It was boring, tedious work, except when it wasn’t. Elizabeth Bigelow, an aspiring architect recruited from Vassar, also began working on JN-25 when she came in with a later university-trained cohort. She at one point was given an urgent but badly garbled cipher and asked to decipher it, which she did, within a matter of hours. It told of a convoy sailing later that day. When she was told that her work had helped to sink the convoy, she said later, “I felt terribly pleased.”
The operation developed the swiftness and efficiency of an assembly line. To the extent that space provided, the women were assigned to groups, or “rooms.” The write-up room would prepare work sheets. The “key” room placed the messages; the classification room salvaged garbled intercepts; a priority room with “expert additive workers” attacked “hot or priority messages.” Hotlines were set up to convey additives to translators. Some messages were tagged “routine,” others “urgent.” There was a special category marked “frantic.” As the number of messages increased, the ranks of people solving them grew steadily more female. By the fourth quarter of 1943, 183 men and 473 women were working on JN-25 in Washington—more than twice as many women as men. One memo noted that it was impossible to keep the women in the dark as to what the messages said. The memo added that the most important secret was the fact that JN-25 was being worked at all, and this secret was at the “mercy of the humblest worker who ever glanced at a work book.”
The women kept that secret and became integral to the operation. And they shared the outrage when the truth about Midway’s success made its way into the press. On June 7, 1942—while the battle was still going on—the Chicago Tribune published a blockbuster story in its Sunday edition, headed: JAP FLEET SMASHED BY U.S. 2 CARRIERS SUNK AT MIDWAY: NAVY HAD WORD OF JAP PLAN TO STRIKE AT SEA; KNEW DUTCH HARBOR WAS A FEINT. The article noted that the makeup of the Japanese forces “was well known in American naval circles several days before the battle began.” The article appeared in several other Tribune -connected papers but was hushed up by the office of censorship out of fear that the Japanese would take note.
Then the syndicated gossip columnist Walter Winchell compounded the risk, saying in a July 5 radio broadcast that “twice the fate of the civilized world was changed by intercepted and decoded messages”—meaning Coral Sea and Midway. Two days later he wrote in his “On Broadway” column, which ran in the New York Daily Mirror, that Washington was abuzz over the Tribune’s item, which, he argued, “tossed safety out the windows—and allegedly printed the lowdown on why we won at Midway.” Of course, he had done the same thing.
The Navy was so apoplectic that it ended up making things worse. When it emerged that a reporter aboard the USS Lexington, Stanley Johnston, was the source of the Midway story, the Navy decided to go after Johnston. The hearings resulted in more publicity, and the code breakers worried that there was no way the Japanese could ignore it. The Japanese made another major overhaul of JN-25 not long after, and many code breakers were convinced this new changeover was the result of the Johnston investigation. “Our crucial battles in the Solomons were conducted without the aid of enemy information that had been available up to this moment,” wrote Laurance Safford bitterly.
Whether the Japanese did take notice is a matter of dispute. The Japanese periodically changed JN-25 books anyway, and their mid-August change most likely had been planned for some time. Regardless, it occurred just as the U.S. Navy began to press its new advantage in the Pacific, taking the offensive and launching an invasion to retake the Solomon Islands. This brave push began with the Battle of Guadalcanal, an amphibious invasion that went well, at first, then bogged down into a bloody months-long quagmire. During the battle, U.S. Marines came upon a cache of codebooks buried six feet in the ground; gallingly, they were in a version of JN-25 that was no longer in use.
Fortunately—given how often JN-25 changed over—the U.S. Navy had made the wise decision to set up a smaller unit to tackle what were called “minor ciphers.” The minor ciphers were lesser systems, but they were anything but unimportant. In the vast Pacific Ocean, not every message could travel in the main fleet code. The Japanese used scores of auxiliary systems, some brief and temporary, to communicate between captured islands, or between weather lookouts and rice ships, or even just to broadcast water levels and fishing conditions. They also devised temporary “contact codes” for use in battle. The ability to read some of these systems had been useful during Midway; the Japanese changed their JN-25 cipher on May 28, just before the battle, so all through that actual engagement, the Americans were able to follow the conversations of Japanese combatants only thanks to contact codes and other minor systems.
As luck would have it, the minor-cipher unit was under the charge of Frank Raven, the very code breaker who destroyed what remained of the career of Agnes Driscoll. His “German Navy miscellaneous” team was now working “Japanese miscellaneous,” plucking intercepts out of random piles in the Navy building that were accumulating in a junk box marked “W.” The crew was a good one—according to Raven, they broke at least one system per week beginning in March 1942—and now the men on that crew were replaced by women. In May 1942 Raven had twenty-three male sailors working under him, and by June, just one month later, “approximately ten civil service girls” came on to form the nucleus of the new team.