Fax: The Accidental Fortress

The Machine That Will Not Die

Somewhere in an American hospital right now, a nurse is standing next to a beige plastic box, watching pages feed through it one by one. Each page takes about six seconds. The machine makes a sound like a dial-up modem mating with a dot matrix printer. The document being transmitted is a referral for a cancer screening, and it is being sent this way because federal law says this is fine. More than fine. Compliant. Secure, even. Meanwhile, if the same nurse tried to email that document, she would need end-to-end encryption, a signed business associate agreement, audit trails, and enough paperwork to make the fax look positively modern.

Nine billion. That is how many fax pages American healthcare alone exchanges every year. Not nine billion in the nineties, when fax machines sat on every desk. Nine billion now. In an era of cloud computing, encrypted messaging, and blockchain medical records that Silicon Valley keeps promising will fix everything. Seventy to ninety percent of all communication between healthcare providers still travels by fax. Not email. Not a secure portal. Not a direct message in an electronic health record. Fax. The technology that a Scottish clockmaker invented thirty three years before Alexander Graham Bell picked up a telephone.

This is a mystery. Not a murder mystery or a spy thriller, but a genuine puzzle about how institutions work. Fax should be dead. Every obituary has been written. The National Health Service in England banned fax machines. Japan's most popular politician declared war on them. Every technology analyst in the world has called them obsolete. And yet the protocol hums on, processing billions of pages through cloud services that most people do not even know exist, wearing a disguise so thorough that the machine itself has nearly disappeared while the protocol it runs refuses to stop.

The question is not why fax was invented. The question is why it is still here. And the answer has almost nothing to do with the technology. It has everything to do with the decisions that institutions made around it, one by one, each reasonable in its moment, until the accumulated weight of those decisions became harder to move than the technology itself. But before we follow the trail of those decisions, we need to know what we are looking at.

The Clockmaker and the Pendulums

Alexander Bain was born in eighteen ten in the village of Watten, in Caithness, at the very top of Scotland. His father was a crofter, a tenant farmer working a small plot of land in one of the most remote parts of the British mainland. Alexander was one of thirteen children, six brothers and six sisters, though sources disagree on the exact count. He did not distinguish himself at school. He left at twelve to work alongside his father, and it was only at twenty that his talent for fixing things earned him an apprenticeship with a clockmaker in the nearby town of Wick.

Clocks were the high technology of the age. Precision mechanics, escapements, gear ratios, the physics of oscillation. Bain absorbed it all, moved to Edinburgh, and then in eighteen thirty seven to London, where he set up a workshop in Clerkenwell, the traditional watchmaking district. He attended public lectures at the Polytechnic Institution and the Adelaide Gallery, cramming his mind with the new science of electricity that was transforming everything. He was a tinkerer, a builder, the kind of person who saw a new principle and immediately asked what he could make with it.

On May twenty seventh, eighteen forty three, Bain received British patent number nine seven four five. The title was dry, as patent titles always are. Improvements in producing and regulating electric currents and improvements in timepieces, and in electric printing, and signal telegraphs. Buried inside that patent was something that would take the world a century and a half to fully exploit. Bain had described a method for scanning an image line by line using a pair of synchronized pendulums, converting each line into electrical signals, transmitting those signals over a wire, and reconstructing the image at the other end on chemically treated paper that darkened where the current flowed.

He had invented the fax machine. In eighteen forty three. The same year Charles Dickens published A Christmas Carol. Thirty three years before Bell's telephone. Twenty one years before the Geneva Convention. The typewriter had not been invented. The American Civil War had not started. Abraham Lincoln was a thirty four year old country lawyer in Illinois. And a Scottish clockmaker, working from the mechanics of pendulums and the new understanding of electrical conduction, had figured out how to send an image across a wire.

In eighteen forty eight, Bain traveled to America to seek his fortune. Scientific American reportedly observed working image transmission machines in his New York offices that year, though accounts differ on how successfully Bain's devices actually functioned. The images, by all accounts, were crude. The synchronization between the transmitting pendulum and the receiving pendulum drifted, smearing the output. But the principle was sound, and it was decades ahead of its time.

Bain died in eighteen seventy seven, the year after Bell's telephone patent, largely forgotten and mostly impoverished. He did not get rich from fax. He did not get rich from the electric clock, which he also invented. The world was not ready for either idea. But the idea survived him. An Italian physicist named Giovanni Caselli built a refined version called the pantelegraph, and in eighteen sixty five launched the first commercial fax service between Paris and Lyon. In its first year of operation, the service transmitted nearly five thousand faxes, at a peak rate of a hundred and ten per hour. Eleven years before the telephone existed, people in France were faxing each other. Bain's principle was elegant. But elegance does not explain survival. What kept fax alive was not the pendulums. It was what happened when institutions started making decisions around them.

The Standard That Locked the Door

For the next hundred years, fax technology existed in a strange twilight. Each decade brought an improvement. Shelford Bidwell built the first photographic scanner in eighteen eighty one. Arthur Korn transmitted the first long distance photograph in nineteen oh four. By the nineteen twenties, newspapers were using fax to transmit photographs across oceans, and the Associated Press had built a dedicated wire photo network. But fax remained a specialty tool, expensive, slow, and confined to newsrooms and government offices.

The machine that would colonize every office in the world did not arrive until nineteen sixty six, when Xerox released the Magnafax Telecopier. It weighed forty six pounds, connected to a standard telephone line, and could transmit a page in about six minutes. Before the Telecopier, fax machines were room-sized, temperamental, and required dedicated leased lines. After it, any office with a phone line could send a page to any other office with a phone line. The price of entry dropped from thousands of dollars to hundreds. But even the Telecopier was not the decision that locked fax into the infrastructure. That came fourteen years later.

In nineteen eighty, the International Telegraph and Telephone Consultative Committee, the body that would later become the International Telecommunication Union, published Recommendation T dot four, the Group Three fax standard. The standard was developed primarily by Japan's domestic telephone company, Nippon Telegraph and Telephone, and the overseas carrier known by its initials K D D I. It promised something revolutionary. Interoperability. Any Group Three fax machine, from any manufacturer, could talk to any other Group Three fax machine on earth. It compressed images digitally. It transmitted a page in about forty seconds instead of six minutes. And it used a handshake protocol called T dot thirty to negotiate the connection, agree on capabilities, and confirm delivery.

Remember that protocol name. T dot thirty. It will matter later. Because interoperability, once achieved, is an institutional decision that is almost impossible to reverse. Every fax machine that shipped after nineteen eighty spoke the same language. Every workflow, every legal filing, every hospital referral that was built on the assumption that any fax machine could reach any other fax machine, was built on T dot thirty. The standard did not just make fax convenient. It made fax load bearing.

The nineteen eighties and nineties were the golden age. Fax machines became as common as photocopiers. Every business card had a fax number. Every office supply store sold thermal paper. The sound of a fax handshake, that descending warble of tones, was as recognizable as a telephone ring. In Japan especially, where the Group Three standard had been born, fax became woven into the fabric of daily life in ways it never quite did in the West.

The Hanko and the Fax

To understand why fax will not die, you have to understand Japan. Not just the technology, but the culture of documentation that technology became entangled with.

The hanko is a personal seal, a small cylindrical stamp carved with a person's name in kanji characters. One of the first evidences of writing in Japan is actually a hanko, a solid gold seal dating from fifty seven AD, given to the ruler of Na by the Chinese Emperor Guangwu. For more than a thousand years, the hanko has been the Japanese equivalent of a signature. Not a metaphorical equivalent. A legal one. You stamp your hanko on contracts, marriage certificates, bank documents, employment agreements. The act of pressing a carved seal into red ink and then onto paper carries a gravity that a scribbled signature does not. In a culture that values formality and mutual commitment, the hanko is not a bureaucratic inconvenience. It is a ritual of trust.

And here is the thing about a hanko stamp. It exists on paper. To get a hanko onto a document, you need a physical document. To send a document with a hanko on it, you need a machine that can transmit physical documents. You need a fax machine. This is the loop that kept Japan faxing long after other countries started moving on. The hanko required paper. Paper required fax. Fax justified the hanko. Schools communicated with parents by fax. Hospitals sent records by fax. Government offices required faxed forms. In twenty twenty, a survey by the Ministry of Internal Affairs and Communications found that thirty three point six percent of Japanese households still owned a fax machine. Among people in their fifties, nearly half owned one. Elementary schools, government offices, and hospitals would often only communicate by fax. Then the pandemic arrived, and the cost of that loop became impossible to ignore.

The Three Day Fax

In the spring of twenty twenty, as the virus spread through Japan, public health centers became the front line of the national response. Doctors needed to report each new case to local health authorities. The reporting system was simple. A doctor would fill out a form by hand. Then fax it to the public health center. The public health center would compile the faxes, tabulate the numbers, and report them upward to prefectural and national authorities. When case numbers were low, this worked. When they were not low, it became a catastrophe.

Doctors were spending hours each day handwriting reports and standing by fax machines. Public health centers were drowning in paper, their fax lines jammed, busy signals echoing through offices where exhausted workers sorted through piles of thermal paper trying to count infections that had happened days earlier. The gap between a case being confirmed and the government knowing about it stretched to three days. Three days during which contact tracing could not begin. Three days during which the infected person was walking around untracked. Three days because the fastest communications technology that Alexander Bain could imagine in eighteen forty three was still the backbone of Japan's pandemic response in twenty twenty.

It took until May of that year, months after the virus had already gone global, for the Ministry of Health, Labour and Welfare to launch an online reporting portal. Many hospitals and health centers were unable or unwilling to switch. The fax machines kept whirring. Enter Taro Kono, the politician who thought he could change all of it.

The Minister Who Declared War

Kono was already one of Japan's most recognizable politicians, a former defense minister and foreign minister known for his blunt style, his fluent English, and a social media presence unusual for a Japanese lawmaker. In September twenty twenty, Prime Minister Yoshihide Suga appointed him minister for administrative reform, a post that sounds bureaucratic but was actually a license to challenge the way the Japanese government had always done things.

Kono wasted no time. He announced that he would take on the twin pillars of Japanese bureaucratic culture, the hanko stamp and the fax machine.

I do not think there are many administrative procedures that actually require printing out paper and faxing. Why do we need to print out paper? In many cases, it is simply because the hanko stamp is required. So if we can put a stop to that culture, then it will naturally do away with the need for printouts and faxes.

The ambition was staggering. Kono was not just proposing to replace hardware. He was proposing to unwind a cultural practice that stretched back more than a thousand years. The hanko was identity. The fax was how identity traveled. Pull one thread and the entire tapestry of Japanese bureaucratic procedure would have to be rewoven.

In June twenty twenty one, the Headquarters for the Promotion of Administrative Reform issued a directive to all central government ministries and their affiliated agencies. Stop using fax machines by the end of the month. Report back if there were reasons you could not comply. The responses poured in. Four hundred formal objections from government agencies, each explaining why they could not possibly give up their fax machines. Security concerns. Anxiety over the communication environment. Court procedures that required faxed documents. Police work that depended on the physical transmission of paper. Some agencies argued that the fax was actually more secure than email, that the point to point nature of the telephone network made interception harder. Others simply said they were not ready. The government backed down. The deadline passed. The fax machines stayed.

When Kono was later appointed digital minister, he broadened his crusade to include floppy disks, another technology that Japanese bureaucracy had kept alive long past its expiration date. He eventually declared victory on the floppy disk front in twenty twenty four. But the fax machine, the older enemy, the more entrenched one, remained undefeated. Speaking at a news conference, Kono captured his frustration in a single question.

Where does one even buy a floppy disk these days?

He might have asked the same question about fax machines. But the answer, in Japan at least, was everywhere. The floppy disk had no regulatory fortress around it. No legal procedures depended on it. No thousand year old identity system fed into it. Fax had all of those things, and that is why it survived where the floppy disk did not.

The Accidental Fortress

Now let us cross the Pacific and talk about the decision that did more than anything else to keep fax alive in the twenty first century. Not a cultural tradition. Not a personal seal. A legal interpretation.

When the United States Congress passed the Health Insurance Portability and Accountability Act in nineteen ninety six, it created a framework for protecting patient health information. The law required the Department of Health and Human Services to develop detailed rules. The Security Rule, finalized in February two thousand three under Secretary Tommy Thompson, required healthcare providers to implement safeguards when transmitting protected health information electronically. Email had to be encrypted. Electronic records had to be secured. Access had to be logged, controlled, and auditable. The penalties for violations were severe. One of the largest fines ever issued for faxing patient records to the wrong number was two point five million dollars.

But here is where the logic gets strange. The Department of Health and Human Services, through its Office for Civil Rights, drew a line between electronic transmissions and analog transmissions. An email is electronic. A file uploaded to a portal is electronic. But a piece of paper fed into a fax machine and sent over an analog telephone line? That, the regulators decided, was not an electronic transmission. Even though the fax machine digitized the page, modulated it into audio tones, and sent it across a telephone network that had been increasingly digital since the nineteen seventies. The reasoning was explicit in the preamble to the Security Rule. Copy machines, fax machines, and telephones, even those that contain memory and can produce multiple copies, were not intended to be covered. The information did not exist in electronic form immediately before the transmission. The paper was the source. The telephone line was just a conduit.

Think about what that means. In the eyes of federal regulators, the act of a fax machine converting a document into a signal and sending it across a telephone network was essentially the same as a voice telephone call. And since no one would argue that a doctor discussing a patient on the telephone needed to encrypt the phone call, no one could argue that faxing required encryption either. The logic was internally consistent, in the way that a building can be architecturally sound even though it was built on a swamp.

A facsimile machine accepting a hard copy document for transmission is not a covered transmission even though the document may have originated from printing from an electronic file.

The practical consequence was breathtaking. To send a patient record by email, a hospital needed to negotiate a business associate agreement, implement end-to-end encryption, maintain audit logs, train staff on the encryption software, and document the entire chain of custody. To send the same record by fax, the hospital needed to dial a number and press a green button. Guess which one they chose. Nine billion pages a year chose.

The Defenders and the Critics

This is the part of the story where we have to be fair. The logic behind the exemption was not entirely indefensible, and the people who argue that fax has genuine security advantages are not delusional.

The traditional public switched telephone network, the PSTN, creates a dedicated circuit between two points for the duration of a call. The signal does not bounce through servers or get cached in intermediate storage. Intercepting a fax on a PSTN line requires physical access to the telephone infrastructure, a wiretap on the actual copper wire. That is not easy. It requires proximity, equipment, and either criminal intent or a court order. Compare that to email, which traverses multiple relay servers, any one of which could be compromised, and which may sit in plaintext on a server thousands of miles from either the sender or the receiver.

There is truth in this argument. Point to point analog transmission is genuinely harder to intercept at scale than unencrypted email. A mass surveillance program can sweep up email traffic far more easily than it can tap thousands of individual phone lines. For an individual document going to a known recipient, the analog fax has a narrower attack surface than an email bouncing through a chain of relay servers. Doctors who defend fax on these grounds are not being irrational. As one physician put it to a medical journal, the fax is dumb, but it is direct. You dial a number, the document goes to one machine in one room, and you get a confirmation page.

But the argument has a hole in it large enough to drive a fax machine through. The physical security cuts both ways. Faxes arrive on paper, in a tray, often in a shared space. Anyone walking past can read them. Thirty percent of medical tests are reordered because faxes are lost, arrive on a busy signal, or simply sit in a tray until nobody remembers who they were for. One in five faxed referrals never receives a response. Eighty eight percent of healthcare practitioners say fax related delays negatively affect patient care.

And then there is the problem that demolishes the entire security defense. The PSTN, that dedicated circuit network that made the security argument plausible, has been dying for years. Telephone companies have been replacing copper lines with voice over internet protocol infrastructure. AT and T announced plans in twenty twenty four to retire the bulk of its copper lines by twenty twenty nine. The FCC issued orders in twenty twenty five streamlining the copper retirement process, waiving costly notice requirements, and allowing carriers to move faster. The copper sunset is not a theoretical future. It is underway. When a fax travels over voice over internet protocol, it is no longer a dedicated circuit. It is data packets on the internet, exactly like email, but without the encryption. The phone number authentication that the old network provided is gone too. In a voice over internet protocol world, caller ID information can be set to any value the sender chooses, making it trivially easy to impersonate a legitimate fax source.

The security advantage that regulators used to justify the exemption has been evaporating underneath the technology it was supposed to protect. And the Office for Civil Rights knows this. In June twenty twenty two, the office issued new guidance explicitly stating that voice over internet protocol communications are electronic and must comply with the Security Rule. The guidance names voice over internet protocol technologies alongside smartphone apps and messaging services as electronic communications requiring encryption safeguards. But the guidance did not revisit the fax exemption itself. It did not ask the obvious question. If voice over internet protocol calls require Security Rule compliance, and most fax transmissions now travel over voice over internet protocol infrastructure, then what exactly is the exemption still protecting? Nobody in the regulatory apparatus has publicly answered that question. Nobody appears to have formally asked it. The institutional decision made in two thousand three, when the PSTN was still a copper network and the analog exemption made technical sense, has simply continued by inertia, even as the technical foundation it rested on has been dismantled wire by wire.

Axe the Fax

The British tried a different approach. Blunt force. In the summer of twenty eighteen, the Royal College of Surgeons filed a freedom of information request with the National Health Service. The answer startled even the people who expected bad news. There were over eight thousand fax machines still in active use across the NHS. Eight thousand two hundred and nine, to be precise. The largest healthcare system in Europe, serving sixty seven million people, was running on technology that most businesses had abandoned in the early two thousands.

Matt Hancock, the Health and Social Care Secretary, seized the moment. Hancock had come to politics from a background in economics and data analytics, including time at the Bank of England and his family's software business. He was the kind of minister who used the word "interoperability" in speeches and meant it. He had already launched NHSX, a dedicated unit for NHS digital transformation, and described himself as obsessed with technology. When the fax numbers landed on his desk, he saw his opportunity. In December twenty eighteen, he announced a ban. No NHS organization could purchase new fax machines after January twenty nineteen. All existing fax machines would be phased out by March thirty first, twenty twenty.

We have got to get the basics right, like having computers that work and getting rid of the archaic fax machines still used across the NHS when everywhere else got rid of them years ago.

At a technology conference in Paris, a journalist asked Hancock how he would measure the success of innovation in the NHS. His answer was perfect.

When I am no longer the world's largest owner of fax machines.

He set the deadline. The deadline arrived. The NHS missed it. By October twenty nineteen, five months before the cutoff, only about forty two percent of fax machines had been eliminated. Then the pandemic hit, and suddenly no one was thinking about fax machine logistics anymore. The pandemic that exposed Japan's fax dependency also torpedoed England's plan to escape its own. Hancock, who described tech transformation as the only way the NHS could meet people's needs, ended up presiding over the same fax infrastructure he had vowed to destroy.

The Protocol That Shed Its Skin

Here is the twist in the story. When your doctor's office faxes your records to a specialist, there is a good chance that no fax machine is involved at either end. The sending system thinks it is talking to a fax machine. The receiving system thinks it is a fax machine. Neither is. Both are software running in a data center, speaking a protocol designed in nineteen eighty, pretending to be hardware that no longer exists.

The fax machine is dying. Global sales have been declining for decades. The beige boxes in hospital corners are aging out, their thermal print heads wearing down, their handset cradles cracking. Manufacturers have stopped making most models. But the protocol, T dot thirty, that handshake language that Group Three machines used to negotiate connections, is thriving. It just no longer lives inside a machine. It lives in the cloud.

Cloud fax is one of those industries that most people do not know exists. Companies like eFax, RingCentral, Consensus Cloud Solutions, and etherFAX operate vast digital infrastructure that receives fax transmissions, converts them to digital documents, and routes them to email inboxes, electronic health records, or web portals. No beige box. No thermal paper. No handset. The document arrives as a PDF attached to an email or injected directly into a hospital's record system.

The market is not small. Estimates vary, but the cloud fax industry is projected to be worth somewhere between one and a half billion and four point five billion dollars by the mid twenty thirties, depending on which analyst you ask and how they draw the boundaries. Paul Banco, the cofounder of etherFAX, saw the opportunity early. He cofounded the company in two thousand nine, betting that healthcare's regulatory dependence on fax would outlive the hardware. He was right. etherFAX's Secure Exchange Network now connects over six million endpoints, routing documents between hospitals, insurers, and pharmacies without a single telephone call.

And here is the beautiful irony. Many of these cloud fax services do encrypt the transmission. etherFAX uses what they describe as military grade encryption. When a fax travels through their network, it never touches the public telephone network at all. It goes from one data center to another through a closed, encrypted channel. The security argument that justified fax's regulatory privilege, the analog point to point circuit, has been replaced by something better. The protocol kept the compliance status it earned from the old telephone network while quietly upgrading to encryption that email providers have offered for years.

The T dot thirty eight protocol, published by the International Telecommunication Union in nineteen ninety eight, was designed specifically for this transition. Its job is to take the T dot thirty handshake, the one that Group Three fax machines have been speaking since nineteen eighty, and carry it over internet protocol networks. T dot thirty eight is, in essence, a translator. It fools the sending system into thinking it is talking to a fax machine on a telephone line, while actually routing the data through the internet. The fax machines at both ends believe they are having an analog conversation. They are not. They are packets on the wire, just like everything else. The protocol shed its hardware. Like a hermit crab moving to a new shell, T dot thirty crawled out of the beige box and into the cloud. The machine died. The protocol survived. And because the regulations were written around the protocol, not the machine, the compliance status traveled with it.

The Weight of Reasonable Decisions

So why does fax survive? Not because it is the best technology. Not because doctors love standing by beige machines. Not because Japanese bureaucrats are Luddites. Not because British health secretaries lack the will.

Fax survives because institutional decisions hardened around it. Each decision was reasonable in its moment. The Security Rule's drafters, writing regulations that Tommy Thompson signed in two thousand three, drew a sensible distinction between electronic and analog transmission at a time when fax really was analog and email really was insecure. Japan's hanko culture gave fax a purpose that no digital alternative could replicate without dismantling a thousand year old identity system. Legal systems accepted faxed signatures as valid because the technology produced a physical artifact with a transmission confirmation, something that early email could not reliably prove. Courts built procedures around fax filing. Hospitals built workflows around fax referrals. Insurance companies built claims processing around fax intake.

Each of these decisions became load bearing. Remove fax from American healthcare and you have to renegotiate compliance for every provider, rewrite every interoperability workflow, retrain millions of workers, and replace a system that, whatever its flaws, actually delivers nine billion pages a year. Remove fax from Japanese government and you have to resolve the hanko question, which means resolving questions of identity, formality, and trust that run deeper than any technology choice. Remove fax from the legal system and you have to replace a century of precedent about what constitutes valid delivery.

Nobody wants to do this. It is not that nobody can. It is that the cost of unwinding all those institutional decisions, all at once, is higher than the cost of keeping the fax machine plugged in for one more year. And then one more year. And then one more. Until the protocol has been alive for forty five years and the machine that created it is gathering dust in a museum while the protocol itself processes more pages than ever, humming invisibly through cloud data centers, wearing the compliance badge it earned in an era of copper wire and thermal paper.

Alexander Bain died in eighteen seventy seven, broke and forgotten. He never saw the Magnafax Telecopier. He never saw Group Three. He never saw nine billion pages flowing through hospitals. He certainly never imagined that the principle he patented, scanning an image line by line and transmitting it as electrical signals, would still be in daily use nearly two centuries later, long after the pendulums were replaced by lasers and the telegraph wires by fiber optic cables and the chemical paper by PDF files rendered on screens. But the principle endures. Not because it is elegant. Because it got tangled up in the way human institutions make decisions. And those decisions, once made, become the hardest technology of all to replace.

The fax machine is a fossil. But the protocol is alive. And as long as regulations say fax is compliant, as long as courts accept fax filings, as long as one hospital needs to send a referral to another hospital that uses a different electronic records system, the protocol will keep going. Not because anyone chose it. Because nobody chose to stop it. And in the world of institutional technology, that is the most powerful survival strategy of all.