applied cryptography and in the habit of assuming that any system would be attacked.

That habit, adversarial thinking applied to everything, was perhaps the cypherpunks’ most lasting contribution to the design of digital money. A system was not judged by how well it worked when everyone behaved, but by how it held up when some participants did their worst.

Two manifestos

Tim May had written “The Crypto Anarchist Manifesto” in 1988, before the group existed. In a page of deliberately provocative prose, it predicted that cryptography would let individuals and groups communicate and do business anonymously, and that this would change the nature of government regulation, taxation and the control of information. May saw this as inevitable and, on balance, liberating.

Eric Hughes’s “A Cypherpunk’s Manifesto”, dated 9 March 1993, was calmer and more widely quoted. It began with a sentence that would be repeated for decades.

“Privacy is necessary for an open society in the electronic age.”

Eric Hughes, A Cypherpunk’s Manifesto, 1993

Hughes distinguished privacy from secrecy: privacy, he argued, is the power to reveal oneself selectively to the world. He argued that in an electronic society this power would require cryptography, and that anonymous transaction systems were needed if people were to pay for things without creating permanent records. And he stated the group’s method in three words that became its motto: “Cypherpunks write code.” The point was that rights would not be granted by governments or companies; they would have to be built, as software, and given away.

The crypto wars

The cypherpunks did not work in a vacuum. In the United States at the time, strong encryption software was treated for export purposes as a military item, and its export without a licence was restricted. The government was also worried that widespread encryption would make lawful wiretapping impossible.

PGP

In 1991 a software engineer named Phil Zimmermann released Pretty Good Privacy, or PGP, a program that let anyone encrypt e-mail and files with public-key cryptography. Copies spread across the internet and beyond the United States. Zimmermann became the subject of a federal criminal investigation over the export of the software. In 1995 MIT Press published PGP’s source code as a printed book, a way of highlighting that a book, unlike software, was clearly protected as speech. The investigation was dropped in 1996 without charges.

A web of trust

PGP also embodied an idea about trust that would matter later. Public-key cryptography has a weak point: how do you know that a public key really belongs to the person it claims to belong to? The usual answer, adopted for secure websites, is a hierarchy of certificate authorities, organisations whose job is to vouch for keys.

PGP took a different approach, which Zimmermann called the web of trust. Users could sign one another’s keys to vouch that they belonged to their owners, and each user decided whose signatures to rely on. There was no central authority; trust spread outward through personal connections. Cryptographers held key-signing gatherings at conferences, where people checked each other’s identification and fingerprints and then signed keys.

The web of trust was never easy to use, and most people never used it. But it was an early, practical attempt to replace a central authority with a network of peers who vouch for each other, an idea that would reappear, in a very different technical form, in later designs for digital cash.

The Clipper chip

In April 1993 the Clinton administration announced the Clipper chip, an encryption device for telephones that would hold a copy of each device’s key in escrow with government agencies, so that law enforcement could decrypt conversations with a court order. The cypherpunks, along with many others, campaigned against it. In 1994 the cryptographer Matt Blaze, then at AT&T Bell Laboratories, published a flaw that could let users defeat the escrow mechanism. The scheme was never widely adopted and was effectively abandoned within a few years.

The legal fight over export controls continued through the 1990s, notably in a case brought by the mathematician Daniel Bernstein, in which a panel of a federal appeals court held in 1999 that software source code was a form of protected speech; the decision was later withdrawn for rehearing, and the case faded as the rules changed. In 2000 the United States substantially relaxed its export rules. Strong encryption, which had been treated as a weapon, became a normal feature of web browsers and, later, of messaging apps used by billions.

The dream of private money

For a large part of the cypherpunks, encrypted communication was only half the project. If every payment left a record, privacy in conversation would not be enough. Chaum’s eCash was much discussed on the list, admired for its cryptography and criticised for its dependence on a central bank. Members built anonymous remailers, which forwarded messages while stripping identifying information; Hal Finney and Eric Hughes wrote some of the earliest of them. Others wrote about how digital cash might work without any bank at all.

Some of the most important precursors of later systems were first posted to the cypherpunks list or circulated among its members: Adam Back’s Hashcash in 1997 and Wei Dai’s b-money in 1998, which we discuss in the next essay. Nick Szabo’s writings on smart contracts and bit gold were read and argued over by the same circle.

Disagreements and decline

The cypherpunks were never a single movement with a single programme. Members ranged from libertarians to civil-liberties campaigners to engineers who simply liked hard problems, and some ideas aired on the list were extreme and widely rejected even there. By the end of the 1990s, after disputes over moderation and the departure of many key members, the list had fragmented.

Legacy

The cypherpunks’ practical legacy is all around us: encrypted web connections, end-to-end encrypted messaging, anonymity networks, and the general assumption that ordinary people should have access to strong cryptography. Their influence on the idea of digital cash is equally clear.

When a pseudonymous author calling themselves Satoshi Nakamoto announced the Bitcoin whitepaper in October 2008, the announcement went to a different list, the cryptography mailing list run by Perry Metzger, where many former cypherpunks were active. Among the first to respond seriously and to run the software was Hal Finney, a veteran of the cypherpunk years. The dream of private money, discussed on a mailing list in the 1990s, had found a new form, although whether it matched what the cypherpunks had hoped for is still a matter of debate.