On a June morning in 1930 William Friedman, Chief of the Army’s Signals Intelligence Service (SIS), and three members of his staff headed down a long corridor in a deserted part of the Munitions Building. Throwing open the two large doors that led into the vault area, Friedman lit a match and announced, “Welcome, gentlemen, to the archives of the American Black Chamber.”Established in 1919 the Chamber had once been America’s main cryptologic workhorse. Friedman hoped to determine if any of the information found in the Chamber’s files could shed light on current Japanese systems.

purple2
William Friedman

By the mid-1930s, Frank Rowlett, the senior member of the group, headed up a robust effort to break the first of two important Japanese diplomatic systems. Dubbed the Red Code, the elements of the system were a mystery. For months, Rowlett and his team endured long days and even more sleepless nights. Finally, a breakthrough occurred one evening. Rowlett remembered that from the hundreds of messages he had examined, three in particular were exceedingly long. If by chance those messages were enciphered on the same particular machine, it might be possible to discern a pattern that resembled certain Japanese words.

purple3
Frank Rowlett

The following morning, the team put Rowlett’s epiphany to the test. By noon they were on their way to solving the Red Code. But the benefits culled from the Red Code would only last so long. In 1939 the Japanese upgraded to a new machine driven cipher that SIS called “Purple.” Despite their earlier success, the team realized that Purple was more complex than the previous Red system. For the next 18 months the codebreakers were in the dark.

Thankfully, for a brief period the Japanese used both the Red and Purple machines on some diplomatic circuits. By monitoring specific stations, the team could predict the first few words of each message. Using the many “cribs,” as the cryptologic clues were called, the team discerned that, in the text of Purple messages, six letters were always treated differently than the other 20. Through pen and paper analysis, the team regularly uncovered the shuffling sequence and determined which of the six letters stood apart from the other 20 each day.

But pen and paper analysis was not fast enough to crack the system on a timely basis. To conquer Purple, the team needed to build a device that could mimic the machine’s scrambling pattern. The fact was, however, that no westerner had ever seen a Purple machine.

The challenge of building the device fell to Leo Rosen, a MIT educated Army officer. One day, while leafing through an electrical supply catalog, Rosen noticed a device called “the uniselector.” The device consisted of six telephone stepping switches bunched together that ran through 25 contacts. Quickly, he ordered two of the devices built, what would come to be known as, the “six buster.”

purple4
 Leo Rosen

Conquering the “sixes” was critical, but there was still the even more difficult challenge of the “twenties.” Luckily, by September 1940, the Purple team knew enough to uncover the general patterns required to crack the remaining 20 letters.

Using the insights culled from the conquering of the sixes and the twenties, Rosen made the additional modifications to his original prototype. Each section of the “Purple analog” connected to the next through 500 wires. Lastly, Rosen added a typewriter to feed the encrypted traffic into the device and an additional typewriter to spit out the deciphered text.

Incredibly, after countless hours of painstaking effort, the work was finally done. It was time to test the system. Historian Stephen Budiansky notes.

“Late one night, Rosen and Rowlett plugged in the power supply and flipped the main switch. Rowlett began to type in the cipher text of a Purple message. The two cryptanalysts watched in awe as deciphered Japanese text began to emerge from the printer.”

The breaking of Purple would be of immense help in understanding Japan’s diplomatic strategy before Pearl Harbor. But as great a success as Purple was, there was a distinct downside. Generals and admirals dwell in far different worlds than those who negotiate treaties. The stunning success of Purple distracted the U.S. cryptologic community from the true indicator of Japanese intentions, the naval code.

Thus on the evening of December 7, 1941, the Pacific Fleet lay in ruins and those like Frank Rowlett, who were responsible for predicting just such an event, faced the future with a mixture of regret for past sins of omission and a willing determination to do better.

Sources:
cryptologicfoundation.org
NSA.gov