Finnish Radio Intelligence

[paspartoo]

Has anyone read ''Suomen Radiotiedustelu 1927-44'' by Pale Erkki and Ahtokari Reijo ? I'm interested in the crypto systems that the Finns could read,especially Soviet 5-figure codes and the M-138 strip cipher.

[Mangrove]

I'm interested in the crypto systems that the Finns could read,especially Soviet 5-figure codes and the M-138 strip cipher.

Cryptos that Finnish intelligence could read as of 1944, as per the book. This is not complete list as I have personally seen examples of other nation's diplomatic messages.

- Soviet 2-figure code used by some Red Army units and NKVD.
- Soviet 3-figure code used by NKVD battalions and regiments.
- Soviet 4-figure code used by the Baltic Navy and some Red Army units ("OKK-5").
- Soviet 5-figure code.

- United States M-138A ("strip") -code.
- Unites States "Brown" -code.
- United States "Gray" -code.
- United Kingdom Government Telegraph Code.

[paspartoo]

Thanks for this information! Can i ask a few questions?
Did the Finns use Hollerith/IBM punch card machines for cryptanalysis?
Can you provide more details on the extent of their success with the Soviet 5-figure in the period 1943-45 and the American strip in 1944? Are specific strips mentioned?
Also according to postwar German reports from 1943 the Soviets used one time pad extensively on their 5-figure code, so solution was practically impossible. Is this confirmed by the book?

[Mangrove]

Did the Finns use Hollerith/IBM punch card machines for cryptanalysis?

The work was done by hand during the beginning of the Continuation War but later Hollerith machines were bought from Germany.

Can you provide more details on the extent of their success with the Soviet 5-figure in the period 1943-45 and the American strip in 1944? Are specific strips mentioned?

It seems that the Soviet troops ended the use of the 5-figure code at the Finnish front during late-1941 as the result of the discovery that Finns had cracked it during the Winter War and the beginning of trench war where phone lines were more useful (although Finns were also capable of listening them).

I have few examples of strips from year 1943 which I found from the National Archive (translation from Finnish to English by me), e.g. http://forum.axishistory.com/viewtopic.php?f=59&t=140703&p=1548020

From Ankara to Washington, March 11th 1943. No 442.
Sgt. Carl Durfee and Cpl. David D. Coward, two of our internated pilots who, according to instructions from the embassy, have not given their words of honor, have fled from their accommodations from Ankara and are now on British s/s Bendera, which have left Iskenderun to Egypt. Report this to General Arnold.
Steinhardt.

Another from the book (translation again by me):

Helsinki-Washington
6 July 1942

SECRET. For the first time during this War, arrived Finnish coastal defence ship Väinämöinen to the Bay of Helsinki on Saturday evening. She was anchored to a point half a mile directly south of the eastern end of Kulosaari island. Schoenfeld.

When combining the information I have seen at the Archive and read from the book, it's quite clear that Finns could read at least some of the diplomatic messages they captured from following countries: Brazil, Portugal, Romania, Soviet Union, Turkey, United States of America, Vatican, Vichy France and Yugoslavia (General Draža Mihailović's troops).

[paspartoo]

I have info on the Finnish successes from ’Finland's Codebreaking in World War II’ in ‘In the Name of Intelligence: Essays in Honor of Walter Pforzheimer’ written by David Kahn. As you mentioned the codes of Turkey and Tito and Mihailović were among those read.
The M-138 strip was a high level code and it was used with two different sets of strips. The general ones for communications to all embassies (0-1,0-2 etc) , the special ones for communications between a specific embassy and Washington ( for example 38-1 Washington-Moscow).
Can you check historical details on the strip? How and when did they break it. What links were broken, what help did they receive from the Germans and/or the Japanese.
Also how much of the strip did they read in 1944? Book ‘Swedish signal intelligence’ says one of the US messages they decoded on 30 May 1944 revealed the date of ‘Overlord’!

[Mangrove]

I have info on the Finnish successes from ’Finland's Codebreaking in World War II’ in ‘In the Name of Intelligence: Essays in Honor of Walter Pforzheimer’ written by David Kahn.

On the contrary what I wrote based on Pale's book, Soviet did use and Finns did capture and decipher them during the whole War. Ohto Manninen's 2002 book "Stella Polaris" contains reproduction original monthly radio intelligence reviews which were not available during the time Pale wrote his book.

The first review is from December 1941. During the month Finns captured 731 "general" 5-figure coded messages, of which 171 or 23 % were deciphered. "Personal" 5-figure coded messages were harder to break, only 64 of 2768 messages (2 %) captured during the month were deciphered. The same percent figures for 4-figure codes was 16 %, 68 % for 3-figure and 72 % for 2-figure codes. In total 3321 (39 %) of the 8609 messages captured during the month were deciphered.

The same statistics are available for almost every month. Most of the 5-figure codes (74 % of them) were coded using the "personal" code and only 0,5 % with the "general" code which was the easiest one to break. During the year 1943 the Finnish intelligece deciphered only 172 (1 %) of the captured 22208 Red Army 5-figure coded messages and 0 of the 10701 captured NKVD 5-figure coded messages. These statistics are for messages that used some kind of addition during the crypting process. On the other hand at the same time 48 % of the Red Army 5-figure messages crypted with some kind of replacement method were deciphered by the Finns.

The message concerning the route and timetable of Convoy PQ18 in September 1942 was crypted with ordinary 2-figure code through Soviet Air Force radio network and thus was easily deciphered by the Finns. The message was captured on 6 September and was recorded on Seekriegsleitung's war diary on the following day.

Can you check historical details on the strip? How and when did they break it. What links were broken, what help did they receive from the Germans and/or the Japanese.

The Finnish intelligence broke the STRIP in summer 1942. The Americans made some rudimentary mistakes that helped Finns. For example they might have sent the same message from Washington to all embassies using the individual strips. If you knew one key, then you could naturally start to crack the others. They also frequently used the same word as a starter ("SECRET" or "SPTEL"). The Overlord case is not mentioned in any of my books but following the Madrid embassy Finns could also follow the start of Operation Torch in North Africa.

Finns and Germans changed some information regarding some codes but not always the most highest level codes. Some material regarding the STRIP has said to be given by Admiral Canaris to Finns in 1942. For example both countries shared their knowledge on Soviet armour turret coding and unit code numbers.

[paspartoo]

Thanks for the interesting information. The Soviet 5-figure code was numerical and enciphered with either General or Individual additive tables. Individual were supposed to be used only once but this sometimes did not happen.
According to the Germans in 1941-2 the Russians used General tables extensively:
http://chris-intel-corner.blogspot.com/ ... codes.html

If the 5-figure was also enciphered with substitution tables it's something i've read only for 1941.
It's interesting to see detailed statistics. If it's not a bother i could use the stats for 5 -figure for each month.

For the strip: http://chris-intel-corner.blogspot.com/ ... rmany.html
I have read in a report that the Germans gave the Finns strips 0-1. These had been given to them by the Japanese
who copied them from the American consulate in Kobe. The Germans solved 0-2 to 0-5 and many specials. Altogether 22 are mentioned solved by Erich Huettenhain ,chief cryptanalyst of OKW/Chi.
I am interested in the success the Finns had with it in 1944 .

[Mangrove]

The Soviet 5-figure code was numerical and enciphered with either General or Individual additive tables.

Finnish Intelligence recognized three different bloknots:
General = общий блокнот
Individual = индивидуальный блокнот
Circular = циркулярный блокнот

If it's not a bother i could use the stats for 5 -figure for each month.

The reviews are not complete due of the nature of Operation Stella Polaris after the War. However I collected these figures from the reviews for you.

G = General
I = Individual
C = Circular
***/*** = captured/deciphered

Red Army & Soviet Air Force 5-figures:
December 1941:
G-5-fig = 731/171 = 23 %
I-5-fig = 2768/64 = 2 %

January 1942:
G-5-fig = 383/45 = 12 %
I-5-fig = 2759/183 = 7 %

June 1942:
G-5-fig = 41/29 = 71 %
I-5-fig = 78/14 = 18 %
C-5-fig = 14/5 = 36 %

July 1942:
G-5-fig = 0/0 = 0 %
I-5-fig = 9/3 = 33 %
C-5-fig = 18/0 = 0 %

September 1942:

G-5-fig = 0/0 = 0 %
I-5-fig = 38/? = ? %
C-5-fig = 41/? = ? %

January 1943:
5-fig-additive = 2513/47 = 2 %
5-fig-substitution = 80/0 = 0 %

February 1943:
5-fig-additive = 4185 / 17 = 0,5 %
5-fig-substitution = 50 / 0 = 0 %

March 1943:
5-fig-additive = 3625 / 19 = 0,5 %
5-fig-substitution = 107 / 59 = 55 %

April 1943:
5-fig-additive = 2435 / 48 = 2 %
5-fig-substitution = 29 / 23 = 80 %

May 1943:
5-fig-additive = 1007 / 7 = 0,7 %
5-fig-substitution = ? / ? = ? %

June 1943:
5-fig-additive = 713 / 0 = 0 %
5-fig-substitution = 56 / 0 = 0 %

July 1943:
5-fig-additive = 669 / 0 = 0 %
5-fig-substitution = 33 / 0 = 0 %

August 1943:
5-fig-additive = 1651 / 0 = 0 %
5-fig-substitution = 25 / 0 = 0 %

September 1943:
5-fig-additive = 1160 / 28 = 3 %
5-fig-substitution = 37 / 34 = 92 %

October 1943:
5-fig-additive = 1075 / 6 = 1 %
5-fig-substitution = 78 / 68 = 87 %

November 1943:
5-fig-additive = 1887 / 0 = 0 %
5-fig-substitution = 88 / 82 = 93 %

December 1943:
5-fig-additive = 1288 / 0 = 0 %
5-fig-substitution = 25 / 24 = 96 %

January 1944:
5-fig-additive = 1620 / 0 = 0 %
5-fig-substitution = 58 / 56 = 97 %

February 1944:
C-5-fig-additive = 708 / 0 = 0 %
I-5-fig-additive = 1919 / 0 = 0 %
5-fig-substitution = 99 / 99 = 100 %

March 1944:
5-fig-additive = 2965 / 10 = 0,3 %

April 1944:'
5-fig-additive = 2728 / 10 = 0,4 %

May 1944:
5-fig-additive = 3153 / 21 = 0,7 %

June 1944:
5-fig-additive = 1487 / 0 = 0 %

July 1944:
5-fig-additive = 3120 / 0 = 0 %

August 1944:
5-fig-additive = 3386 / 7 = 0,2 %

[Mangrove]

Some additional observations of the Red Army 5-figure codes at the Finnish Front and their decipherment by Finns based on Manninen's book.

Red Army:
011-A (September 1941 to Mid-April 1942)
023-A (1 April 1942 to 1 November 1942)
045-A (2 November 1942 to May 1943, "C"-coded continued to Mid-February 1944). Captured by Germans and sent to Finns.
062-A (May 1943 to Mid-February 1944, "C" and "I"-coded only). Captured by Germans and sent to Finns.
091-A (1 February 1944 to the end of the War). Captured by Germans and sent to Finns.

[paspartoo]

Dettmann of the German army’s Horchleitstelle Ost (central cryptanalytic and evaluation center for
Russian traffic located at Loetzen, East Prussia from late 1941 to late 1944) says in Ticom DF-112:
011-A Jan '41 - Oct '41
023-A Oct '41 - Mar '42
045-A Mar '42-Mar '43
062-A Mar '43-Mar '44
091-A Mar '44-Mar '45

[Juha Tompuri]

I have info on the Finnish successes from ’Finland's Codebreaking in World War II’ in ‘In the Name of Intelligence: Essays in Honor of Walter Pforzheimer’ written by David Kahn. As you mentioned the codes of Turkey and Tito and Mihailović were among those read.
The M-138 strip was a high level code and it was used with two different sets of strips. The general ones for communications to all embassies (0-1,0-2 etc) , the special ones for communications between a specific embassy and Washington ( for example 38-1 Washington-Moscow).
Can you check historical details on the strip? How and when did they break it. What links were broken

At the autobiography of Finnish War-time president Risto Ryti, he several times mentions that he was given fresh deciphered US telegrams from routes Helsinki-Washington, Washington-Helsinki, Stockholm-Washington and Washington-Stockholm.

Regards, Juha

[Mangrove]

At the autobiography of Finnish War-time president Risto Ryti, he several times mentions that he was given fresh deciphered US telegrams from routes Helsinki-Washington, Washington-Helsinki, Stockholm-Washington and Washington-Stockholm.

These are kept on his private collection at the Finnish National Archive. I believe Ohto Manninen wrote about these in his article published on Sotilasaikakauslehti 6-7/2007 (Mongoliasta ja Puolasta Suomussalmen taisteluihin - radiotiedustelumenestyksen isäehdokkaita riittää.). He has recently written about the participation of the Finnish intelligence to the attack on PQ18 (Avasiko radiotiedustelu kaksi "saattuesanomaa" 1942? - Sotilasaikakauslehti 2/2012) and the Stella Polaris files in Sweden (Stella polariksen sisältö hahmottuu - Sotilasaikakauslehti 10/2011). He has previously also published articles on Nordic countries' pre and post-war radio intelligence.

[Juha Tompuri]

At the autobiography of Finnish War-time president Risto Ryti, he several times mentions that he was given fresh deciphered US telegrams from routes Helsinki-Washington, Washington-Helsinki, Stockholm-Washington and Washington-Stockholm.

These are kept on his private collection at the Finnish National Archive.

Finnish translations of several of them are also at the autobiography.

Regards, Juha

[Mangrove]

Examples of captured, deciphered and translated Turkish diplomatic messages from year 1943. Marked with red "secret" stamp.

[Juha Tompuri]

Today's Turun Sanomat has the obituary of Lauri Kalevi Loimaranta. During the Continuation War he served as a cryptology expert in the General HQ Radio Battalion. He was a key figure in the cracking of the American strip cipher, which he later presented to the American intelligence (OSS) in Stockholm together with Captain Erkki Pale (chief of cracking section).

viewtopic.php?f=59&t=123984

Regards, Juha