Pearl Harbor and JN-25B - questions

[Caustic Logic]

Note to mods: Technically this might fit better in a science forum, or history, or something, but it does tie in with the PH conspiracy angle, so I’m posting it here.

Hey, I'm back after a long break. Thank you, thank you all. I’ve been looking more in-depth into the Pearl Harbor attack (yes, the "FDR knew" angle), and at the moment I’m trying to sort out the issue of the Japanese naval code dubbed JN-25B. The main story of this code for the war is the secret window onto Japanese navy plans that was finally broken in early ’42, giving us a surprise win at Midway and Yamamoto’s head, but the crack came too late to see the Pearl Harbor attack coming. I’d like to offer a mix of observations and questions for some of the more learned here.

I understand that as used, it was a code using tens of thousands of 5-digit sequences to represent Japanese Kana, and further superencyphered with a book of random-number ‘additives’ that was changed at intervals. In essence, it seems, the encryption must be neutralized, at least in part, to even see the code, which then had to be analyzed for patterns to narrow down to the possible meanings of the number strings.

The original JN-25 was instituted 1939, which U.S. cryptanalysts had made some progress on before it was replaced by the JN-25B version, essentially a new code, on Dec 1 1940

Q1) minor question - why was it not called JN-26? Is it a different code altogether, or a variant of the first? Did the change invalidate the previous progress?

As it appears so far, subsequent changes that frustrated cracking efforts were not with the code itself, but with the encrypting additive books. This occurred at intervals – once in early 1941, again on June1, again on August 1, and again on Dec 4, just before the surprise attack. If I understand right, the code-breaking progress until then is still valid and ready to proceed once the cipher is cracked anew.

Q2) Is this correct, or did the additive changes alter the underlying code, invalidating previous work?

Evidence it was not broken by US cryptographers appears pretty strong: Budiansky claims to have found multiple primary sources detailing the actual progress on the code and its ciphers, indicating that no JN-25 transmissions were read as current intelligence prior to Dec 7. This may be an oversimplification (eg; there may have been particles of useful intel, rather than none, so far as I know), but it seems to me essentially correct.

Contradictory evidence is weak; Robert Stinnett – whose work seems to be full of flaws I’m beginning to latch onto - claims the US had broken the code entirely by mid-November 1941 at latest. His main evidence for this is a letter he found that seems to say something else entirely – something closer to “we’re trying hard to break the code.” http://www.nybooks.com/articles/14086

So it appears the US, at least in the known channels of analysis (Hypo, Cast, etc.) did not get more than a smidgen of the code readable – but there were people other than the US working it, especially the British and the Dutch. James Rusbridger and Eric Nave claim the British broke the code and were able to read JN-25 messages at will well before 12/7. The whole thing sounds fishy but I haven’t looked close yet.

Q3) Can anyone debunk the British JN-25 decode prior to 12/7 proposal before I do and save me work?

It may mean nothing, but the 1946 Congressional joint committee's report made no mention of JN-25, or any Japanese coding system. Decodes of intercepts were possible by then, but not called on as evidence – the Japanese plans were instead assembled from memories of their military survivors, and some stray copies that escaped destruction. Adm. Layton found in the 1980s the National Security Archive had “no JN-25 traffic that was actually decrypted before war broke,” “not one single original decrypt made at the time of any JN-25 message” and even looking at later decodes “we evidently did not pick up Yamamoto’s 25 November sailing message” at all. Layton also found British official sources equally non-forthcoming on 1941-era JN-25 decodes. [See “And I was There” available for reading online – pages 206-207, 231-32, 534]

Q4) Am I reading Layton right, that this very important order is not in the archive as received?
Q5) Does anyone have any newer information on where a pre-Dec 7 intercept can be seen so we can see what a “10-15%” readable message looks like?

I have other questions and thoughts for later, but this should generate enough answers to keep me busy for a while.

[JoeyDonuts]

I wasn't a codebreaker, I was an ELINT guy when I was in the Navy...

The Center for Information Dominance down at Corry Station in Pensacola, FL has a Cryptologic Museum that has a lot of stuff from that time period. They even have an Enigma machine cyphered for the Japanese - extremely rare.

Anyway, it might be worth it to file a FOIA when you really get it narrowed down what you're looking for.

Also, IIRC, the book "At Dawn We Slept" has a fairly good section on JN-25 that would be worth a read if you haven't read it yet.

[Caustic Logic]

Originally Posted by JoeyDonuts

Anyway, it might be worth it to file a FOIA when you really get it narrowed down what you're looking for.

Also, IIRC, the book "At Dawn We Slept" has a fairly good section on JN-25 that would be worth a read if you haven't read it yet.

Cool, thanks! I hope it doesn't come to that, I've had a bad history with FOIA stuff. So far I am already outta my depth a bit, but I've only been really researching about a month now. I hope I can follow the info I'm sure someone here knows waaaay too much about.

That book is on my list for the next round - even I can afford used paperbooks on Amazon in plural form! It's cool - right to the index for now - Day of Deceit (Stinnett), Battle of Wits (Budiansky), Pearl Harbor Myth (Victor), and I was There (Layton) so far. Deeper stuff as required...

[Caustic Logic]

May as well slip in this point too.

I'm a little confused by Budiansky’s explanation how after the war, Hypo technicians went back to “unread AN [JN-25] traffic that had piled up in the months just before Pearl Harbor,” and while finding some clues of the attack, “None specifically mentioned Pearl Harbor.” [BOW 8-9]

That seems to me either a disingenuous dodge or else an interesting clue. Multiple orders to the fleet sent in JN-25 mention things like (from Yamamoto’s sailing order, 11/25):“The task force […] shall advance into Hawaiian waters and […] attack the main force of the United States Fleet in Hawaii and deal it a mortal blow.” the same traffic should have shown that the attack was slated for “dawn of X-day,” which was already set as the 8th (Tokyo time) as far back as Nov. 7. http://www.ibiblio.org/pha/myths/jm-097.html#1

Q3) Were these orders not included in the JN-25 traffic Hypo received, or is Budiansky playing word games here? Is there good reason to think Hypo should not have intercepted this?

[UNLoVedRebel]

I bet ULTIMA1 can help you out. He does, after all, work for the NSA.

[UNLoVedRebel]

[OT] I recently bumped your 77's shadow thread to see if CIT's water boy would take a stab. He wouldn't touch it with a 40 foot pole. [OT]

[Caustic Logic]

Originally Posted by UNLoVedRebel

[OT] I recently bumped your 77's shadow thread to see if CIT's water boy would take a stab. He wouldn't touch it with a 40 foot pole. [OT]

Cool mang, and thanx for the welcome backs. I'll have to check that bump.

As for Ultima1, who claims better luck than I with FOIA requests

awww, too bad he seems to have disappeared.

Say hi to Nevada for me.

[UNLoVedRebel]

Originally Posted by Caustic Logic

Say hi to Nevada for me.

You're the first to figure out my SN.

[Caustic Logic]

Originally Posted by UNLoVedRebel

You're the first to figure out my SN.

Somehow I doubt that. But it does tie back in with code-breaking, sort-of.
---
So until someone else has anything, here is a timeline of additive changes from Dec 40-Dec 41 with crypto-analytic progress estimates from various sources:
[Very detailed timeline here, the best I've seen if it's reliable]

Dec 1 1940 – JN-25B introduced to replace the completely decrypted JN-25A
Feb 1 1941 - “In this additive book the Japanese simply rearranged the key additives, rather than changing them, making the key recovery simple.”
June (?) - “Six months [after Dec 1] the Japanese Navy replaced the additive book.” [not listed at above link]
“By August 1 1941, they had recovered 4,800 additives in the current book.”
“prior to […] August 1941, the cryptanalysts had recovered only 2,000 code groups in JN-25 -- about 4% of the codebook -- and these were mostly numerals and stereotyped phrases.”
August 1
“Washington's success at reading JN-25B in November was probably on the order of 10 to 15 percent.”
“only 2,500 additives and 3,800 code groups had been recovered” before Dec 4.
Dec 4 (Dec 1 by some sources) - final additive change before attack.

[Biscuit]

Not sure if this is helpful or not but I know that the U.S. had broken the Japanese diplomatic code prior to PH. In fact the Japanese suspected this and broadcast fake ship positions to make the U.S. think that their fleet was in Japanese waters when they were actually steaming towards PH.

I have not seen anything that indicates we had broken their military codes.

If you haven't seen it before you might find this site helpful. I know I have.

http://www.nsa.gov/about/cryptologic...ew/index.shtml

[JoeyDonuts]

IIRC, The reason why not much progress had been made by Station HYPO is because JN-25 had been used very sparingly, if it all, prior to Pearl Harbor. They used secure courier as much as they could get away with it. After commencement of hostilities, JN25 activity surged, giving the cryptanalysts at HYPO and points distant much more to work with.

Here are a few of MY interesting points about the whole shebang:

1. The Japanese spy Takeo Yoshikawa made his reports using PURPLE diplomatic code.
2. The pissing contest between HYPO's CO LCDR Rochefort and the OP-20-G office.
3. The railroading of LCDR Rochefort post-Midway, not to mention that flatout careermongering glory theft of the Redman brothers.
4. Relationship between FLTADM Nimitz and LCDR Rochefort.

This would be a fantastic movie in the right hands, you know. The real in depth dramatic tales of the strong personalities involved.

You know I hate to say it, but the mentality of the Washington Naval Intel bureau actually reminds me of what you saw pre-9/11. Combination of head-in-the-sand, shoot the messenger, careermongering, and ass-covering.

[JoeyDonuts]

Originally Posted by Biscuit

Not sure if this is helpful or not but I know that the U.S. had broken the Japanese diplomatic code prior to PH.

The Japanese spy on the islands was transmitting using the broken PURPLE code. He also didn't know that Pearl was for sure a target, but he had a strong inclination. Anyhow, one intercept out of all of them stands out - he had recommended that the harbor be delineated in a grid format in preparation for bombing and shallow-water torp runs.

The problem is, the sheer volume of mundane run-of-the mill traffic being transmitted in the Japanese diplomatic PURPLE code made it very easy for this crucial little tidbit to get lost in the tide.

I believe that the IJN, who essentially was running the government in late 1941, did not trust the embassies with critical information like attack plans. This sort of thing would have never gone out over PURPLE code. It would have been via secure courier or one of the few-and-far-between JN-25 transmissions pre-December 7th.

[Caustic Logic]

Originally Posted by Biscuit

I have not seen anything that indicates we had broken their military codes.

Me neither, at least the main one I'm looking into.

Quote:

If you haven't seen it before you might find this site helpful. I know I have.

http://www.nsa.gov/about/cryptologic...ew/index.shtml

Yes, the JN-25 article is one of my sources, pretty good, and the main Pearl Harbor one has some good points as well.

Originally Posted by JoeyDonuts

IIRC, The reason why not much progress had been made by Station HYPO is because JN-25 had been used very sparingly, if it all, prior to Pearl Harbor. They used secure courier as much as they could get away with it. After commencement of hostilities, JN25 activity surged, giving the cryptanalysts at HYPO and points distant much more to work with.

True, it came out big once ships started moving around more - but this happened at least somewhat before 12/7. They had to get their orders and such, and as one source points out:
"By the end of 1942, the Japanese Navy employed fourteen different minor systems which generated over 40,000 messages per year in addition to messages obtained from the general-purpose system which, by November 1941, had reached 7,000 messages per month." [general purpose here does mean JN-25B]
http://www.history.navy.mil/books/comint/ComInt-2.html
and another:
"In 1941 Washington was receiving on average 7,000 JN-25 original messages plus 11,000 duplicates per month. This average level continued until mid 1942, [ref 1, page 7] These made up approximately 60 to 75% of the IJN messages (depending on the month) received in Washington."
http://home.comcast.net/~r2russ/midw...5-sinclair.htm

So I'm not sure if app. 7,000/mo is considered heavy or light, but it could give some kind of depth to analyze and plumb. I'm of the semi-informed opinion the code could well have been broken in time and probably should have been. The main problem, according to the best-informed sources, was not a lack of transmissions, but of manpower and priority. HYPO for example was kept primarily busy on the relatively low-traffic Flag Officer's Code until after the attack - when they get serious, they go for JN-25. Before 12/7, they were not so serious for some reason.

Thanks for the thoughts. Next post I have some more responses.

[Caustic Logic]

Originally Posted by JoeyDonuts

Here are a few of MY interesting points about the whole shebang:

1. The Japanese spy Takeo Yoshikawa made his reports using PURPLE diplomatic code.
2. The pissing contest between HYPO's CO LCDR Rochefort and the OP-20-G office.
3. The railroading of LCDR Rochefort post-Midway, not to mention that flatout careermongering glory theft of the Redman brothers.
4. Relationship between FLTADM Nimitz and LCDR Rochefort.

This would be a fantastic movie in the right hands, you know. The real in depth dramatic tales of the strong personalities involved.

Yeah, like a nice [ST]feel-good[/ST] feel-not-as-bad fictional film - Bruckheimer, not Burns. Seriously, maybe I'm wrong to think this, but it seems like personality foibles/ego/drama/random mental problems ('sabotage psychosis' is my fave) seem to play an inordinately large role here. It's crazy to consider FDR's and others' desire to enter the war as a possible reason, but a whole string of random derelictions and confusion and technical problems is okay. Coincidence theorists are writing the predominant narrative! Just so long as nobody on our side meant any harm, it flies...

Quote:

You know I hate to say it, but the mentality of the Washington Naval Intel bureau actually reminds me of what you saw pre-9/11. Combination of head-in-the-sand, shoot the messenger, careermongering, and ass-covering.

I'm less keen now than I would be a year ago to say 'hey, you said it not me,' but... the parallels are a bit unnerving.

Quote:

The Japanese spy on the islands was transmitting using the broken PURPLE code. He also didn't know that Pearl was for sure a target, but he had a strong inclination. Anyhow, one intercept out of all of them stands out - he had recommended that the harbor be delineated in a grid format in preparation for bombing and shallow-water torp runs.

Interesting. I knew that one was known, but didn't know it was in Purple. I hear that no spies were charged, maybe that's wrong.

Quote:

The problem is, the sheer volume of mundane run-of-the mill traffic being transmitted in the Japanese diplomatic PURPLE code made it very easy for this crucial little tidbit to get lost in the tide.

I believe that the IJN, who essentially was running the government in late 1941, did not trust the embassies with critical information like attack plans. This sort of thing would have never gone out over PURPLE code. It would have been via secure courier or one of the few-and-far-between JN-25 transmissions pre-December 7th.

Yes, a lack of trust is app. why the diplomats were left out of the loop except in the most general sense. War footing for Navy means ships A, B, C go here or there at such-and-such time, and blow up this or that. War footing for diplomats means cut-off negotiations and smash your machines at such-and-such time. Way more vague. It was a bad priority, IF actionable defensive intel was desired, to focus on Purple and such rather than Naval codes.

No harm meant, of course, just another cock-up pile-up of history.

Thanks for the springboards, don't take me as attacking you personally, JD.

[JoeyDonuts]

Originally Posted by Caustic Logic

Thanks for the springboards, don't take me as attacking you personally, JD.

What? No. You'll have to try harder than that.

WRT the spy, I believe he received a three-word coded message that tipped him off that attack was imminent. He destroyed all of his codebooks and basically anything that could ID him as a spy.

From the wiki-wiki:

Quote:

When he heard the "East wind, rain." code phrase on the short wave radio bringing the news from Tokyo signaling an attack against America, Yoshikawa destroyed all evidence of his activities. When the FBI picked him up on the day of the attack there was no incriminating evidence of his espionage. Yoshikawa eventually returned to Japan in August 1942 in a diplomat prisoner exchange. It was not known for some time that he was the chief Japanese agent in Hawaii.

[Caustic Logic]

Originally Posted by JoeyDonuts

WRT the spy, I believe he received a three-word coded message that tipped him off that attack was imminent. He destroyed all of his codebooks and basically anything that could ID him as a spy.

From the wiki-wiki:

Hmm, I'd consider that a realistic enuff chain of events.

Holy crap, why am I still awake?
zzzzz

[Caustic Logic]

Last post was not sarcastic. That does sound reasonable.

Just got my copy of Layton's book finally. Will be back with info.

[Caustic Logic]

Alright, I have read a good chunk of this mammoth 550 page tome
Layton, Edwin T. with Roger Pineau and John Costello "And I Was There": Pearl Harbor And Midway -- Breaking the Secrets.New York. Quill. 1985.

Layton was the Pacific Fleet's main intel officer, of course, responsible foor keeping Kimmell informed of Japanese plans, intent, etc. If there were a plan to keep intel from Pearl, Layton would have to be in on it, or right up against it but outside. The book is full of interesting clues indicating the latter, although he never draws that conclusion himself, speaking of 'tragic miscalculations,' 'inexplicable blunders,' etc...

An amazing document on many different levels. Anybody done a 'de-bunking' of Layton, or found he made stuff up or something? Because if not, wow.

Some quotes and notes of interest to JN-25 (and other intel) from pages 93-95:

Quote:

“although a red machine had been sent to Hypo, it was not utilized because there was no local traffic in this diplomatic code. Much more serious was that the Purple machine originally destined for Pearl Harbor had been diverted in January to the British”

This was in a trade-off for help on German codes. [this is all '41]

Quote:

“The net result of Washington’s code-breaking trade with the British was frustration for Rochefort and the Navy’s best cryptanalysts: Dyer and Wright. It was also ultimately to prove disastrous for the Pacific Fleet.”

Oops, but it did coincidentally help the US get its casus belli to help England in the war. Ironic, that fate stuff.

Quote:

“When station Cast informed Washington that they now intended to make JN-25 their first priority, there was a twofold objection. They were told by OP-20-G that their cryptographic unit “was not sufficiently supplied with the necessary statistical machinery,” and that it has been intended to “transfer this attack to Pearl Harbor in July.”

24 April – Cast ordered to “concentrate on breaking the the current JN-25 cipher anyway, and that “project will not be transferred to fourteenth as planned.”” [That means Pearl] 2 May – even after Rochefort’s arrival at Hypo, that team was again ordered to focus solely on the Flag Officer's Code (designated AD).

Quote:

"the denial of access to Hypo of the main Japanese operational cipher which resulted from the wrangling over JN-25 was a major blunder..."

It looks like station Hypo - at Pearl Harbor - was denied their own Purple machine, and left reliant on summaries from Washington, which rarely if ever arrived. Naval intercepts were unreadable anywhere, but some at Hypo wanted to help learn but weren't allowed.

Quote:

"If Hypo had set to work on JN-25 as originally intended in July, it is probable that the cipher could have been penetrated in time — especially with the cooperation of the cryptanalysts at at Cavite and Singapore – for the course of history to have been changed. This could be why the various Pearl Harbor inquiries dealt only with our ability to read the Purple traffic, and avoided all but the vaguest reference to JN-25."

Quote:

“Just how badly Washington had goofed in dividing its cryptographic resources would become apparent after 7 December 1941."

Those goofy goofballs, always goofing up!

Anyone have any thoughts on the significance of this stuff? I promise not to jump all over you right off. I'm really not quite ready to argue the point yet, like I tried above, just fishing for thoughts.

[dudalb]

Ronald Lewin's "The American Magic" has a good criticism of Layton's book.
It has some valuable information, but also a lot of "The Navy was not at fault" or more precisely "Kimmel was not at fault" pleading that a lot of ex naval officers tended to indulge in.

[Caustic Logic]

Originally Posted by dudalb

Ronald Lewin's "The American Magic" has a good criticism of Layton's book.

Thanks for the tip, that's new to me. I'd be interested to see what he has to say, and it's available cheap, but originally published 1982, three years before Layton's book. Must be this is in a later edition, but I'm not finding that on book sites (not apparently edition-specific). Any clues? Have a copy handy?

Quote:

It has some valuable information, but also a lot of "The Navy was not at fault" or more precisely "Kimmel was not at fault" pleading that a lot of ex naval officers tended to indulge in.

"indulge" in "pleading." Another odd behavior attributal to anyone but FDR et al. So do you or Lewin think this fault-shifting is inaccurate, and it was Kimmell's/navy's fault?

It's true that Layton makes a lot of claims that are hard or impossible to double-check - some of the points of importance are supported by other sources, others... I just can't say. IF there was some kind of gross negligence that kept PH clueless, Layton himself would be a prime suspect and thus likely to try extra hard to shift the blame. And he does it effectively.

[Caustic Logic]

Originally Posted by JoeyDonuts

1. The Japanese spy Takeo Yoshikawa made his reports using PURPLE diplomatic code.

This may be correct, but I'd like to see a source if you have one. Acc. to Layton, ensign Yoshikawa was a Japanese Third Bureau (Navy intel) operative, working at the consulate, and app. reporting via Consul Kita, who sent the info back on the J-19 consular code. Philippines and other places used PURPLE, and made messages easier to read – 5 times the number of Manila messages as Honolulu were read at this time. For some reason, Hawaii did not have a Purple machine, but used a different system.

ch 14 is all about this stuff:
September 24 the original transmission was sent to Honolulu asking that the waters to be divided according to a grid and ship locations in that grid pinpointed + updated. Kita supplied the first info first 9/29, then weekly then bi-weekly “ships-in-harbor reports.” This was diplomatic traffic, but acc. to Layton sent in J-19 Consular code. This was mailed unread to Washington via courier, then at least some of them were decrypted and read.
Army decrypted the original 'bomb-plot message' 9 oct, navy cracked the 1st response giving the coordinates the next day. The fleet, and eventually whole surrounding base area had an invisible tactical bombing grid placed over it and intel in Wash knew about it (or could/should have).

Other J-19 messages from just before this had indicated naval intel and foreign ministry were arranging espionage ‘couriers’ be sent to Honolulu, helping give the context. Of these and the later bomb plot messages Layton writes: “Neither the Fourteenth Naval District nor our headquarters was ever told about these early signs […] Washington’s failure […] was tantamount to folly.” [164]

Some (Kirk, Bode, Safford) wanted to specifically inform Kimmell but were stopped by others (Noyes). In the investigations, some professed an “impression” that these were somehow sent to Pearl, while others (Stark, Turner) professed to not recalling the messages at all, and dismissing the significance if they HAD seen them as showing the Japanese "attention to detail". Oh you know the Japanes, they love to, um, make really accurate scale models of our harbors, and - um - real-world accurate ship placements worked in. Yeah, it's a zen thing, that's the ticket.

Kimmell said if He'd heard about this stuff, it would have totally changed his assessment. Layton concurs while citing his old boss.

Quote:

“…the failure of the office of naval operations to ensure that the bomb plot messages were sent to us at Pearl Harbor was blind stupidity at the least, and gross negligence at the best.” [167]

Sounds like more sour grapes from another wheel-sleeper-atter. Bomb plot message never sent on, ridiculous! Of course it was, right? Like the "war warning?" Anyone counter this?

[JoeyDonuts]

Originally Posted by Caustic Logic

This may be correct, but I'd like to see a source if you have one.

Scratch the PURPLE code portion. That's actually from the wiki page on Yoshikawa. However, the source cited for this info is a Time magazine article that doesn't say anything about what code was actually used to pass the information along. I'd go with your source. Yoshikawa used the J-19 consular encryption.

We do know one thing: He didn't know an attack was coming until he heard the phrase "East Wind, Rain." Apparently every single Japanese consular employee payed AWFULLY close attention to the weather forecast back in those days.

[Caustic Logic]

[quote=JoeyDonuts;4455838]Scratch the PURPLE code portion. That's actually from the wiki page on Yoshikawa. However, the source cited for this info is a Time magazine article that doesn't say anything about what code was actually used to pass the information along. I'd go with your source. Yoshikawa used the J-19 consular encryption.[/QUOTE}

And it's not even a major point! Readable diplomatic traffic it was. Thanks for the link, another for the list.

Quote:

We do know one thing: He didn't know an attack was coming until he heard the phrase "East Wind, Rain." Apparently every single Japanese consular employee payed AWFULLY close attention to the weather forecast back in those days.

In fact there's been some controversy over this issue, the winds message(s). I haven't looked at it close yet, and not sure I get what it's saying, but there's some new findings from late last year.
http://www.upi.com/Top_News/2008/12/...1811228612980/

Also some points from Layton regarding my Q3 about British and Dutch decodes of JN-25 forthcoming. Interesting stuff...

[fezzic]

RE: #18

I don't think the presence of Purple at Pearl Harbor would have meant anything. My limited understanding was that the Consulate communicated with Japan via the cable companies which observed the privacy of entrusted cablegrams. I understand that it was supposedly a plea from USN to a president of a cable company that brought cooperation which would leaking copies starting December 1 (the Consulate rotated business among the companies).

I would take issue with the idea that with something as potentially fateful as war and a early (tho after declaration) surprise attack that it would be communicated to anyone without a strict need to know. If the Washington DC embassy had, in fact, been alerted to the attack it would have been made aware of the importance of getting the break in negotiations (defacto declaration of war) presented at the prescribed time (to remain 'legal'). If there was any doubt on the ability of the embassy to do so, chances were that measures would have been taken to ensure compliance.

BTW, from TCB (abridged edition), it gave an example of a message sent from the Consulate by Yoshikawa shortly before the attack. It was encrypted in the PA-K2 code because (it was said) the higher level codes had been destroyed on orders from Tokyo.

An explanation of the wind code was that because communications cease between belligerents, it was a way to remind outlying diplomatic posts to destroy their codes and ciphers in case they didn't get the word that war had started. That Yoshikawa might have used it but I don't see why. He was a diplomat (and spy). He can use the Consulate to send his messages. Diplomats go home or are interned. Diplomatic immunity covers that.

Certainly a lot has been brought out that I didn't know before.


TCB = The Codebreakers by David Kahn

[Caustic Logic]

Originally Posted by fezzic

RE: #18
My limited understanding was that the Consulate communicated with Japan via the cable companies which observed the privacy of entrusted cablegrams. I understand that it was supposedly a plea from USN to a president of a cable company that brought cooperation which would leaking copies starting December 1 (the Consulate rotated business among the companies).

Thanks for the thoughts. I've heard about the cable company aspect but don't know much, and didn't catch Layton's take on that part - that may be how they were sent - coded/encyphered, of course - and how intercepted. Stinnett has a little conspiracy theory about RCA's Sarnoff and FDR having a secret deal. I don't know, if he's the only one saying that it's probably bunk. Another thing to look into besides the winds controversy and a thousand others.

Quote:

An explanation of the wind code was that because communications cease between belligerents, it was a way to remind outlying diplomatic posts to destroy their codes and ciphers in case they didn't get the word that war had started. That Yoshikawa might have used it but I don't see why. He was a diplomat (and spy). He can use the Consulate to send his messages. Diplomats go home or are interned. Diplomatic immunity covers that.

I doubt Yoshikawa needed to know this code, but perhaps did. I don't think it was all a major secret - the consulate people would know about it, and he worked there as an "assistant", and he already knew or guessed PH was a target.

Regarding JN-25 decoding, Layton reports on page 206 that British had good luck with penetrating JN-25, but is not specific. Other sources vaguely estimate about equal to US progress.
He finds reason however to believe the Dutch cryptographers had better luck – their files were lost in the chaos of invasion, but many have testified to a knowledge that could almost only have been gleaned from JN-25 intercepts.

They knew of a Japanese buildup in the Kuriles, which can only be the PH striking force. Finding no evidence of Dutch submarines that far north, the strict radio silence of the striking force, Layton reasons the transmitted, intercepted, and read order-to-sail of Nov 25 "would have been the only source of information" tipping them off to this true fact, a fact directly supported by some recollections.

Did they know Hawaii was the target? Unsure, but the same traffic they seem to have been reading revealed this. Via close cooperation with the British on code work, the Dutch news could have been passed on to London, which could be quite interesting for reasons I'll explain later.

[JoeyDonuts]

Originally Posted by Caustic Logic

Did they know Hawaii was the target? Unsure, but the same traffic they seem to have been reading revealed this. Via close cooperation with the British on code work, the Dutch news could have been passed on to London, which could be quite interesting for reasons I'll explain later.

That dog don't hunt. Even if there were cable traffic that equivocally named Pearl as the target, it must not have been voluminous enough to warrant a plan of action. Japanese aggression in the Pacific Rim was noted, but it was so far removed from our way of thinking at the time that CINCPAC could have received several decrypts and still not given the fleet orders to emergency sortie to clear the harbor.

Their distrust of the new radar system that supposedly picked up the first wave of attack aircraft certainly didn't help either. In all fairness, the system was relatively new and they were on the lookout for a returning flight of B-25s IIRC. Then again, the radar skin paint on a group of Mitchells vs. 300 + enemy aircraft is completely different. But I digress, new technology - and not likely to be relied on to the degree it is now.

[Caustic Logic]

Originally Posted by JoeyDonuts

That dog don't hunt. Even if there were cable traffic that equivocally named Pearl as the target, it must not have been voluminous enough to warrant a plan of action.

Now wait - I'm not sure you understand. If this traffic was intercepted and even partly understood, it could have revealed any or all of the following:
- Striking force contained six aircraft carriers
- Force was to head east on the morning of Nov 26
- The final target was the US Fleet in Hawaii, which was at the moment weakly guarded, with an eye to destroying it.
- the attack was planned for dawn on Dec 7.
How voluminous would this info need to be? If this vague, unsubstantiated info was intercepted and known, why hide the fact? Because it may well have been available but hasn't been acknowledged. I feel it would have been taken seriously by anyone who knew anything and wasn't trying to miss the clues, but that's just my opinion I guess.

For a little more insight, Layton cites Dutchmen familiar with their intel as knowing the location of this force in the Kuriles, which shows they had some insight into location signifiers i JN-25. Also, one Dutch source I don't recall where I saw it, said he felt the Japanese were planning an attack on Siberia.
This might indicate they could not make out the target location codes for "Hawaii" or "US Fleet," but might also mean they knew it was to head east from there. East from Hitokappu bay could mean attack Siberia or Aleutians (left turn), Seattle-ish straight ahead with lots of fuel, or Hawaii with a right turn. Or these statements might have been some kind of 'limited hangout,' reflecting the political situation, or made up, or whatever.

But here we are in the land of "citing recollections of long-dead military people..." It's probably impossible to know for sure what was known here.

Quote:

Japanese aggression in the Pacific Rim was noted, but it was so far removed from our way of thinking at the time that CINCPAC could have received several decrypts and still not given the fleet orders to emergency sortie to clear the harbor.

Opinion, and one that could be countered by all kinds of evidence. Specific Japanese moves were anticipated and fretted over - they were a real enemy that was taken seriously. We can't say for sure what they would have done at that time - Kimmel and Layton et al. insisted after the fact they would have done much more if they had the right intel, but they would say that either way.

Quote:

Their distrust of the new radar system that supposedly picked up the first wave of attack aircraft certainly didn't help either. In all fairness, the system was relatively new and they were on the lookout for a returning flight of B-25s IIRC. Then again, the radar skin paint on a group of Mitchells vs. 300 + enemy aircraft is completely different. But I digress, new technology - and not likely to be relied on to the degree it is now.

I have my doubts as well about the radar confusion, but am willing to chalk that up to error/coincidence/unfamiliarity if that makes most sense. OTH, Since I suspect officials wanted total surprise, they might have to consider radar, which had the capacity to screw that up at the last minute. IIRC the planes were first picked up about 7:02, and didn't get over the harbor til app 7:55. That's enough time to spoil the surprise and get Pearl's planes up. Luckily (??) the report of B-25s due in plus whatever else was sufficient to mentally short-circuit the radar alert system.

[JoeyDonuts]

Originally Posted by Caustic Logic

Now wait - I'm not sure you understand. If this traffic was intercepted and even partly understood, it could have revealed any or all of the following:
- Striking force contained six aircraft carriers
- Force was to head east on the morning of Nov 26
- The final target was the US Fleet in Hawaii, which was at the moment weakly guarded, with an eye to destroying it.
- the attack was planned for dawn on Dec 7.
How voluminous would this info need to be? If this vague, unsubstantiated info was intercepted and known, why hide the fact? Because it may well have been available but hasn't been acknowledged. I feel it would have been taken seriously by anyone who knew anything and wasn't trying to miss the clues, but that's just my opinion I guess.

I understand the traffic that was intercepted. I also know how long it took the military decisionmakers to rely on cryptography and any information derived from it. Another thing you have to consider (and this is PURELY speculation on my part) is that perhaps the Japanese high command issued orders for multiple different targets in order to obfuscate their true motives. Disinformation messages, if you will, to divert the adversary's attention towards a bogus target. This was employed by the Allies prior to D-Day IIRC.

You are dead on about one thing - the fog of war and time will prevent us from ever definitively proving one way or the other exactly WHAT information CINCPAC or his staff had in their hands at the time.

Interesting thread. I'll do some looking of my own and look for any non-woo tidbits I can bring to the table.

[Caustic Logic]

Originally Posted by JoeyDonuts

I understand the traffic that was intercepted. I also know how long it took the military decisionmakers to rely on cryptography and any information derived from it.

This has been stated a lot, emphatically. Only after Pearl 12/7...

Quote:

Another thing you have to consider (and this is PURELY speculation on my part) is that perhaps the Japanese high command issued orders for multiple different targets in order to obfuscate their true motives. Disinformation messages, if you will, to divert the adversary's attention towards a bogus target. This was employed by the Allies prior to D-Day IIRC.

Sure, that's something they'd think about too I'd guess. Not to mention they were mobilizing in all directions, since they attacked like a million places in the hours and days after hitting Hawaii. So there was tons going on, and mobilizations, transmissions, stuff.

[JoeyDonuts]

Originally Posted by Caustic Logic

This has been stated a lot, emphatically. Only after Pearl 12/7...

Something about the horse getting out of the barn, and pants around the ankles...that sort of thing.

[Corsair 115]

Originally Posted by JoeyDonuts

Their distrust of the new radar system that supposedly picked up the first wave of attack aircraft certainly didn't help either. In all fairness, the system was relatively new and they were on the lookout for a returning flight of B-25s IIRC.

It was a flight of B-17s, not B-25s.

Originally Posted by JoeyDonuts

Then again, the radar skin paint on a group of Mitchells vs. 300 + enemy aircraft is completely different.

It's more than that. The B-17 flight was coming in from an entirely different direction from that picked up on the radar return. But that little fact wasn't really appreciated. That same radar site could have also tracked the outgoing Japanese aircraft back towards their fleet.

Originally Posted by Caustic Logic

- The final target was the US Fleet in Hawaii, which was at the moment weakly guarded, with an eye to destroying it.

Would the coded have transmission have stated outright the target was Pearl Harbor? That would seem unlikely to me. Wouldn't they have used an additional code for each target, similar to the way Midway Island was target 'AF'?

[Caustic Logic]

Cool, two responses!

Originally Posted by JoeyDonuts

Something about the horse getting out of the barn, and pants around the ankles...that sort of thing.

Of course, I didn't mean to totally dismiss that argument, just qualify it. It didn't go from 0% interest to 100%, and IMO the impression of codelessness pre-12/7 has been exaggerated after the fact.

Originally Posted by Corsair 115

It was a flight of B-17s, not B-25s.

Was it 48 of them by chance?

Quote:

Would the coded have transmission have stated outright the target was Pearl Harbor? That would seem unlikely to me. Wouldn't they have used an additional code for each target, similar to the way Midway Island was target 'AF'?

Correct, it would have been in code, so it's almost certain if the decryption was broken some code words and not others would be readable - the code was neither 0% nor 100% broken by anyone (as far as we know). Decryption was likely, since the latest additive change was nearly 4 months old by the time of the order to sail, and the code itself was almost a year old, with continuous progress that was notably slow on the US end anyway.

Also, the order to sail (see below) was preceded by about two weeks of differently-detailed transmission saying about the same things in different parts. The info should have been about like this reconstruction of the final order cited in the JCC's '46 report (no actual JN-25 decodes to compare):

Quote:

”(a) The task force, keeping its movements strictly secret and maintaining close guard against submarines and aircraft, shall advance into Hawaiian waters and upon the very opening of hostilities, shall attack the main force of the United States Fleet in Hawaii and deal it a mortal blow. The first air raid is planned for dawn of X-day (exact date to be given by later order).

Upon completion of the air raid the task force, keeping close coordination and guarding against enemy counterattack, shall speedily leave the enemy waters and then return to Japan.

(b) Should it appear certain that Japanese-American negotiations will reach an amicable settlement prior to the commencement of hostile action, all the forces of the combined fleet are to be ordered to reassemble and return to their bases.

(c) The task force shall leave Hitokappu Bay on the morning of November 26 and advance to 42° N. And 170° E. (standing-by position) on the afternoon of December 4, Japan time, and speedily complete refueling. “

Here are reconstructions of more communications with the striking force, not necc. all in JN-25.
http://www.ibiblio.org/pha/myths/jm-097.html

Thanks all for engaging!

[fezzic]

Personally, I don't see why even that last would be put on the air in any code. There were only about 30 ships in the task force and coordination within the task force would be fairly easy. It is not as if the plan of attack was being formed up and only finalized shortly before the task force departed (though I admit maybe it was) such as with Midway.

I would point out that date/time was enciphered within JN-25 messages separately (I think). There was reference to the cipher used in TCB (The Codebreakers) where it had been unsolved until an overnight effort at Pearl Harbor finally solved it during the runup to Midway. According to TCB, that confirmed the estimates by the intell staff about when the Midway operation was going to start. This might not be relevant to Pearl Harbor because supposedly it was seen only three other times (one with a garble that caused problems) of which once was at the Battle of the Coral Sea. something like that.

[Caustic Logic]

First, to clarify from before - location indicators were in a different code-within a code than the normal numbers-for-kanji scheme. I think these were given roman letter abbreviation, as AF was found to mean Midway. (at least that's how I remember it as well), and so maybe "A" has a 5-digit string and "F" another. I don't know what Hawaii's indicator would be or who may have known it in late Nov 41.

Originally Posted by fezzic

Personally, I don't see why even that last would be put on the air in any code. There were only about 30 ships in the task force and coordination within the task force would be fairly easy. It is not as if the plan of attack was being formed up and only finalized shortly before the task force departed (though I admit maybe it was) such as with Midway.

Apparently it was transmitted - the whole point of a naval code and encypherments was so they could send secure stuff out. In retrospect at least, they shouldn't've relied on that, but it would seem they did.

Quote:

I would point out that date/time was enciphered within JN-25 messages separately (I think).

I think that's correct, actually. Numbers were the first thing that popped out of JN-25A, 000-999 in one mathematical insight. Since dates, times, and geo coords are given in numbers, it would make sense to again uses a code-within-a-code for these so even when they get the numbers, they don't seem to mean what they should.

So yeah, these considerations need to be taken into account when wondering what might have been extractable.

Quote:

There was reference to the cipher used in TCB (The Codebreakers) where it had been unsolved until an overnight effort at Pearl Harbor finally solved it during the runup to Midway. According to TCB, that confirmed the estimates by the intell staff about when the Midway operation was going to start. This might not be relevant to Pearl Harbor because supposedly it was seen only three other times (one with a garble that caused problems) of which once was at the Battle of the Coral Sea. something like that.

JN-25 was used a lot, but not seen much at Hypo until after the attack. As noted above (post 13), there were thousands of messages in this code each month. They did have 'garble checks' built in, but I'm not sure what that is, and if you can't read it, it would look garbled.

[JoeyDonuts]

Originally Posted by Caustic Logic

They did have 'garble checks' built in, but I'm not sure what that is, and if you can't read it, it would look garbled.

IIRC the "garble check" was some kind of internal sequencing inherent to the code - sort of a checksum that was exploited and opened the door to cracking the entire cypher.

ETA: Our friends at the No Such Agency have a little insight to offer on this.

Quote:

Navy cryptanalysts used IBM machines to correlate and compare values in JN-25-based communications. This allowed them to identify depths, which were the key to recovering the numerals used in messages. This method also enabled them to identify a "garble check" the Japanese used to ensure each message had been copied and decoded correctly. This was of prime value in helping to strip away the additive numbers and get to the underlying codebook values.

Once the additive had been stripped away, U.S. cryptanalysts took advantage of stereotyped messages to recover some basic text. Commonly repeated texts included vessel port entry requests, daily position data for convoys, and medical reports.

[Caustic Logic]

Total side note: I decided to look up the authors of this book I keep citing. IIRC it was actually finished after Layton's death, and carried on by co-authors Capt. Roger Pineau and John Costello. All three are dead by now.

Layton's wikipedia page:
http://en.wikipedia.org/wiki/Edwin_T._Layton

Pineau's obituary:
http://query.nytimes.com/gst/fullpag...52C1A965958260

Now Costello... I suspect he was the main wordsmith, and a bit of a CTist - believed in adding "color, light, action" to his writings, which included the Pacific War, Pearl harbor (Days of Deceit), Soviet Spies, homosexual/Soviet rings or something, public morals and communist hedonism, and Rudolph Hess' bizarre 'peace mission' to England, etc. Quite an unusual catalog.

He died in 1995 of food poisoning - obituary:
http://query.nytimes.com/gst/fullpag...260&sec=&spon=
Four months after testifying to Kimmel's innocence:
http://users.erols.com/nbeach/kimmel.html
Was he whacked? Illuminati news thinks maybe, which means YES, the jews did it!
http://www.illuminati-news.com/091006b.htm
David Irving suspects foul play for a writer of "real history":
http://www.fpp.co.uk/Letters/fan/Orourke_020605.html
Was he hardcore gay? Somebody with JFK secrets says so:
http://www.think-aboutit.com/thinker...he-white-house


So yeah... that surely doesn't help the book with many here, but I have to consider the info itself as it stands. Eve oddballs can be correct when they're trying to, but I've got my grains of salt handy.

[fezzic]

My understanding of how to strip the additive groups was that the USN used the "difference" method (the unabridged TCB IIRC had a short description of the process, I haven't gone to the library to read that passage again in many years so this is all I really remember). As described in TCB (abridged), the computers assisted and sped up the cryptanalysis process in so far as things were known thereby freeing up cryptanalysts to tackle the unknown. Anyway, I have never heard of a garble check as such. Checksums, yes or something similar, but usually you can derive the check from the content of the message or code group(s) at some point.

Well, I have no real background in crypto so I would have to defer to others with that kind of background.

[JoeyDonuts]

Originally Posted by fezzic

Well, I have no real background in crypto so I would have to defer to others with that kind of background.

Neither do I as such. I was an Electronic Warfare technician in the USN for 6 years...but that mostly deals with ELINT derived information and countermeasures. However, we were all converted to CT's my third year or so. That opened the door to learning more about this kind of thing. One of the really-real CT's assigned TAD to the Barry was a real crypto-history buff. Apparently it was common and an internal function of JN-25 to have those incidental bits of code that worked as sort of a "handshake" protocol.

The IBM machines did work on codifying and categorizing the known parts of the code, thus freeing up human resources to focus on the unknown parts.

Sorry if I'm relating this incorrectly, but it was kinda hard to hang in there with her as she was describing this stuff. I dealt with radar analysis and ASMD countermeasures, not nuts-and-bolts code stuff. The idea fascinates me, but I think the execution would most likely make me fall asleep.

[fezzic]

Just thought I'd add a couple thoughts.

Prior to the start of WWII, the number of cryptanalysts was certainly quite limited. One may note that the Combat Intelligence Unit (aka Station HYPO) at Pearl Harbor had 3 actual cryptanaysts. The majority of personnel assigned were devoted to direction finding and traffic analysis -- useful in themselves, but not code breaking. This would mean that the amount of intercepts that the unit could handle was limited and choices were made with respect to what intercepts in what cryptosystem would be handled first. Once the war broke out, the number of cryptanalysts was rapidly expanded (along with everything else) so the ability of the CIU to handle the increased intercept traffic would likewise expanded greatly.

It should also be noted that competent (reliable) translators were also not easy to come by. Telegraphic japanese was described in TCB as being pretty hard to translate mostly because of the lack of the kanji -- the translator takes the syllables, groups them into what appears to be 'words', finds the kanji, then translates into english, assuming the japanese makes sense. So being able to read, write, and speak Japanese did not make a person into a translator especially when one considers what might happen with an incorrect translation.



I don't know if anybody really realizes (maybe it is just me) but it seems that people tend to conflate "breaking a code" and "breaking a cipher" as equivalent. Namely that if a code is "broken" then any message sent in it can be essentially read with maybe small gaps. I think a code is "broken" when the first code groups are recovered (I think it may also be called 'solved').

Whereas breaking a cipher, such as the ADFGVX cipher of WWI, means that the structure and form of the cipher has been determined with secrecy residing solely in the keys, a code is not so vulnerable. Even if a crib reveals a number of code group equivalences, it merely helps in recovering additional groups. So a code may be "broken" but yield no useable information other than to the cryptanlysts, but eventually it may be "broken" or "solved" to the point where a fair portion of traffic is read.

So the idea that JN-25b was "broken" by a particular time, with the implication that everything could be read, has to be taken with a bit of salt because codes can be "just a little bit broken". (IMO)

[Caustic Logic]

And for the record, I know probably less about cryptography than either Fezzic or JD, so a lot of what I'm saying here is guesswork. Good points in the previous post.

I was going to post the little mystery that got me looking onto JN-25, but I wound up getting too drunk to start. If anyone knows what I'm getting at and wants t debunk it already, you go. I'm talking Churchill-to-FDR telegram of 3 am Nov. 26 1941. It might be nothing, but...