“The woman who read everyone else’s secrets, then vanished from the story”
You may not know the name Elizebeth Smith Friedman.
That is not evidence that she was unimportant.
It is evidence that history, like any badly governed information system, can lose critical records while continuing to insist that the process worked exactly as designed.
There is no EU AI Act markup on this particular article. Elizebeth has been so invisible for so long that the only images I am using in this, Artefact 013, are real photographs of her, not AI generated images. There are not very many.
Elizebeth Friedman helped create American cryptanalysis as a profession. She broke the communications of international smugglers, dismantled criminal networks, testified as an expert witness, trained codebreakers, exposed narcotics traffickers, tracked Nazi spies and worked against clandestine Enigma systems. She turned intercepted radio chatter into arrests, convictions, diplomatic evidence and wartime counterintelligence.
Yet she is still commonly introduced as the wife of William Friedman.
This is particularly absurd because she introduced William to cryptology.
The historical record has managed to place the person who opened the door behind the man who followed her through it. Even by its own standards, this is impressive work.1
Elizebeth, with an e
Elizebeth Smith was born in 1892, the youngest of nine children in a Quaker family in Indiana. Her unusual spelling was deliberate: her mother reportedly wanted to prevent anyone shortening her name to Eliza.
She studied English literature at Hillsdale College, along with Latin, Greek and German.
She briefly worked as a school principal and then travelled to Chicago in 1916, hoping to work somewhere that would bring her closer to books and literature. At the Newberry Library, her interest in an original Shakespeare folio led indirectly to a meeting with the wealthy and spectacularly eccentric George Fabyan.2
Fabyan owned Riverbank, a large private estate and research laboratory near Geneva, Illinois. His collection of researchers investigated acoustics, genetics, cryptography and whatever else had caught his attention before breakfast.
He was also convinced that Francis Bacon had written Shakespeare’s plays and concealed proof of this inside the printed texts using ciphers.
This was nonsense.
Productive nonsense, as things turned out, but nonsense nonetheless.
Fabyan hired Elizebeth to work with Elizabeth Wells Gallup, who claimed that differences in the typefaces used in early editions of Shakespeare encoded Baconian messages. Elizebeth approached the problem seriously, learned how ciphers worked and gradually concluded that Gallup was finding patterns because she expected patterns to be there.
Long before cybersecurity acquired machine learning, it had already discovered the danger of training the analyst on the desired answer.
At Riverbank, Elizebeth met William Friedman, a geneticist and photographer who had been assigned to produce images of the Shakespeare texts. Their shared scepticism about the Bacon theory developed into a shared fascination with cryptography. They married in 1917.
The popular shorthand calls William Friedman a foundational figure in American cryptology.
He was.
But Elizebeth was not standing beside him holding his coat. She was already doing the work, and it was through her that he entered the field. The NSA’s own biography states this plainly, although even that biography begins by describing her through a list that includes “wife” before reaching “cryptanalyst”. Some habits have impressive cryptographic resistance.
America’s accidental cryptologic laboratory
When the United States entered the First World War, there was little organised federal cryptanalytic capability. Riverbank effectively became a national cryptologic laboratory, solving government messages and producing training materials for military personnel until formal government organisations could assume the work.
Elizebeth and William taught themselves by doing. There was no established American career path called cryptanalyst, no mature profession to enter and no helpful recruitment portal asking whether applicants had five years’ experience with a discipline that barely existed.
They collected cipher systems, studied methods, solved real messages and helped turn secret writing from an antiquarian curiosity into an analytical trade.
After the war, they moved to Washington and worked for the War Department. But the paths of the two Friedmans soon diverged. William remained within military cryptology and eventually became one of its most celebrated institutional figures.
Elizebeth moved towards a rather different battlefield.
Alcohol.
Criminals discover radio
The United States enacted Prohibition in 1920, thereby demonstrating the enduring governmental technique of banning a popular commodity and acting surprised when organised crime develops a distribution network.
Smuggling became an international industry. Large “mother ships” carried alcohol from the Caribbean, Canada and elsewhere to positions beyond US territorial waters. Smaller boats collected the cargo and brought it ashore. Networks of suppliers, ship captains, shore stations, corrupt officials and criminal organisations coordinated movements across thousands of miles.
Radio transformed these operations.
Ships and shore stations could exchange instructions rapidly, redirect cargoes, warn of Coast Guard patrols, arrange fuel and coordinate landings. Radio also presented a small drawback: anyone with a suitable receiver could listen.
The rumrunners therefore used codes and ciphers.
The Coast Guard accumulated hundreds of intercepted messages but lacked the capability to read them. In the mid-1920s, its entire cryptanalytic operation effectively consisted of Elizebeth Friedman, a clerk, pencils and paper.3
The criminals were not all using nursery substitutions.
Their systems evolved as operations grew more profitable. They used codebooks, substitutions, transpositions, enciphered codes and frequent changes intended to prevent earlier solutions remaining useful. Different stations employed different systems, and a recovered message might contain vessel identifiers, quantities, locations and operational phrases disguised behind layers of transformation.
Elizebeth attacked the systems structurally.
She studied repetitions, message lengths, call signs, recurring positions, likely operational vocabulary and relationships between separate transmissions. A ship requesting fuel yesterday might use the same format when requesting supplies tomorrow. A repeated group might represent a location, a vessel or a predictable phrase. Traffic intercepted across a network could illuminate meanings that one isolated message concealed.
Cryptanalysis did not require clairvoyance.
It required patience, language, pattern recognition, disciplined hypotheses and the willingness to remain irritated by a group of characters until it confessed.
Twelve thousand messages
Elizebeth and her small team eventually solved around 12,000 encoded rumrunner and smuggling messages. NSA records credit this work with contributing to approximately 650 criminal prosecutions, while she personally appeared as an expert witness in 33 cases.4
Those numbers matter, but they obscure what made her work unusual.
Military intelligence could often remain secret. Criminal prosecutions required evidence and public scrutiny.
Elizebeth could not merely tell investigators that a vessel was carrying contraband. She had to show how an intercepted sequence of apparently meaningless groups became an operational instruction. Defence lawyers could question her methods, qualifications and conclusions in open court.
She therefore had to do something many technical experts find considerably harder than solving the original problem:
She had to explain it clearly to people who had not spent years studying it.
A class in cryptology
In 1933, Elizebeth appeared as the government’s principal cryptanalytic witness during the prosecution of members of the Consolidated Exporters Corporation, known as Conexco, a major international smuggling organisation.
The evidence included hundreds of messages passed between suppliers in British Honduras, headquarters in New Orleans, radio stations in the Louisiana bayous and a fleet of ships waiting offshore.
A defence lawyer attempted to undermine her expertise.
Elizebeth asked for a blackboard.
She then demonstrated the solution of a randomly selected section of cipher, recovering the message:
Anchored in harbour, when and where are you sending fuel?
Reporters described what followed as a “class in cryptology”. The performance helped link the apparently abstract code groups to the physical ships and criminal organisation on trial. Senior members of the network were convicted.5
This is a magnificent piece of courtroom theatre, but the important point is not that she humiliated a lawyer. Lawyers have built natural resilience against this; they get paid win or lose; heads they win, tails you lose.
The important point is that she transformed cryptanalysis into legally comprehensible evidence.
She established that intercepted communications could be collected, analysed, explained and connected to real-world activity strongly enough to survive adversarial challenge.
Modern digital forensics follows the same path:
- acquire the data;
- preserve its provenance;
- analyse it reproducibly;
- connect it to actions and identities;
- explain it without requiring the court to become a specialist laboratory.
The technology changes.
The evidential burden does not.
The case of I’m Alone
The wonderfully named I’m Alone was a Canadian-registered schooner used as a rum-running mother ship. In 1929, after refusing to stop for the US Coast Guard, it was pursued and sunk. One crew member died, and Canada sought damages from the United States.
The dispute turned partly on ownership. If the vessel was genuinely Canadian, the diplomatic and legal consequences were serious. If it was secretly controlled by Americans using Canadian registration as cover, the case looked rather different.
Elizebeth decrypted 23 messages connected with the vessel’s operations. The traffic helped establish that American interests owned and controlled it, despite the Canadian flag fluttering helpfully above the enterprise.
Cryptanalysis had moved from catching speedboats to resolving an international dispute.6
This is worth pausing over.
The cipher system did not merely conceal cargo movements.
It concealed relationships:
- who owned the vessel;
- who issued instructions;
- which organisations were connected;
- where authority actually sat beneath the visible paperwork.
That is still one of the most important purposes of communications intelligence. Content tells you what somebody said. Patterns of communication tell you what the organisation is.
And organisations are often most interesting precisely where their official diagrams become fictional.
Solving Chinese without speaking Chinese
In 1937, Canadian authorities requested Elizebeth’s help with an opium-smuggling investigation involving coded messages written using Chinese words.
She did not speak Chinese.
She solved the cryptographic system anyway.
A Chinese-language specialist could then translate the recovered content, helping secure convictions against members of the network.
This sounds almost supernatural until one remembers that cryptanalysis and translation are different tasks.
A cryptanalyst may identify repeated units, message structures, substitutions and relationships without knowing what the recovered language means. The cipher imposes one structure; the language supplies another. Remove the artificial structure and someone else can interpret the natural one.
It is a useful lesson in specialist collaboration.
One person does not need to understand everything.
The system needs to bring together people who understand different necessary things without insisting that one of them grow three additional careers before lunchtime.
From rumrunners to Nazis
By the end of the 1930s, the problem had changed.
Elizebeth’s unit began following communications associated with German ships and clandestine intelligence networks. During the Second World War, the Coast Guard cryptanalytic organisation, often identified as Unit 387, worked against German espionage traffic, particularly networks operating across Latin America.
These were not smugglers asking when the whisky would arrive.
German agents gathered information about Allied shipping, military movements, industrial production and political conditions. Clandestine radio stations connected local agents with German intelligence. Some networks used increasingly complex cipher systems, including Enigma-based communications.
Elizebeth’s team broke thousands of transmissions associated with German military intelligence and helped map networks of agents, radio operators and handlers spread across South America and beyond.
The scale changed, but much of the craft remained recognisable.
Call signs changed.
Operators made mistakes.
Messages repeated structures.
Networks had habits.
Agents had schedules, relationships and administrative needs.
Even spies must occasionally ask headquarters for money.
The Queen Mary
In March 1942, Elizebeth’s unit recovered intelligence showing that German submarines were attempting to locate the liner Queen Mary, then carrying roughly 8,400 American troops.
German authorities had reportedly offered a million-Reichsmark reward for sinking it.
The ship escaped its hunters.
One decrypted message does not by itself win a war.
But intelligence does not need to win a war single-handedly to matter.
Sometimes it needs to alter one route.
Delay one departure.
Warn one ship.
Find one transmitter.
Arrest one operator before the next message is sent.
The grand history of intelligence is built from thousands of decisions whose success is measured by events that did not happen.
No sinking.
No convoy intercepted.
No agent report delivered.
No headline.
Prevented outcomes are notoriously bad at building reputations.
Hoover discovers publicity
The FBI entered the counter-espionage story with considerably greater institutional visibility than cryptanalytic competence.
In 1938, following a poorly handled German espionage investigation, FBI Director J. Edgar Hoover asked Elizebeth to train agents assigned to South America. Her Coast Guard unit possessed radio-interception and codebreaking capabilities the FBI did not.
Elizebeth and her team subsequently supplied decrypted intelligence to the Bureau.
Hoover then presented successes arising from that work as FBI achievements. According to the Coast Guard’s official history, he publicly claimed credit for spy-catching operations, failed to acknowledge Friedman’s unit and revealed enough about the compromised German communications that the enemy changed its systems.
In other words, the intelligence was used.
The source was exposed.
The adversary adapted.
Artefacts 011 and 012 have just spent considerable time explaining why one should not do this. Learning the lessons of history is a lesson from history that needs to be errr learned.
Apparently Hoover had skipped ahead to the publicity section.
After further premature arrests and public claims, German networks changed codes and moved key personnel. Elizebeth’s unit broke the revised systems as well, but details were then withheld more carefully from Hoover’s organisation. The resulting intelligence helped close clandestine stations and dismantle Axis networks across Latin America.
This is more than an entertaining story about an objectionable man appropriating credit, although it is certainly that.
It is an operational-security lesson.
Intelligence consumers can endanger intelligence sources.
The people receiving the product may not understand the fragility, cost, nor uniqueness of the collection method. They may see information only as something to be acted upon, circulated, briefed or converted into a press release bearing the correct agency crest.
Protecting intelligence therefore requires controlling not merely who sees it, but how recipients may use it.
Classification is not just secrecy.
It is behaviour management for officials who believe the laws of consequence apply mainly to other departments.
The Doll Woman
Elizebeth also contributed to the case of Velvalee Dickinson, a New York doll-shop owner who passed information to Japan using apparently innocent correspondence about dolls.
The letters were examples of open code. References to dolls, families and repairs were suspected of representing ships, fleets and naval damage. Elizebeth analysed the messages, identified likely hidden meanings and provided expert guidance connecting the domestic language to military information.
Dickinson eventually pleaded guilty and received a prison sentence.
Public accounts credited the FBI or unnamed government cryptographers. Elizebeth largely disappeared from the story.
The case illustrates a different kind of cryptography.
The message was not necessarily transformed into incomprehensible symbols. It was disguised as an ordinary conversation.
This remains deeply familiar. Criminals and spies still use coded vocabulary, euphemisms, memes, product names and apparently harmless discussions to convey meaning understood only within their group.
Encryption hides the text.
Open code hides the significance.
Sometimes the most dangerous message is perfectly readable.
Why did she disappear?
Elizebeth was not entirely unknown during her working life.
During Prohibition, newspapers and magazines reported on her courtroom appearances and remarkable career. Unfortunately, much of the coverage treated her as a novelty: a woman who could, astonishingly, perform difficult intellectual work while apparently remaining recognisably female. The horror!
Her skill made her newsworthy.
Her sex was treated as the surprising part.
Then several forces combined to erase her.
Secrecy
Her wartime work was classified. She signed commitments preventing her from discussing it and died in 1980 before the full record became public. Important files concerning Unit 387 and her work against Nazi espionage were not declassified until decades later, with significant material emerging in the early twenty-first century.7
Institutional credit
Cryptanalytic results passed to investigators, prosecutors, military services and the FBI. The arresting or announcing agency became visible; the analysts who made the evidence intelligible remained behind the classification boundary.
Her initials, ESF, appeared at the bottom of decrypted messages. The public saw the institution whose stamp was placed above them.8
Hoover
The FBI had a publicity machine and a director with no documented aversion to using it. Hoover claimed or absorbed credit for operations dependent upon work performed by Elizebeth and her Coast Guard team. She was forbidden from correcting the public record. The asymmetry of power is shocking.
William
William Friedman’s achievements were substantial, better institutionalised and gradually declassified. He received major awards and became widely described as a father of American cryptology and of the organisation that became the NSA.
Elizebeth increasingly appeared as his wife, collaborator or supporting biographical detail.
The NSA now explicitly acknowledges that she had a major and distinct career and that she introduced him to cryptology, but the older framing proved remarkably persistent.
Gender
She led men, trained investigators, built a cryptanalytic unit, testified as a national expert and solved systems that defeated whole agencies.
Yet official authority was often described through male supervisors, agency directors and uniformed recipients. As a civilian woman moving among military, Treasury and law-enforcement organisations, she was easily filed under somebody else by institutional histories.
Nobody needed to organise a conspiracy to erase her.
They merely needed to follow the normal assumptions about who counted as the principal and who counted as assistance.
Systems can discriminate very effectively without ever holding a meeting to approve the design. We, today, now, need to be both cognisant of this and actively defend against it if we are ever to overcome these biases.
Footnote to History
Elizebeth began an unfinished memoir.
She titled it Footnote to History.9
The title feels modest.
It also feels devastating.
She had spent decades extracting meaning from material designed to conceal it. She understood better than most that absence from the visible text does not mean absence from the underlying event.
A footnote may contain the source upon which the entire paragraph depends.
She was not a colourful supporting character in William Friedman’s life.
She was not the woman who helped the Coast Guard with a few difficult messages.
She was not a convenient technical service beneath the FBI’s public victories.
She was a founder of operational cryptanalysis in the United States: a codebreaker, teacher, team leader, intelligence analyst, expert witness and builder of institutional capability.
The outline for this series calls her an essential overlooked figure and links her work not only to cryptography but to prosecution, testimony, intelligence analysis and the habit of forgetting who actually did the work. That is exactly why she belongs here.
The “Footnote to History” was neither finished nor published. It is far from trivial to locate and obtain. I have done so, because I could. Should you wish to read it – see below the footnotes.
The artefact is the pipeline
The artefact is not one particular rumrunner cipher.
It is the system Elizebeth built around intercepted communications:
radio monitoring, collection, traffic analysis, cryptanalysis, training, corroboration, law-enforcement action, courtroom evidence and counterintelligence.
She demonstrated that codebreaking could become a repeatable public capability rather than an occasional act of individual cleverness.
She also demonstrated a less cheerful principle:
Technical work can be indispensable to an institution while the person performing it remains dispensable to the institution’s preferred history.
The rumrunners transmitted.
The spies transmitted.
Elizebeth read them.
The Coast Guard acted.
The FBI held the press conference.
History remembered the men.
We can correct that.
Elizebeth Smith Friedman.
With an e.
Despite the invisibility, she made a long-lasting difference.
00001101
References & Links
- https://www.nsa.gov/History/Cryptologic-History/Historical-Figures/Historical-Figures-View/Article/1623028/elizebeth-s-friedman/ ↩︎
- https://www.nsa.gov/History/Cryptologic-History/Historical-Figures/Historical-Figures-View/Article/1623028/elizebeth-s-friedman/ ↩︎
- https://www.history.uscg.mil/Research/THE-LONG-BLUE-LINE/Article/2925266/the-long-blue-line-mrs-friedmanthe-coast-guards-cryptologist-in-charge-and-nsc/ ↩︎
- https://www.nsa.gov/Press-Room/News-Highlights/Article/Article/2299166/pioneering-codebreaker-elizebeth-friedman-honored-by-us-coast-guard/ ↩︎
- https://www.history.uscg.mil/Research/THE-LONG-BLUE-LINE/Article/2925266/the-long-blue-line-mrs-friedmanthe-coast-guards-cryptologist-in-charge-and-nsc/ ↩︎
- https://www.nsa.gov/History/Cryptologic-History/Historical-Figures/Historical-Figures-View/Article/1623028/elizebeth-s-friedman/ ↩︎
- https://www.womenshistory.org/education-resources/biographies/elizebeth-smith-friedman ↩︎
- https://www.smithsonianmag.com/blogs/smithsonian-american-womens-history-museum/2026/07/13/meet-elizebeth-smith-friedman-the-us-governments-first-woman-codebreaker-she-cracked-nazi-codes-saved-8000-lives-and-pioneered-modern-codebreaking/ ↩︎
- https://www.smithsonianmag.com/blogs/smithsonian-american-womens-history-museum/2026/07/13/meet-elizebeth-smith-friedman-the-us-governments-first-woman-codebreaker-she-cracked-nazi-codes-saved-8000-lives-and-pioneered-modern-codebreaking/ ↩︎
©️1977-2026 Sophie Baskerville 