The Bletchley Park Codebreakers (61 page)

Edward Simpson would like to point out that this account of Bletchley Park’s cryptanalytical work on the main Japanese naval cipher JN-25 between August 1943 and August 1945 was written from memory in 2010, and by reference to the ‘History of the Fleet General Purpose System (JN-25): Cryptographic Party’, an
internal GC&CS history which he wrote in August 1945. This was released to the PRO in 2009 as HW 8/149. It is not a history of the cryptanalysis. That is believed to be in a separate internal history which he and Ian Cassels wrote: ‘GC&CS Naval Cryptanalytic Studies Volume IX: The Japanese Fleet General Purpose System II’, which is still withheld as HW 43/34. Simpson is most grateful to both editors for the opportunity to put this part of Bletchley Park’s work on the record, he believes for the first time, and for much help in preparing it.

Page
1
The Hagelin C-38 cipher was read currently: For the intelligence derived from the Italian Hagelin traffic, see ‘C38m’ in the index to F. H. Hinsley, with E. E. Thomas, C. F. G. Ransom and R. C. Knight,
British Intelligence in the Second World War: Its Influence on Strategy and Operations, vol. 2
(London: Her Majesty’s Stationery Office, 1981).

Page
2
‘Commander Belfield RN … but without any known Christian name’: Editor’s note: Belfield’s Christian name was unknown at the time to Simpson and his colleagues but should be recorded as Felix.

Page
3
divisible by three in virtually all: in the N code scanning by three disappeared. Instead, no group contained zeros

Page
4
the traffic went first to the Processing Party which served all the Japanese Naval parties. This was led by Mrs E. Parsons: see ‘History of the Japanese Naval Subtractor Systems (JN-11, JN-23, JN-25): Processing Party’, by Mrs E Parsons, PRO HW 8/152.

Page
5
‘Freebornery’ named after its head ‘Mr Freeborn’: Frederic Freeborn.

Page
6
we relied heavily on the ‘Freebornery’ to provide: for an account of the Freebornery by Mr Freeborn’s Deputy, Ronald Whelan, ‘The Use of Hollerith Equipment in Bletchley Park’, see PRO HW 25/22.

Page
7
Referring to an earlier era… Gordon Welchman wrote: see Gordon Welchman,
The Hut Six Story: Breaking the Enigma Codes
(Allen Lane, London, 1982), p. 181

Page
8
Frank Birch: Frank Birch
The Official History of British Sigint 1914–1945
(John Jackson (ed.)) (Military Press: Milton Keynes, 2007), 2: 101 (a published version of ‘The History of British Sigint 1914–1945. Vol II’, PRO HW 43/2).

Page
9
The resolution of our role came on 23 October 1944 with the signing … of an Agreement: see Ralph Erskine: ‘The 1944 Naval BRUSA Agreement and its Aftermath’,
Cryptologia,
30 (2006), 1.

Page
10
British keys proved to be: ‘History of GYP-1’, 127: NACP, RG 38, CNSG Library, Box 116, CNSG 5750/202. The Washington method was a weighting process based upon the statistical effects of the code’s scanning properties.

Page
11
would be good to know how they did it: perhaps this is in the internal history ‘GC&CS Naval Cryptanalytic Studies Volume IX: The Japanese Fleet General Purpose System II’ (see introductory note above).

David Alvarez would like to thank Ralph Erskine and Michael Smith for providing copies of certain documents from the Public Record Office.

Page
1
a staff of about sixty-five, etc.: GC&CS annual report for 1940, Diplomatic sections, 13 January 1941 (PRO HW 14/11).

Page
2
The Foreign Office received, etc.: D & R Berkeley Street, Diplomatic Section [nd] (PRO HW 3/162).

Page
3
recipient was Major Anthony Blunt: Chart, Sources of Material (as at 31 March 1944) (PRO HW 3/32).

Page
4
The Russians have claimed: David Kahn, ‘Soviet Comint in the Cold War’,
Cryptologia
, 22 (1998), 11.

Page
5
On 27 July 1942: Carl Boyd,
Hitler’s Japanese Confidant: General Oshima Hiroshi and Magic Intelligence, 1942–1945
(University Press of Kansas, Lawrence, 1993) p. 63; Kahn, ‘Soviet Comint in the Cold War’, 13.

Page
6
brief existence in the 1920s: David Alvarez,
Secret Messages: Codebreaking and American Diplomacy, 1930–1945
(University Press of Kansas, Lawrence, 2000), p. 38.

Page
7
five Japanese diplomatic ciphers: ‘History of the Signal Security Agency, vol. 2: The General Cryptanalytic Problem’, pp. 31–2 (NACP, RG 457, SRH-361).

Page
8
at least one Mexican cipher: Alvarez,
Secret Messages
, p. 54.

Page
9
low-grade consular and administrative traffic: ibid.

Page
10
neither of these systems was readable: ‘Italian Codes and Ciphers, 1939–1943’ (HCC Box 1388).

Page
11
still struggled with its target: Alvarez,
Secret Messages
, p. 61.

Page
12
was hardly better: ibid., p. 62.

Page
13
window into Japanese diplomacy had closed: ibid., p. 63.

Page
14
joint staff conference in London: Alan Harris Bath,
Tracking the Axis Enemy: The Triumph of Anglo-American Naval Intelligence
(University Press of Kansas, Lawrence, 1998), p. 25.

Page
15
exchange information on their cryptanalytic operations: Alvarez,
Secret Messages
, p. 76.

Page
16
‘please expedite reply’: quoted in Robert Louis Benson,
A History of US Communications Intelligence during World War II
(NSA, Fort Meade), p. 17.

Page
17
concerning specific foreign codes and ciphers: Alvarez,
Secret Messages
, pp. 78–9. The Akin-Friedman memorandum is undated but is generally thought to have been written on or about 1 September 1940.

Page
18
a few Red Army and Comintern systems: Robert Louis Benson and Cecil Phillips,
History of Venona
(NSA, Fort Meade, 1995), 1: 29. I am indebted to Lou Benson for arranging for the declassification of parts of this multi-volume history.

Page
19
it was the breakthrough: Stephen Budiansky,
Battle of Wits: The Complete Story of Codebreaking in World War II
(Free Press, New York, 2000), pp. 165–6.

Page
20
quashed their opposition: Alvarez,
Secret Messages
, p. 83.

Page
21
‘have a look around’: Prescott Currier oral history, NSA-OH-38-80 (National Cryptologic Museum Library, NSA, Fort Meade).

Page
22
‘almost empty’: Donald Gish, ‘A Cryptologic Analysis’,
International Journal of Intelligence and Counterintelligence
, 6(4) (1993), 387, fn 17. By January 1941, OP-20-G had recovered only 3 per cent of the values in JN-25B.

Page
23
had been working this system: Michael Smith,
The Emperor’s Codes: Bletchley Park and the Breaking of Japan’s Secret Ciphers
, (Dialogue, London, 2010) pp. 78–9.

Page
24
‘a fairly large staff: [Abraham Sinkov], ‘Report of Cryptographic Mission’ (NACP HCC Box 1296).

Page
25
committed to the Italian problem: ‘Cryptographic Codes and Ciphers: Italian’ (NACP HCC Box 1388); ‘Annual Report of the Chief Signal Officer and Miscellaneous Studies, FY 1942–1943’ (NACP HCC Box 832).

Page
26
abandoned work on these formidable ciphers: ‘Report on German Diplomatic Section’, 8 January 1941 (PRO HW 14/4).

Page
27
ciphers used by Germany in the First World War: Alvarez,
Secret Messages
, p. 130.

Page
28
‘all our technical documents’: ‘Report by Lieut. Colonel J. H. Tiltman on his visit to North America during March and April 1942’ (PRO HW 14/46). The material passed by Tiltman to the Americans is enumerated in this report.

Page
29
‘results from me’: [Solomon Kullback], ‘The British GC&CS’, 1 August 1942, 6–7, Sinkov Papers (NACP HCC Box 1413).

Page
30
had been especially fruitful: ‘Notes on Diplomatic Liaison with U.S.’, 8 March 1942 (PRO HW 14/46); ‘Report on present position of legibility of Foreign Government Cyphers’, 25 July 1942 (PRO HW 14/38).

Page
31
‘we wonder if they are actually deeply interested’: Denniston to Stevens, 22 December 1942 (PRO HW 14/62).

Page
32
direct requests for assistance: for a survey of exchanges between Berkeley Street and Arlington Hall in the period 1943–5, see ‘Governments on which there has been no substantial impediment to liaison’, Clark Files (British Liaison, 1940–1945) (NACP HCC Box 1413).

Page
33
including American policies: Cadogan to ‘C, 3 July 1944 (PRO HW 37/4).

Page
34
South American systems: for an overview of areas in which collaboration faltered, see ‘Governments on which there is a present obstacle to liaison’, Clark
Files (British Liaison, 1940–1945) (NACP HCC Box 1413).

Page
35
Russian civil traffic: Benson and Phillips,
History of Venona
, p. 31.

Page
36
‘very great indeed’: (NACP, RG 457. SRH-349), ‘Achievements of the Signal Security Agency in World War 11’, 9.

Ralph Erskine wishes to thank David Alvarez, Stephen Budiansky and Frode Weierud for supplying some of the documents used in Chapter 11.

Page
1
Even Knox found: Dillwyn Knox, undated notes on senior staff [nd c. end 1939] (PRO HW 14/1).

Page
2
‘almost as conveniently as if…’: C. H. O’D. Alexander, ‘Cryptographic History of Work on the German Naval Enigma’, 90 (PRO HW 25/1).

Page
3
were never broken: Hinsley et al.,
British Intelligence
, 2: 664.

Page
4
Poles had given GC&CS a reconstructed Enigma: on the Polish contribution to GC&CS’s Enigma work, see Appendix 30 to F. H. Hinsley et al.,
British Intelligence in the Second World War
, (HMSO, London, 1984) 3(2).

Page
5
Birch and Turing on breaking Enigma: Alexander, ‘Cryptographic History of Work on the German Naval Enigma’, 19; ‘all German codes were unbreakable’: Birch, as quoted in A. P. Mahon, ‘History of Hut Eight’, 14 (PRO HW 25/2).

Page
6
Turing solved: ‘Mathematical theory of ENIGMA machine by A. M. Turing’, 136 (PRO HW 25/3). This version of ‘Prof’s Book’ is much more legible than the poor photostat copy, ‘Turing’s Treatise on the Enigma’ (NACP HCC Nr. 964), which lacks many figures; see also Alexander, ‘Cryptographic History’, 20.

Page
7
little headway: Alexander, ‘Cryptographic History’, 24.

Page
8
7 rotors, VI and VII were recovered: Hinsley et al.,
British Intelligence
, 3(2): 957.

Page
9
Schiff 26
captures: Ralph Erskine, ‘The First Naval Enigma Decrypts of World War 11’,
Cryptologia
, 21(1) (1997), 42; Hugh Sebag-Montefiore,
Enigma: The Battle for the Codes
(Weidenfeld & Nicolson, London, 2000), p. 74.
Schiff
26 is erroneously called VP 2623 in Hinsley et al.,
British Intelligence
, 3(2): 959.

Page
10
Rotor VIII was captured: Hinsley et al.,
British Intelligence
, 3(2): 957. For full details of the numerous captures of Enigma material, see Sebag-Montefiore,
Enigma
, but no one has discovered how rotor VIII was captured.

Page
11
‘Foss’s day’ etc.: Alexander, ‘Cryptographic History’, 22.

Page
12
diagonal board: see p. 48.

Page
13
not getting ‘fair does’: Birch, 21 December 1940, as quoted in Mahon, ‘History of Hut Eight’, 29.

Page
14
Banburismus: see Alexander, ‘Cryptographic History’, chap, ix; Mahon, ‘History of Hut Eight’, 16. For an outstanding article on Banburismus, see Steven Hosgood ‘All You Ever Wanted to Know About Banburismus but were Afraid to Ask’ (http://tallyho.bc.nu/~steve/banburismus.html (accessed 5 November 2010). For a detailed contemporary account, see ‘Home Waters Enigma’ (NACP RG 38, Radio Intelligence Publications, Box 172, RIP 610).

Page
15
Enigma key-list for February: documents at NHB; on this and related captured documents, see Ralph Erskine, ‘Captured
Kriegsmarine
Enigma Documents at Bletchley Park’,
Cryptologia
, 32 (2008), 199.

Page
16
‘Germet 3’: on this cipher, see ‘German Naval Meteorological Cypher’, Met 65 (NACP HCC Box 187 Nr. 874).

Page
17
broke the DAN meteorological cipher: G. C. McVittie, diary entry for 8 February 1941, with ‘Autobiographical Sketch’ prepared for Royal Society of Edinburgh (RLEW, CCAC).

Page
18
delay of fifty hours: Hinsley et al.,
British Intelligence
, 2: 174.

Page
19
reconstructing … new bigram tables: Alexander, ‘Cryptographic History’, 31.

Page
20
a further two months: ibid.

Page
21
dummy signals … falsified: ibid., 30.

Page
22
unable to solve: ibid., 31.

Page
23
changed every two days: ibid., 5.

Page
24
only officers were permitted:
Der Schlüssel M. Allgemeine Bestimmungen
(M. Dv. Nr. 32/3), para. 130 (NHB).

Page
25
within thirty-six hours: Hinsley et al.,
British Intelligence
(1979), 1: 338.

Page
26
Werftschlüssel
, as well as Enigma: see e.g., identical
Werft
and Enigma signals in ‘German Reports of British Mining on 29/7/41’, ZG 45 of 31 July 1941 (PRO ADM 223/2). On the
Werftschlüssel
(M. Dv. Nr. 103) (NHB), see Michael van der Meulen, ‘
Werftschlüssel
: A German Navy Hand Cipher’,
Cryptologia
, 19(4) (1995), 349 and 20(1) (1996), 37.

Page
27
Shark … introduced … 5 October: MND signal 27 September 1941 (NACP microfilm T1022/2325). Alexander, ‘Cryptographic History’, 34, suggests that the change occurred on 3 October, but the German signal is very specific.

Page
28
bigram tables … changed: Alexander, ‘Cryptographic History’, 35.

Page
29
tables … captured from
Geier
: documents at NHB. The tables, but not the
Kenngruppenbuch
, had also reached Bletchley Park from
Donner
on 30 December, when they were immediately borrowed by Turing.

Page
30
M4: see Ralph Erskine and Frode Weierud, ‘Naval Enigma: M4 and its Rotors’,
Cryptologia
, 11 (1987), 235.

Page
31
M4 lid … from U-570: documents at NHB. The lid was from machine
M 3172.

Page
32
already solved the wiring: ‘Memorandum No. 3. “
Schlüssel M
(Form M 4)”’ (NACP RG 38, Radio Intelligence Publications, Box 169, RIP 403); cf. Alexander, ‘Cryptographic History’, 36.

Page
33
only three Shark keys: Alexander, ‘Cryptographic History’, 36.

Page
34
–72 referred to the Y Board: DMI, minute, 19 March 1942, ‘Brief for C.I.G.S. on 20.3.42’ (PRO WO 208/5027).

Page
35
Soaring shipping losses: see table in Appendix O in S. W. Roskill,
The War at Sea, Vol. II
(HMSO, London, 1956).

Page
36
Wetterkurzschlüssel
… seized: documents at NHB. For a description of the incident, see Sebag-Montefiore,
Enigma
, pp. 218–21.

Page
37
Hut 8 believed: McVittie, diary, 2 December 1942.

Page
38
Germet 3 additive tables … were repeated: ‘WW and other crib processes into U-Boat traffic’, nd, but between 12 December 1942 and 11 January 1943 (PRO HW 14/64).

Page
39
begin to be repeated on 8 December: memorandum of 15 December 1942 by RAF Section, ‘Report of Research and Exploitation’ (PRO HW 14/61).

Page
40
‘hotted up’: McVittie, diary, 2 December 1942.

Page
41
sent a teleprint: ZTPGU 1 (PRO DEFE 3/705).

Page
42
eighty-eight out of the following ninety-nine days: Mahon, ‘History of Hut Eight’, 77.

Page
43
pessimistically advised: as quoted in Rear-Admiral J. H. Edelsten, minute of 9 March 1943 (PRO ADM 205/29).

Page
44
could not build: Alexander, ‘Cryptographic History’, 50.

Page
45
out of the 112: ibid., 49.

Page
46
stretched Allied HF-DF: on shore HF-DF, see Ralph Erskine, ‘Shore High-Frequency Direction-Finding in the Battle of the Atlantic: An Undervalued Intelligence Asset’,
The Journal of Intelligence History
, 4(2) (2004), 1.

Page
47
proved invaluable: Alexander, ‘Crytographic History’ 50.

Page
48
serious delays: see e.g. PRO ADM 223/OIC SI 641, 648 (weeks ending 19 and 26 July, when Shark had been partially read up to 19 June and 16 July, respectively); also OP-20-GM war diary (NACP RG 38, Crane Library, 5750/176), entries in July and August.

Page
49
British and US Navy … bombes: as to June, Hinsley et al.,
British Intelligence
, 2: 751; as to August, ICY 51, 82 (NACP HCC Box 705, Nr. 1736). An early British four-rotor bombe had been available since April 1943, but for some reason (perhaps a lack of cribs) did not give any results against Shark until June: Alexander, ‘Cryptographic History’, 43.

Page
50
forty-five hours: OP-20-GY-A-1, memorandum, 3 November 1944
(NACP RG 38, Crane Library, 5750/205).

Page
51
Feynman adopted: Richard P. Feynman,
Surely You’re Joking, Mr. Feynman
(Unwin Hyman, London, 1985, pbk), p. 128.

Page
52
score over 100 tetras: Alexander, ‘Cryptographic History’, 108.

Page
53
Decibans played a vital part: they were also used in breaking Tunny, the Lorenz SZ 40/42 teleprinter cipher attachment.

Page
54
‘to succeed where it would otherwise have failed’: Alexander, ‘Cryptographic History’, 108.

Page
55
Initially, Hut 8 consisted of: this section is based on ibid., 86–92.

Page
56
‘… veil of considerable opacity’: ibid., 38.

Page
57
visitor from Hut 6 pointed out: ibid., 39.

Page
58
Süd: Kenngruppenverfahren Süd Januar
1944 (M. Dv. Nr. 608) (PRO ADM 223/331).

Page
59
Alan Turing had written: ‘Mathematical theory of ENIGMA machine by A. M. Turing’, 130.

Page
60
Sunfish … was broken: ‘German Supply Ship [‘
Versorgungsschiff
’] Traffic’, ZG 240, 19 August 1943 (ADM 223/4).

Page
61
broke Seahorse: ‘War diary of
SCHIFF 28
, Mar–June’, ZG 244 of 15 September 1943 (PRO ADM 223/4); on breaking Seahorse and the other throw-on ciphers, see Ralph Erskine and Philip Marks, ‘Naval Enigma: Seahorse and other
Kriegsmarine
cipher blunders’,
Cryptologia
, 28 (2004) 211.

Page
62
broken, ‘[a]fter a certain amount of trouble’: Alexander, ‘Cryptographic History’, 71.

Page
63
Kriegsmarine … fully informed: see
Deckblätter Nr. 1-8 ‘Schlüsselanleitung zür Schlüsselmaschine Enigma vom 13.1.40’; cf. Der Schlüssel M Verfahren M Allgemein
(M. Dv. Nr. 32/1 – August 1940 edition), para. 140, referring to single encipherment for
Heer
Enigma.

Page
64
about 460 intercepts each day …: Mahon, ‘History of Hut Eight’, 7.

Page
65
by about one year: Alexander, ‘Cryptographic History’, 9 (although not mentioning HF-DF).

Page
66
twenty-three … stations: ‘The Communication Intelligence Organizations of the British Empire’, 73–6 (NACP RG 38, Radio Intelligence Publications, Box 65, RIP 99).

Page
67
Tina: see ‘Tina, 1946’ (NACP HCC Box CBCB 43 Nr. 901), ‘Tina, 1941’ (ibid., Nr. 902), ‘Tina, 1942–1943’ (ibid., Nr. 903); ‘Rep/Tina Historical Documentation’ (ibid. Box CBKH 68 Nr. 1540); on RFP, see ‘Report on R.F.P. Identifications’, 15 November 1942 (PRO HW 14/58); ‘Sample Oscilligrams – RFP Groups, 1944’ (NACP HCC Box 585 Nr. 1451); ‘RFP Manuals’ (ibid., Box 806 Nr. 2330).

Page
68
seldom do more than distinguish: ‘Report on R.F.P.’, 19 November 1944
(PRO HW 18/89).

Page
69
Ultra: ‘Ultra’ was a short title for ‘special intelligence’: signal of 20 August 1944 (quoted in full in G. E. Colpoys, ‘Admiralty Use of Special Intelligence in Naval Operations’, 43 (PRO ADM 223/88)).

Page
70
all attempts … should be abandoned: minute of 19 November 1944, ‘Summary of R.F.P. Classification of Naval Units during the German War’ (PRO HW 18/89).

Page
71
US Navy RFP gave worse results: minute of 9 March 1944, ‘R.F.P. and Tina Effort on German U-boats’, 9 March 1944, para. 8 (PRO HW 18/89).

Page
72
Mrs Agnes Driscoll: ‘Naval Security Group History to World War II’, 400 (NACP RG 457, SRH 355).

Page
73
exasperated by GC&CS’s failure: Cdr. J.N. Wenger, Cdr. H. T. Engstrom, Lt-Cdr. R. I. Meader, memorandum of 30 May 1944, ‘History of the Bombe Project’, para. 5 (NACP HCC, Box 1414, Nr. 4584).

Page
74
an extensive bombe programme: J. N. Wenger, memorandum of 3 September 1942, endorsed as approved by Admiral Home, the Vice-Chief of Naval Operations (NACP RG 38, Crane Library, 5750/441).

Page
75
Desch … had completed: memo of present plans for an electromechanical bombe, 17 September 1942 (NACP RG 38, Crane Library, 5750/441). The report can be downloaded from http://cryptocellar.web.cern.ch/
cryptocellar/USBombe/desch.pdf.

Page
76
Holden Agreement: PRO HW 3/193, f. 69; on the Agreement, see Ralph Erskine, ‘The Holden Agreement on Naval Sigint: The First BRUSA?’,
Intelligence and National Security
, 14(2) (1999), 187.

Page
77
British official history errs: see Hinsley et al.,
British Intelligence
, 2: 57.

Page
78
Adam … delivered: serial 003 of 27 May 1943 (NACP RG 38, Crane Library, 5830/116).

Page
79
Neither machine was running well: see e.g., serials 852 of 3 July 1943 (‘Adam out of operation’), 439 of 31 July 1943 (‘Eve still out’) (ibid.); cf. OP-20-GM war diary, 3 June 1943.