GHQ  Historic Corregidor  | Harbor Defense of Manila & Subic Bays  |  Corregidor Under Siege  Retaking Corregidor  |  Rediscovering Corregidor  | Units & Personnel  |  Map Room Secret Corregidor PX  |  Now Showing |  Archives  |  Bulletin Board | Galleries  |  Mail Call | Links 503d Heritage Bn.










Well done, you're with us still! This part shall concentrate on  JN-25, which was the code system used by the Japanese to transmit the final instructions for the attack on Pearl Harbor.  The issue for your consideration is "what was the state of knowledge of this code amongst the various Allies prior to 7 December 1941?"

When you finish reading this, you might change your mind about the history of WW2. 

But don't jump too quick, for we'll then re-examine these revisionist views, and come around full circle. 

The story so far:...)





Paul F. Whitman




Even prior to the opening of hostilities, the Corregidor unit had, together with the Singapore unit, commenced the attack and breakdown of JN25. This most widely distributed and extensively used of Japan's cryptosystems, in which about half of her naval messages were transmitted, comprised a code with five digit code numbers to which were added a key of other numbers to complicate the system. The Navy called it the "five numeral system, " or more formally, JN25b - the JN for "Japanese Navy," the 25 an identifying number, the b for the second (and current) edition. It had made the difficult initial entries, and was in the best position to make new assumptions or confirm or disprove old ones, intercepting messages that the others might not have picked up.

The phrase "even prior to the opening of hostilities" is a curious one. How much prior? And did Corregidor (the US) necessarily know what Singapore (the British) knew? And when they knew it? Who knew about the Japanese Fleet Codes first, and did they later profess not to have known it?

On 23 April 1943, Adolph A. Berle Jr. of the State Department wrote a secret letter to John G. Winant, the American ambassador in London, concerning a planned visit by code breaking expert Colonel Alfred McCormack to GCCS (Government Code & Cipher School), warning: "A feeling has grown up in certain circles that while there is full interchange on our side, certain information has not been forthcoming from the British side." 

Berle was alluding to the manner in which the US had been open concerning it's code-breaking successes, going so far in January 1941 as to supply the British with two Purple machines, along with a treasure trove of other code breaking secrets.  Included in the handover were two sets of JN-25 fleet codes with current keys (additive tables) and techniques of solution produced by OP-20-G since the initial breakthrough in October 1940.   In return, the British Foreign Office had, during 1941,   vetoed the handing over of  a captured Enigma cryptograph, as had been promised, because it was against British policy to share its code-breaking secrets with a neutral. The Americans had been justifiably furious at this blatant double-cross, particularly the US Navy which had earmarked those very machines to go to Station Hypo at Pearl Harbor.  They had also earmarked the JN-25 codebooks to Hypo and Station Cast (Cavite).  In so doing, the British had not only pretended to be unaware of the Japanese Fleet Code, they feigned knowledge that the US had solved it, and that they had no use for it in London.

The fact (was) that after January 1941 Churchill was getting from GCCS and FECB [Far East Combined Bureau - Colombo & Singapore] decrypts of Purple messages between Tokyo and Washington and Berlin as fast as, and probably faster than, Roosevelt.



So? The British were reading Japanese operational messages before the US entered the war.  Bearing in mind that GCCS had 300 people working solely on JN-25, whereas OP-20-G had to split its far smaller staff between the monthly roster of handling Purple traffic and the navy signals, it is not surprising that the United States took a year longer than the British to break JN-25.  There are JN-25 decrypts in Australian archives dated January-February 1942 which were read immediately and were all complete and thus show that even five weeks into the Pacific War, the British had easily mastered JN-25. But how long had the U.S. been working on the Japanese Fleet Codes?

What is particularly significant (about the gift in January 1941) is that if the U.S. Navy was able to give GCCS reconstructed JN-25 codebooks, even though incomplete, in January 1941, with current additive tables, and to show them how to continue to break the system, then OP-20-G had made remarkably quick progress breaking the code from their first decrypt only three months earlier.  Furthermore, OP-20-G had obviously been reading some JN-25 intercepts during the previous months, and so one would expect to find these today in the archives along with the Purple diplomatic decrypts for the same period. But they are not there.

In April-May 1941, the Americans severely restricted the distribution of their Purple decrypts as the result of their decoding a message from Tokyo to Washington on 5 May 1941 warning: "According to a fairly reliable source of information it appears almost certain the United States government is reading your code messages."  As a direct consequence, Admiral Husband E. Kimmel (C-in-C Pacific Fleet) and Lieutenant General Walker C. Short (Commanding General U.S. Army, in Hawaii) were removed from the distribution list of Magic decrypts. Extraordinary as it may seem, even President Roosevelt was removed from the Magic List, and thereafter he was given suitably paraphrased summaries by the State Department.   Roosevelt didn't get back on the list until 12 November 1941. But this is digression. When did the U.S. begin to decrypt JN-25? Why, in 1979, when President Carter authorized the NSA to release a mass of Japanese intercepts, was not a single JN-25 decrypt that had been read prior to 7 December 1941 released?  Because OP-20-G were preoccupied with the Japanese diplomatic codes and not the Japanese Navy's operational codes, and  none of the latter were ever decrypted?  That's the official view.  The alternative answer comes from  "Betrayal at Pearl Harbor" by James Rusbridger and Eric Nave, published by Summit Books 1991, who quote from A Brief History of Communications Intelligence in the United States, written by Safford in 1952 and released into their National Archives in March 1982.

On 1 June 1939 the Japanese Navy introduced a new type of numerical code referred to by Navy COMINT personnel as [censored] the Operations Code. [The next two lines are totally censored.] Mrs. Driscoll and Mr. Cutter spearheaded the attack and we were soon [censored] reconstructing the code.

Recovery of the [censored] keys, [the word  missing here is probably additive] however, involved much more labour and required many more crypto-personnel than the earlier transposition keys. Main work of solution was undertaken at Washington [OP-20-G].

By December 1940 we were working on two systems of keys with this book; the "old" keys for code recovery and the "new" keys for current information [five lines completely censored].

The inference from the need for more personnel appears to be that the code was more tedious than hard to crack. The Safford report continues:

On 1 December 1941 the system [JN-25] became unreadable...this could have been a tip-off as to coming hostilities but it could have also been a mere routine change of system. After all, the code had been in use for 2 years. Two weeks later Corregidor [Station Cast] flashed the good news that the same old code was still in use but that new keys were being used with its was the third or fourth set of keys used with this same codebook.

There's a caveat here.   We will later hear from Duane Whitlock, who from November 1940 through March 16, 1942 was a radioman first class doing decryption and preparing intelligence reports based on Japanese traffic analysis for the U.S. Navy at Cavite and Corregidor.    Rusbridger and Nave then comment...



This passage is particularly interesting for several reasons. First, Safford is mistaken that the date when the key changes were made was 1 December, when in fact the change was made  on 4 December. second, he confirms that JN-25 was broken soon after its introduction and was read throughout the two-and-a-half-year period to late 1941. And third, that the basic code remained unchanged and only the additive tables (or keys) altered.

They then cite a still censored message from Station Cast (Corregidor) on 15 December 1941 which reads:

"Com 16 to OPNAV info CINCAF. TOP SECRET - 151250. Two intercepts in [censored] plain code [December] 6 and 13 followed within a few hours by enciphered versions confirmed indicator [censored] already recovered by mathematical elimination code remains unchanged (.) Will send recoveries this system if you desire work on current period."

If Station Cast's code breakers knew that JN-25 remained unchanged then it must mean that they were reading it during the previous six month period, from 1 June 1941 through 4 December 1941.

In a memorandum Safford wrote on 17 May 1945 he stated:

Com 16 [Station Cast in Corregidor] intercepts were considered most reliable ..not only because of better radio interception, but because Com 16 was currently reading messages in the Japanese Fleet Cryptographic System (5-number code or JN-25 and was exchanging technical information and translations with the British at Singapore [FECB].

Rusbridger and Nave then make the conclusion,  (against opposition from other writers):



Taken together, these... accounts of U.S. Navy Code breaking show beyond doubt that between 1 June 1939 and 7 December 1941 some JN-25 messages were definitely decoded by the U.S. Navy.  But not a single pre-Pearl Harbor JN-25 intercept or decrypt can be found in any American archive. Every single scrap of evidence relating to JN-25 between June 1939 through late November 1941 has vanished from US records.

What is allowed to appear for posterity in the National Archives, Rusbridger and Nave say,  are a few pre-Pearl Harbor JN-25 decrypts which were said to have been decoded in late 1945 and 1946.  By early 1942, it is openly acknowledged in the public record that JN-25 was being read quite freely. It is impossible to believe that this could have happened other than by a deliberate policy, beginning in the immediate aftermath of the war to conceal or destroy all the evidence relating to the reading, before hostilities, of this code. The U.S. Navy refused to allow discussion of JN-25 at any of the seven inquiries into the attack on Pearl Harbor.   This is not because of some governmental embarrassment at admitting that it had read another government's secrets during peacetime, as the United States has admitted reading pre-war  Japanese Diplomatic Codes (Purple).


The answer is very simply that the JN-25 messages contained the final operational details of the Pearl Harbor attack, whereas the Purple intercepts did not. It is therefore a legitimate conclusion that:

(a) Churchill deliberately withheld from the Americans vital intelligence about the Japanese Task Force derived from reading the Japanese naval code JN-25;

(b) that Churchill rightly believed that if he told Roosevelt what he knew about the Japanese Task Force that FDR would - as a totally honorable President - immediately warn his commanders.  As a probable consequence, the British would then have had to face the Japanese invasion in Malaya alone;

(c) OP-20-G's first-hand knowledge derived from JN-25  could only have been denied the President by a very senior naval officer;

(d)  persons within the U.S. Navy, knowing that the missing material contained information that is highly sensitive and embarrassing, acted in concert to conceal the existence of pre-Pearl Harbor JN-25 decrypts from all  inquiries, and from history itself - just as had happened with the missing Winds message.

(e) even today fifty years later, persons within the U.S. Armed Forces and the NSA,   still censor any original historic  material relating to the subject.

This is not some casual cover-up but a carefully pre-meditated policy of deceit of the greatest magnitude that can only have originated from the highest authority to deliberately frustrate the truth being told.

Rusbridger and Nave then state:

They then go on to suggest that Roosevelt was  kept in the dark over JN-25 matters, (he wasn't told about Purple until 4 months after it was broken, and was denied access to it from May to November 1941 because of an alleged leak) and that the most likely candidate behind this massive cover-up  was .................


(To find that answer, go buy the book!)   If you're still with us, the next link will take you to the most recent book on the subject of JN-25, Marching Orders by Bruce Lee (which, incidentally, is a recommended read.)  His opening salvo on Rusbridger &  Nave is one of the best in the business...


"Now it's harder to put a stop to a headline-making, money-machine conspiracy theory than it is to kill a rattlesnake with a short-handled hoe. But this writer has done so."  

1999 Paul Whitman










Corregidor - Then & Now GHQ Coast Artillery - Harbor Defenses | 503d  RHQ |  503d PRCT Heritage Bn. | Rock Force | Board

H Version 01.27.13

Capt. Duane Whitlock U.S.N. (Ret'd) has an article on cryptography and the role that Corregidor played in it entitled The Silent War Against the Japanese. He was a radioman on Corregidor.