Navy attack aircraft ambushed the Imperial Japanese Navy’s aircraft carrier force 70 years ago and thus helped shape the rest of World War II in the Pacific. Many factors affected the operation. An indispensable one was the ability of U.S. communications-intelligence officers to decrypt Japanese radio messages. The information they provided enabled carrier-based dive-bombers to be in position northwest of Midway Atoll to deliver a devastating attack and sink all four enemy aircraft carriers. The rest of the Japanese force, part of which was intended to invade Midway itself, slunk home in defeat.
Much of the credit for the intelligence coup has rightly gone to Commander Joseph Rochefort, who led a team of code-breakers in the dank basement of a building at Pearl Harbor. In 2011, Elliot Carlson produced the acclaimed biography Joe Rochefort’s War, which laid out in great detail Rochefort’s achievement in predicting the forthcoming Japanese fleet movements. But the success was really a team effort that also involved the Pacific Fleet intelligence officer, Lieutenant Commander Edwin T. Layton; the gutsy Fleet Commander-in-Chief, Admiral Chester W. Nimitz, who directed U.S. carrier deployments; and Rochefort’s skilled team of subterranean cryptanalysts.
The lead code-breaker on the team was Lieutenant Commander Thomas H. Dyer, who served at Pearl from 1936 to 1945, eventually rising in rank to captain. Nearly 30 years ago—a little more than a year before his death in January 1985—Dyer participated in a Naval Institute oral history. In the autumn of his life, the 81-year-old retired officer was freed from restraints that had earlier prevented him from talking openly about his wartime experiences. He scrupulously censored himself by discussing only unclassified subjects, which meant he would talk about nothing substantive for the period after the war. Among other things, he was involved in site selection in Maryland for what became the National Security Agency. One can surmise that the target of his postwar work was the Soviet Union, but he wouldn’t discuss that in the interviews.
The main impressions that linger from those interviews were of Dyer’s professional competence and his wry sense of humor—a perpetual twinkle in his eyes. He was short in physical stature, and the most noticeable features of his face—besides those twinkling eyes—were his glasses and his white goatee. He could have done well in a Colonel Sanders look-alike contest. Many of his recollections were delivered with a smile—a considerable contrast to the sober-looking official photograph of him at left.
Solving Puzzles
The road to Dyer’s eventual rendezvous with the Japanese codes originated many years earlier with what was at the time the selection of a routine shipboard assignment. As a newly commissioned ensign in 1924, he reported to the battleship New Mexico (BB-40) and planned to serve in the gunnery department. But an enthusiastic shipmate persuaded him to go into radio, and Dyer thought it would be useful to learn about vacuum tubes. In the course of his duty, he saw the monthly bulletins that emerged from the Navy’s Communication Division in Washington.
At the time, crossword puzzles were becoming popular. Lieutenant Laurance Safford, the officer in charge of OP-20G, the communication research desk at the Navy Department, inserted a cryptogram in the form of a crossword puzzle in each issue of the bulletin. Dyer successfully solved one, sent his answer to Washington, and was encouraged to do more. He did so well over the ensuing months that in 1927 he was ordered to OP-20G for a course in the making and breaking of codes. The puzzles he had solved as a pastime changed the entire scope of his career.
When he arrived at OP-20G, he first met Lieutenant Rochefort, who was the section head. The training course was rather basic: a few pamphlets on codes and a supply of paper and pencils. As Dyer put it, “There was not much in the way of instruction. In my opinion, you cannot develop an original cryptanalyst by telling him how to do everything and then put him up against a completely unknown problem.” That observation was apparent when he and two fellow students first tried to decode Japanese navy radio traffic. Initially, they couldn’t make any sense of it.
Independent of the training efforts, naval intelligence had other tools as well. One was gaining access to a physical copy of the Japanese fleet code. Dyer was not part of the operation but recounted learning from other sources that “Certain individuals went in at night and borrowed it and photographed it.” The purloined source contained three-character substitutes for thousands of terms. It came in handy but was no giveaway, because radio transmissions also included a cipher on top of the code, and codes changed over the years.
Dyer Teams with Rochefort
After that duty, Dyer went back to sea and wound up on a fleet staff that also included Rochefort. They pooled their efforts in communications intelligence during a large fleet problem in 1929. As Dyer remembered, that was the last time Rochefort was involved in code breaking until he reported to Pearl Harbor in 1941. In the meantime, Rochefort went to Japan to study the Japanese language. Dyer made it clear that he in no way denigrated Rochefort but said that his leader’s real forte was in analysis of the contents of messages, not in original code breaking: “He could take a message that was only half there and mentally fill in the blanks to the extent of getting the sense out of it.” Both men benefited in the intelligence field because their operational experience at sea gave them an advantage over pure code-breaking technicians.
An irony in the situation is that Dyer had no specific Japanese training himself, and he made no effort to get it because languages had been his stumbling block all along when he was in school. Nor was he able to master the art of sending and receiving messages in Morse code, though that was a skill normally expected of communication officers of the period.
In the early 1930s Dyer went back to OP-20G in Washington and worked with Agnes Driscoll, who was a fixture in the department for many years and the institutional memory. She remained there while the naval officers came and went on their various assignments. Dyer found her expertise most helpful.
Another source of help came as electromechanical devices arrived to assist in the code-breaking process. The head of the research section managed to raise $5,000 in scarce Depression-era funds for the first year’s rental on IBM machines that could sort and tabulate punch cards. Normally, the company that made such machines would provide training to the users, but Dyer had to set up the operating procedures himself because he could ill afford to let IBM know the purpose for which the machines were being used.
In 1929, Secretary of State Henry Stimson had famously eschewed code-breaking by saying “Gentlemen don’t read each other’s mail.” Dyer quipped that Stimson’s statement didn’t deter him and his Navy cohorts “because no one could ever accuse us of being gentlemen.”
Valuable Code-Breaker
In 1936, after serving for three years on board the U.S. Fleet flagship Pennsylvania (BB-38), Dyer went ashore to the 14th Naval District in Hawaii. The outfit he headed would in time be known as the combat intelligence unit and the fleet radio unit—innocuous-sounding titles that did not reveal its true function. Dyer did not go to sea again during the remainder of his career. He was too valuable in the code-breaking specialty but hampered for promotion, partly because sea duty was expected and partly because his specialty could not be mentioned in fitness reports. Finally, there was a work-around in having him designated as an engineering duty officer, even though he certainly was not in the mold of naval architects or marine engineers—those who normally became EDOs.
In the summer of 1941, Rochefort reported to the basement unit and took over as officer in charge. Dyer moved down a notch but was not resentful because Rochefort was senior; they had served together amicably before; and because their skills complemented each other. Rochefort’s knowledge of Japanese was a plus for analysis, as was his understanding of the Japanese mindset from having observed it firsthand. As Dyer explained, “The early messages about Coral Sea, the early messages about Midway, were perfectly capable of being read in a totally different way than the correct way. Knowledge of the character of the Japanese and this, that, and the other enabled him to get the right meaning.”
Over time the number of personnel in the unit increased dramatically because of the heavy influx of intercepted messages to be processed. Dyer observed that a number of the officers and enlisted men were eccentric, and he included himself in that category. During much of the war, he had behind his desk a sign that contained a cartoon of a weird-looking character and the caption, “You don’t have to be crazy to work here, but it helps a hell of a lot.” He opined that “all really successful cryptanalysts are a little bit odd, to say the least.”
In part, his abilities were a natural gift, just as a fine athlete has innate abilities that enable him or her to excel in sports. But Dyer also had a large measure of persistence and dedication that pushed him to make the most of his natural talents. He was a stubborn man, which didn’t always endear him to his co-workers, but that also was a valuable attribute when wrestling with a code that did not yield its secrets readily.
Long Days Pay Off
The wartime days and nights were physically and mentally demanding, particularly in the pivotal year 1942. Rochefort and Dyer alternated standing 24-hour duty watches that were often extended by overlap. Dyer frequently stayed awake for more than 24 hours straight, arming himself with a stimulant called Benzedrine to maintain his alertness. He explained, “I figured there were people out there getting shot at. If it should happen that it turned out to inflict some injury on my health in the long run, so what?” The longest period he worked continuously was 42 hours.
Many, many times Dyer’s efforts resulted in frustration, because he was attempting to unscramble messages that the Japanese had deliberately gone to elaborate lengths to make unreadable. But what was the result when he was figuratively able to open a hole in a seemingly impenetrable brick wall? Dyer replied that the exhilaration was “almost in sexual terms. Physiologically it’s not the same, but the emotional feeling is pretty much the same.”
And, asked whether he agreed with an assessment of him as one of the top few code-breakers in the U.S. Navy at the time, Dyer said, “I think without any false modesty, that in cryptanalysis only, I was number one.” No false modesty there.
The victory at Midway was the most striking example of the success of his efforts.