Naval History and Heritage Command

Although she cracked a multitude of Japanese naval code systems, trained nearly all of the U.S. Navy’s leading World War II cryptanalysts, and is featured in the National Security Agency’s (NSA’s) Cryptologic Hall of Honor, Agnes Meyer Driscoll is still a relatively unknown figure in naval history.

Driscoll (née Meyer) enlisted in the U.S. Naval Reserve at age 28 as a yeoman (F) in 1918. Since she possessed a degree in mathematics and six years of teaching experience, she joined as a yeoman first class. (During World War I, the rate each yeoman (F) was given was based on that recruit’s prior work experience.)

She began her career at the U.S. Navy’s Office of the Chief Cable Censor, where she indexed, filed, and forwarded incoming and outgoing telegrams. In February 1919, she was promoted to chief yeoman and transferred to the Director of Naval Communication’s Code and Signal Section, located in Room 1645 of the Main Navy & Munitions Building on Constitution Avenue in Washington, DC.

It is unknown how her supervisors might have guessed she would excel at cryptography, but she proved to have a natural aptitude for it when she solved every machine cipher that was submitted to the Navy Department for adoption.

After leaving active duty in July 1919, she immediately began work as a Navy civilian in the same office. Although her official title was clerk/stenographer, she performed cryptologic work. Clerical positions were the only jobs available to civilians in the Code and Signal Section at that time. She worked her way up to senior administrative assistant before being promoted to cryptanalyst in 1929, according to her civilian personnel record.

She received formal training in cryptanalysis later that year, according to her January 1943 Application for Federal Employment. This training was likely at the Military Intelligence Branch Section 8’s (MI-8’s) Cipher Bureau, also known as the “black chamber.” This cryptanalytic organization was funded by the State and War Departments and led by Herbert Yardley, a former U.S. Army cryptologic officer who served in the Signal Corps during World War I.

In 1923, she left her government job to work in the private sector for Edward Hebern, a cipher machine inventor. Driscoll had co-developed a cipher machine with Lt. Cmdr. William F. Gresham, which went into production that year. She had developed the cryptographic principle for that machine, although Gresham had invented the machine itself. Hebern wanted to sell his own cipher machine to the Navy, but he needed to make it cryptographically secure first, so he hired Driscoll as a technical advisor. However, Hebern’s company failed to get enough federal contracts, and Driscoll returned to government service a year later.

When she returned to her job with the Navy in 1924, she was assigned to the Office of the Director of Naval Communication’s research desk—a deliberately vague name for the Navy’s new cryptanalysis component. It was widely known by its designator, OP-20-G. The five-person research desk consisted of four civilians led by Lt. Laurance Safford. Safford had just completed a tour of duty commanding a minesweeper on the Yangtze River and had no prior cryptanalysis training. Driscoll, by far the most qualified civilian cryptanalyst of the group, provided Safford with training. Safford later organized his own formal class in cryptology for naval officers assigned to cryptanalytic duties.

For the next 20 years, Driscoll’s codebreaking efforts centered on Japanese fleet operational codes. (A fleet code is a system that ships and shore installations use to exchange information on ship movements, strategy and tactics, logistics, situation reports, personnel, morale, and weather.) The first code she worked on was the Imperial Japanese Navy’s Secret Operations Code of 1918, or “Red Book.” In 1920, the Office of Naval Intelligence (ONI) obtained a copy of this code when a U.S. counterespionage squad broke into the Japanese consulate in New York and made a copy of it. ONI passed it on to the Office of Naval Communications, where it was kept in a red binder after it was translated from Japanese to English. The Red Book code contained nearly 100,000 entries of letters, numbers, words, and phrases with three different code equivalents for each—a five-digit number, an expression in Roman letters, and a three-character Kana (Japanese syllabic characters that are used for sounding out or writing foreign words).

Even with the code book in hand, American cryptanalysts could not read the encoded messages because the Japanese had used a method called super-encipherment. Essentially, the Japanese were using both a code and a cipher to disguise their message content.

Driscoll was the first to crack the code’s cipher. She discovered that the Japanese were using columnar transposition—Japanese coding clerks were writing the code groups out horizontally but transmitting them vertically, using a grid where certain squares were blacked out. Once she figured out the cipher, she created a transposition scheme to unlock the encryption. However, the Japanese changed the grid design periodically, so Driscoll had to continually find solutions and recover a key.

Safford and Driscoll also worked together on expanding OP-20-G’s domain into radio interception. They developed cryptographic systems for naval communications and arranged with the Commander in Chief, Asiatic Fleet, and various naval district commanders to obtain copies of radio intercepts of foreign messages. The Navy began building a network of intercept stations in the Pacific and the Far East so that cryptanalysts could analyze Japanese message traffic in real time.

In 1925, Lt. (j.g.) Joseph Rochefort arrived to take Safford’s place as the officer-in-charge of the research desk. Safford spent four months training Rochefort in collection and the rudiments of cryptanalysis. Once Safford left, Driscoll continued to train her new boss in codebreaking techniques so that he could take over as the research desk officer in charge. (Seventeen years later, Rochefort and his Pearl Harbor–based Station Hypo team would crack the Japanese JN-25B operational code and help secure the Navy’s victory at Midway.)

In Elliot Carlson’s Joe Rochefort’s War, Rochefort recalled that “Aggie” or “Miss Aggie,” (as she was known to some coworkers) made a lasting impression on him. He also found her extremely capable.

“I would say she was a first class cryptanalyst,” he said. “I considered her sort of a teacher to me.”

In December 1930, the Japanese Navy changed the Red Book to the “Blue Book” code. By then, Rochefort had been transferred to sea duty, and Lt. Cmdr. Safford returned to OP-20-G as the head of the research desk. When Safford left in 1932, Lt. (j.g) Thomas Dyer took over as officer in charge of OP-21-G (Dyer later went on to serve as Rochefort’s assistant officer-in-charge at Station Hypo).

Once again, Driscoll trained her new officer-in-charge. Dyer described Driscoll as being “absolutely brilliant” and “at the top of the list [of Navy cryptanalysts]”, according to Kevin W. Johnson’s book, The Neglected Giant: Agnes Meyer Driscoll.

Dyer created new procedures for automating code breaking. He began using IBM machines (key punches, tabulators, and sorters) that compared and overlapped multi-digit code groups. The machines could transpose input, alphabetize, and sort through multitudinous possible solutions. In contrast, Driscoll preferred using pencils and graph paper to solve codes rather than automation. In fact, it was through manual methods that she made the first break in the Blue Book code in 1933. Her success in solving it enabled the Navy to gather valuable intelligence on the Japanese fleet. For instance, decrypted message traffic revealed that the Imperial Navy’s new battle cruiser Nagato had a top speed of slightly more than 26 knots. With this intelligence in hand, the U.S. Navy changed the planned top speed for its new battleships North Carolina (BB-55) and Washington (BB-56) to 27 knots (it had previously been 24 knots).

In October 1937, Driscoll was temporarily sidelined after being in a serious car accident that left her with a broken jaw and right leg. She was unable to return to work until nearly a year later. Her leg didn’t heal properly and remained bowed since it was not immediately put in a cast. From then on, she had to use a cane.

The same year she returned, the Japanese began using a new code for encrypting messages—the “Black” code. This code didn’t last long, however. A year later, in June 1939, the Japanese replaced it with a flag officer’s code and their main operational code, JN-25.

JN-25 was a five-digit enciphered general-purpose code that did not use a transposition system like the previous codes. This time, the Japanese used numbers instead of characters and a codebook with 30,000 five-digit groups. In addition, the Japanese used what was called an additive book, which contained 300 pages of 100 numbers each. The Japanese code makers would add a number from the additive book to each code group in the message using a mathematical method called non-carrying addition. By this point, Driscoll was a senior cryptanalyst in OP-20-G and the resident expert in Japanese fleet codes. As such, she was placed in charge of breaking the JN-25 code. A small team consisting of two civilians, two enlisted personnel, and one officer aided her in the effort, though she did the majority of cryptanalytic work. She did not have time to completely solve the JN-25 code, although she did unravel the system the Japanese were using to encipher it. Unfortunately, the Japanese changed the JN-25 code again in December 1940.

In October 1940, Driscoll was transferred to the Anglo-American Enigma project, the effort to break the primary German operational naval code. This was perhaps considered a promotion since stopping the Germans from attacking Allied shipping interests in the Atlantic had become a higher priority for the U.S. Navy. If the American and British codebreakers could break the Enigma code—used on the German Enigma code machines—they would have access to the Enigma-enciphered radio messages that Admiral Karl Dönitz’s U-boat force was using to communicate with its submarines that were attacking Allied convoys.

It was perhaps Driscoll’s insistence on not using machines to aid her that kept her from solving the Enigma cipher in a timely manner. British codebreakers at Bletchley Park offered her a so-called Bombe, which she refused—for reasons unknown. The Bombe was an electromechanical machine used by the British to help decipher Enigma messages.

She did eventually solve Enigma without employing a machine; however, her solution was resource-heavy and slow. She constructed a paper analog of the German Enigma machine with three sliding paper rotors containing 26 letters on each wheel. The operation of the paper analog was all done by hand, and so labor-intensive it was never used during wartime. Driscoll was eventually moved from the Enigma project once the U.S. Navy began building its own Bombes.

However, Driscoll did contribute to the success of the Navy’s Enigma-breaking machines by helping to design one that automated the operation of the Navy’s Bombes. The Driscoll-Howard Standard Grenade reduced the effort needed to identify the window settings for the Bombes.

Driscoll worked on one more Japanese cipher machine (used by the Japanese naval attaché) before moving on to an assignment in the Russian Language Section (OP-20-G-50) in 1944. There, she led a small group of four civilians and three enlisted WAVES in decrypting Soviet diplomatic messages as part of the Venona project. This was a top-secret counterintelligence program initiated by the U.S. Army during World War II whereby the United States sought to gather and decrypt Soviet diplomatic traffic for intelligence purposes. The project was later absorbed by the NSA and continued another 30 years, until 1980. Driscoll’s team did not have any significant successes on the project.

In 1949, when all military communications intelligence and security organizations were centralized into one national organization, Driscoll and the rest of the Navy cryptology contingent moved to the Armed Forces Security Agency (AFSA). In 1952, the AFSA was redesignated as NSA.

At NSA, Driscoll worked in the Technical Projects/Services Group, a research office for NSA operations and for several other divisions under the NSA umbrella. She maintained her same habits of favoring manual over machine analysis, and coworkers remembered seeing her conducting research work using a huge magnifying glass mounted on a special fixture to analyze the hardcopy files, which were delivered to her in “endless boxes,” according to The Neglected Giant.

She continued to work on deciphering unreadable communications until she retired from NSA on July 31, 1959, at the federal civil service grade of GS-13. She was 70 years old and had 40 years of government service.

Agnes Meyer Driscoll died in 1971 and is buried in Arlington National Cemetery.

     —Wendy E. Arevalo, NHHC Communication and Outreach Division

Further Reading

“A Short History of U.S. Navy Information Warfare.” United States Naval Academy. Accessed 4 June 2024. Cryptologic Warfare History: Information Warfare (usna.edu).

Akers, Regina. “A Historical Overview of the Yeomen (F),” Naval History and Heritage Command, 28 Jan. 2019. https://www.history.navy.mil/browse-by-topic/wars-conflicts-and-operations/world-war-i/people/historical-overview-of-yeomen-f.html.

Burke, Colin. It Wasn’t All Magic: The Early Struggle to Automate Cryptanalysis, 1930s–1960s, Center for Cryptologic History. National Security Agency. Fort Meade, MD, 2002. Accessed 11 June 2024, https://archive.org/details/NSA-WasntAllMagic_2002/page/n4/mode/1up?q=grenade&view=theater.

Carlson, Elliot. Joe Rochefort’s War: The Odyssey of the Codebreaker Who Outwitted Yamamoto at Midway. Annapolis: Naval Institute Press, 2011.

Holtwick, Jack S., ed. “Naval Security Group History to World War II,” part 1 (June 1971). Navy Department Library.

Jacobsen, Philip H. “Navy Cryptology and the Battle of Midway: Our Finest Hour.” Sextant (blog). Naval History and Heritage Command, 14 June 2013. Navy Cryptology and the Battle of Midway: Our Finest Hour > The Sextant > Article View (dodlive.mil).

Johnson, Kevin W. The Neglected Giant: Agnes Meyer Driscoll. Center for Cryptologic History Special Series 10. Fort Meade, MD: United States National Security Agency/Center for Cryptologic History, 2015, Neglected Giant-FINAL-no classification.indd (defense.gov).

Mundy, Liza. Code Girls. New York: Little, Brown and Company, 2018.

“Navy Cryptology: The Early Days.” Historical Events, National Security Agency (NSA) /Central Security Service. 20 Aug. 2021, https://www.nsa.gov/History/Cryptologic-History/Historical-Events/Article-View/Article/2740632/navy-cryptology-the-early-days/.

“The Origin and Evolution of Radio Traffic Analysis: The Period between the Wars,” Cryptologic Quarterly 6, no. 3/4 (Fall/Winter 1987–1988): 25, https://media.defense.gov/2021/Jun/29/2002751711/-1/-1/0/THE_PERIOD_BETWEEN_WARS.PDF.

Parker, Frederick D. Pearl Harbor Revisited: United States Navy Communications Intelligence 1924–1941. Vol. 6. Series IV: World War II. Fort Meade, MD: Center for Cryptologic History, National Security Agency, 2013. https://www.history.navy.mil/research/library/online-reading-room/title-list-alphabetically/p/pearl-harbor-revisited-usn-communications-intelligence/part-1.html.

Zullo, Matt. “On-the-Roof Gang: The U.S. Navy’s Cryptologic Pioneers.” Naval History, June 2022. On-the-Roof Gang: The U.S. Navy’s Cryptologic Pioneers | Naval History Magazine - June 2022, Volume 36, Number 3 (usni.org).