More important now than ever would be Rochefort’s ability to establish credibility — particularly with the new Commander in Chief, U.S. Fleet (COMINCH), Admiral Ernest J. King, and the new CINC PACFLT, Admiral Chester W. Nimitz.

Both could understandably be skeptical of HYPO’s abilities and estimates as they had failed to warn of the attack on Pearl Harbor.  Rochefort’s task would be to change their minds.  Though it would take time, one small success after another began to build the trust and confidence that would carry the Pacific Fleet to eventual victory over the Japanese.

Breaking JN-25(b)

In the days and weeks immediately following the attack on Pearl Harbor the Imperial Japanese Navy began to “run wild” all over the Pacific just as Yamamoto had foretold.  Elliot Carlson writes:  “Japanese forces days [after Pearl Harbor] had seized Guam (eliminating the Navy’s intercept station there), bombed Wake and the Philippines, and attacked Malaya, Shanghai, Thailand, and British Borneo.  Japan’s army and navy were on the move everywhere.”  Further advances must be stopped.  Rochefort knew that simply identifying and locating the enemy would not be enough if the U.S. Pacific Fleet hoped to get ahead of Yamamoto.  He would need to predictively track their movements — informing the operational commander today of the enemy’s intentions tomorrow.


Intentions could not be determined by rudimentary traffic analysis alone.  Station HYPO’s cryptanalysts would need to be able to read and analyze the message contents of the IJN’s main operational code, JN-25(b).  On December 10, 1941, OP-20-G gave Rochefort and his team that very opportunity.  Though the opportunity was welcomed, the challenge would be both harrowing and time consuming.  Craig L. Symonds, in his book, The Battle of Midway, describes the code and its intricacies well:

“The JN-25(b) code consisted of [more than 50,000] five-digit number groups, such that the messages that went out over the air waves looked something like this:

48933        19947        62145        02943        20382        16380

Some of the number groups were dummies, or fillers, added to confuse the code breakers.  In addition to that, however, before sending a message, the Japanese enciphered the code again using a cipher tablet.  The encoder selected a five-digit number from this tablet and added it to the first number group in the message; the next cipher number was added to the second number group, and so on throughout the message.  An indicator buried in the message itself revealed the exact location — page number, column, and line — where the cipher number additives could be found in the secondary tablet.  Thus the code group for “east” might be 10236, but it would be encrypted again by adding another five-digit number from the cipher tablet.  If the coder added the number 45038, the word “east” became 55264… To decrypt the message, the recipient needed the initial code book, the secondary code tablet, and the indicator, showing how to subtract the second from the first.  The puzzle, in short, was extraordinarily complicated…”


The task of cracking the code would be harrowing indeed.  The Japanese believed it to be unbreakable.  Rochefort and his cryptanalysts were determined to prove them wrong.  Working around the clock, Dyer, Wright, and Holtwick analyzed message after message — pouring over them all hours of the day and night — staring at code groups until patterns and clues started to appear.  With some assistance from Station NEGAT and the former inhabitants of Station CAST, slowly but surely they began to fill the code book with thousands of groups.

This effort did not produce results overnight.  In fact, it would be months before they could make sense of many of the messages in a timely manner.  Nevertheless, successes were achieved.  Though many intercepts were ever only partially decrypted, Rochefort was occasionally able to piece together the fragmented code group recoveries in a message and provide operationally relevant intelligence to Nimitz — intelligence that would enable the Commander in Chief to employ his forces in both defensive and offensive positions.

Koryaku Butai

One such key break-through occurred as early as January 18, 1942 — a mere month after Station HYPO had been given responsibility for breaking the operational code.  Dyer and Wright had revealed enough of the code groups in three intercepts to allow for partial translation.  Common to all three was one particular code group which Rochefort translated as “koryaku butai,” or, “assault landing force.”   Preceding the code group was the letter “R.”  Rochefort believed, and Nimitz’ intelligence officer agreed, that the “R’ was Rabaul.  Based on his knowledge of the current disposition of Japan’s forces — courtesy of his traffic analysts and ship plotters — Rochefort deduced that Nagumo was going to head south from Truk, attack New Britain and occupy Rabaul.

Unfortunately, Nimitz did not have forces in a position to repel such an attack.  Rochefort’s analysis was, nevertheless, important to Nimitz.  The Pacific Fleet commander could now carry out America’s first offensive strike in the war against the Gilbert and Marshall islands with relative assurance that Nagumo’s forces could not counter his operations.


Rochefort was right.  On January 20, 1942, aircraft from Nagumo’s carriers attacked the Bismarck Archipelago and occupied Rabaul shortly thereafter.  A few days later, Admirals Halsey and Fletcher would conduct strikes against Japanese positions in the Gilbert and Marshall islands virtually unopposed.  Though the strikes did little physical damage, they met with COMINCH King’s intent to strike back and, in turn, boosted the morale of the Pacific Fleet.  Rochefort had captured the attention of his new Commander in Chief.  Elliot Carlson writes:  “Rochefort’s acumen forecasting the Rabaul attack made an impression on Nimitz… In following weeks Rochefort wouldn’t always bat a thousand, but in this instance he had given Nimitz a hint of things to come.”  Rochefort had laid the foundation for a relationship based on mutual trust and confidence with his operational commander that would pay huge dividends in the months to come.

Center of Codebreaking Activity

Commander Rochefort built that foundation not on flattery or pomp, but on hard work and tangible results.  In many ways, he and his team were aided by the Japanese themselves.  With the increase in activity throughout the Pacific Ocean came an increase in communication.  Yamamoto’s operations had to be coordinated and JN-25(b) was the means by which he did so.  The result for Station HYPO was a flood of intercepts from which to work their cryptanalysis.

In addition to the intelligence provided regarding Japan’s attack on New Britain and the subsequent occupation of Rabaul, HYPO had also broken intercepts indicating a second attack on Pearl Harbor.  Radio traffic emanating from IJN forces in the vicinity of the Marshall Islands provided clues to an offensive operation to occur on March 4 against “AK.”  AK was one of many Kana digraphs in the Japanese designator system used to reference geographic locations in the Pacific.  Rochefort believed AK to be Hawaii.  This belief was not a random guess, but rather a product of weeks and months of studying and analyzing the traffic combining all of the tools at his disposal.  Based on the lack of concrete evidence, some, particularly those at OP-20-G in Washington, doubted Rochefort’s assessment.  This would not be the last time that Rochefort would deduce such a critical correlation, nor would it be the last time that others doubted his analysis.  Nimitz, however, trusted his lead codebreaker and alerted his forces.


The end of the story is somewhat anti-climactic.  Rochefort was right about both the date and the location.  What he was never able to determine was the composition of the force.  As it turned out, the force was comprised of two small sea-planes carrying four bombs each.  Arriving over Pearl Harbor on 4 March, one seaplane over shot Honolulu and dropped it’s bombs in the Ko’olau mountain range near Manoa and the other seaplane missed entirely causing little more than splashes at the entrance of the harbor.

By the end of March, the dungeon had become the center of codebreaking activity for all of the Pacific Fleet.  Traffic analysts were providing current IJN force disposition, location, and tracking.  The cryptanalysts were starting to break enough code groups to allow the linguists to gain insight into the details of future operations.  And the ship plotters were maintaining the overall picture of both blue and red force activity.  Fusing all of the efforts together — along with a little bit of educated guesswork — Rochefort was able to track the Imperial Japanese Navy and, in so doing, provide CINC PACFLT with the battlespace awareness and predictive analysis necessary to anticipate Japan’s next move.



[1]  Carlson, Elliot.  Joe Rochefort’s War:  The Odyssey of the Codebreaker Who Outwitted Yamamoto at Midway.  Annapolis, MD:  Naval Institute Press, 2011.  Print.  [pp. 91-113]

[2] Symonds, Craig.  The Battle of Midway.  New York, NY:  Oxford University Press, 2011.  Print.

[3] Toll, Ian.  Pacific Crucible:  War at Sea in the Pacific, 1941-1942.  New York, NY:  W. W. Norton & Company, 2012.  Print.