Replies

"AN-1 and its group additives were an administrative code like Purple only it was for the IJN. On August 1st the Japanese added 50,000 new additive to the code book that made it unreadable. By Novemeber of 1941 only 3,800 code groups had been identified which still meant the code was unreadable."

I've searched through Stinnett's book, and cannot find any mention of an "AN-1 Code", amongst either the diplomatic codes -- i.e., purple and J-series -- or the Kagun Ango Naval Codes.

Nor does he mention anything about 50,000 "new additives" of which only 3,800 broken by November 1941.

In Stinnett's version, all the codes were easily read throughout the period.
For Stinnett the real question is: who received the intelligence reports?
His answer is: President Roosevelt and his highest level military staff certainly did, but Admiral Kimmel and General Short did not.

(George Victor in "The Pearl Harbor Myth" also makes no mention of an "AN-1 Code").

What Stinnett does discuss in great detail are the Kaigun Ango Japanese Naval Codes of which there were four primary systems (Pgs 71-72):

Code Book D, known to American cryptographers as the 5-Num code.
A radio call sign code assigned to Japanese officials, shore stations, warships, and Marus.
The Ship Movement code, a system for reporting arrival and departures of naval vessels from Japanese ports and anchorages, known to Americans as the SM code, and
The Shin code, used for contacting vessels of the merchant marine, known as S code in America.

Finally, this from Wikipedia (no mention of AN-1):

"By late 1941, those organizations had broken several Japanese ciphers, such as J19 and PA-K2, called Tsu and Oite respectively by the Japanese.[7]
The highest security diplomatic code, dubbed PURPLE by the U.S., had been broken, but American cryptanalysts had made little progress against the IJN's current Naval Code D (called JN-25 by the U.S. after March 1942)..."

"...However, concerning the manning levels, "... just prior to World War II, [the US] had some 700 people engaged in the effort and [was], in fact, obviously having some successes."[8]
Of these, 85% were tasked to decryption and 50% to translation efforts against IJN codes.[9]
The nature and degree of these successes has led to great confusion among non-specialists."

So, we're talking about several hundred code-breakers working on just the Japanese Naval Codes.
And we are to believe that all these hundreds of people produced nothing prior to December 7, 1941?

I don't believe it.
Stinnett says our guys never had a serious problem reading the Japanese 5-Num Naval Code D (aka JN-25).

Here we go. There's the connection. The Kaigun Ango - Sho D was also called AN-1. I also found that it was later referred to as JN25B. The original article that I read on this code, I cant link but I did at least find it's title online. This took a while since I read it at UNM and couldn't for the life of me remember the title. Your mention of Kaigun Ango helped me re-find it.

The original article is called "Closing The Book on Pearl Harbor" by Stephen Budiansky and is in the April, 2000 issue of Crytologia. Stephen Budiansky also wrote "Battle of Wits: The Complete Story of Codebreaking in World War II" which though I have not read it, I did look at portions of it on Google Books when I was researching the "Big Leak" paper I did on the leak of the Rainbow 5 Plan to the Chicago Tribune on December 4th, 1941. The article was actually one of my primary sources on the background I put forth in that essay. (Pretty good essay I thought, though I admit that I could not identify who the source of the leak was.)

So Stinnett does mention the AN-1 after all, just under a different name, which is what I suspected was going on after we went back and forth on this a few times. Also he mentions in his notes that the 5-Numeral or AN-1 to match up the terminology was translated in 1945 and 1946 (p. 334). Too late for Pearl. The only way to verify this would be go the the National Archives and look at the RG 457 documentation. The Japanese decrypts in question are in boxes 286-516 so talk about something that would take a while. :)

What is interesting is that Stinnett says that all these codes were were easily read during this period then points out in his notes that the AN-1 was not translated until 1945. Strange oversight. But again, I'm not criticizing his theory, I just find it interesting that this unbroken code is not more predominant in his research. That doesn't necessarily mean anything.

As to Wiki not mentioning AN-1, I again urge you not to use Wikipedia as a primary source. Just because it's not in it, doesn't mean it does not exist. I had to do some digging to make this connection and I never once looked at Wikipedia to do it.

I found this article in my searching which ties the terminology together and is a pretty well researched essay so far. I'm about half way through it at the moment and will get back to it tonight after my son's graduation party.

Decoding Pearl Harbor: USN Cryptanalysis and the Challenge of JN-25B in 1941