Abstract
This chapter gives a brief account of the early development of telegraphic and radio communication systems and of the use of message interception techniques to obtain intelligence of diplomatic or operational value. Thus modern Signals Intelligence was born.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
In this book, code groups are groups of 3, 4 or 5 digits only, or of capital letters only.
- 2.
Kerckhoffs’ paper is available at www.petitcolas.net/fabien/kerckhoffs in both the French original and in English translation. See the end of Part II. The French original of the free translation given in italics in the main text is ‘Ce ne sera que lorsque nos officiers auront étudié les principes de la cryptographie et appris l’art de déchiffrer’. Section 8.19 gives various other maxims for cryptographers operating with WW2 technology.
Kerckhoffs’ comment may be compared with the extract from a British report of 12 March 1942 quoted at the end of Chap. 3.
- 3.
In WW1 and WW2 the Minister responsible for the Royal Navy was called the First Lord of the Admiralty while the senior admiral was called the First Sea Lord. A few other senior admirals would also be designated as Sea Lords.
- 4.
One interesting example of breaking a change of key on a code whose book had been captured was described by Dr. Frederick Wheatley 19 years after his 1914 achievement. The text is preserved by the Australian War Memorial (AWM) as archival item AWM252 A228. He was able to decrypt, decode and translate German naval messages. The difficulty in recovering (working out) a decrypted enemy code book is discussed in Chap. 11.
In the context of additive cipher systems, which were extensively used by the Japanese in WW2, the word ‘keys’ was sometimes used for what are generally called ‘additive tables’ in this book.
- 5.
Mavis Batey explains in her book Dilly: The Man who Broke Enigmas (Dialogue, London, 2009) that this use of ‘hatted’ originated in the practice of mixing up slips of paper in a hat and then drawing out one to obtain a random choice. Part of the story of the Battle of Midway is that the JN-25B code words for geographical locations were not hatted!
Such codes require two code books in much the same way as, say, a German-English dictionary is complemented by an English-German dictionary.
- 6.
The reference is TNA ADM 223/773 Memo on ‘Political’ Branch of Room 40, by George Young. It is cited in Paul Gannon’s Inside Room 40: The Codebreakers of World War 1, Ian Allen 2010.
- 7.
Reinforcements from the USA were of great significance in the Western Front in Europe in 1917–1918. This was to happen again in 1942–1945.
- 8.
Having symbolism for 26 letters and 10 digits is useful because military messages commonly include numerical information best communicated by digits. Details on how ADFGVX was broken are available in various places. For example, the invaluable series of books produced by Aegean Park Press includes two accounts of the matter. Of course the Google search engine has no trouble finding useful material about ADFGVX on the web. It was not fully broken in WW1.
The book The British Army and Signals Intelligence during the First World War edited by John Ferris (Alan Sutton Publishing for Army Records Society) gives much more information on WW1 Army signals intelligence. In particular it sets out the WW1 use of traffic analysis and describes an interesting technique, based on electric current leakage through the ground, for intercepting front-line telephone communications.
Ferris has recently reviewed in great depth the paper by C. J. Jenner Turning the Hinge of Fate … 1940–1942 in Diplomatic History 32(2) (April 2008). The reference is H-Diplo Article Reviews number 199. Jenner’s paper deals with the situation in North Africa at that time and so is not relevant to this book. However Ferris very usefully explains that the importance of signals intelligence is often exaggerated. The three episodes given pride of place in this book—the Battle of Midway, the USN submarine attacks on Japanese merchant ships and the exploitation of the material captured in Sio, New Guinea, in January 1944 in the ‘Leap to Hollandia’—are exceptions to Ferris’ general comments. Richard Overy’s book Why the Allies Won, Cape, 1995 is useful in assessing the overall importance of signals intelligence in WW2.
- 9.
A useful description of training is given by A. Jack Brown in Katakana Man (Air Power Development Centre, Tuggeranong, 2006), pages 13–18, while its chapters V & VI give a graphic account of front-line interception work on Leyte in late 1944. Such work was generally carried out under difficulties and deserves explicit recognition. Lt.-Col. Ryan, head of the Australian Army interception unit in WW2, wrote on page 3 of his final report (National Archives of Australia (NAA) Canberra item A10908 2):
‘It must be realised that the work of an intercept operator is particularly trying and that efficiency is inevitably governed by such factors as lighting, ventilation, insulation against heat and cold, facilities for proper rest and recreation.’
On page 9, Ryan states that the backs of trucks do not form good workplaces.
Much the same thoughts can be found in NAA Canberra A11093 320/5K5 Part 2: ‘Y Intelligence operators are obliged to listen for five consecutive hours to very faint signals which have superimposed on them other signals and a whole lot of interference.’ The process imposed great strain. Rotation and leave were needed.
Part J of the Central Bureau Technical Records (CBTR) (NAA Canberra item B5436 Part J and accessible on line via the Recordsearch facility) is entitled Field Sections and deals with the RAAF Wireless Units, the Australian Army sections led by Lt.Col. Ryan and the American Signal Radio Intelligence Companies. It specifies the duties of these units as being:
-
1.
Provision of air-raid warning intelligence.
-
2.
Interception of low-echelon Army traffic and provision of tactical intelligence derived therefrom.
-
3.
Interception of raw material, inaudible in rear areas, for cryptographic solution by CBB.
-
4.
Local exploitation by traffic analysis in Kana for intelligence value.
The Allied decryption centres Central Bureau and Frumel, to be introduced later, had interception facilities within short distance of their bases in Brisbane and Melbourne respectively.
The only detailed account of equipment used in WW2 interception stations known to the authors is that by Thomas J. Gray entitled Some Aspects of the ‘Y’ Service—India 1943–1945. This appeared on pages 95–102 of issue 18(2) (1988) of the Journal of the Royal Signals Institution.
The NAA Canberra file A2617 Section 43/14609 Harman: Additions to ‘Y’ Hut has an interesting diagram of the interception building at the Harman Naval Wireless Station near Canberra.
The CBTR fails to mention here the 1st Canadian Special Wireless Group (1CSWG), which operated at Adelaide River south of Darwin in 1945. A short film The Canucks in Aussie was shot by members of the First Canadian Special Wireless Group in August 1945 and shows something of their work at Adelaide River. A copy survives in the Canadian War Museum. It is mentioned in NAA Canberra item A6923 16/6/502.
Kenneth Macksey’s The Searchers: Radio Intercept in Two World Wars provides much information on WW2 interception services, particularly in Britain.
Of course, the other side, transmitting messages, also had problems. Its signal staff would be carrying out repetitive dull work in difficult conditions. Canberra NAA item A705 109/3/817 confirms this. So one cause of the emergence of insecure signalling practices is evident. See also page 6 of the CBTR, part A.
-
1.
- 10.
See Joe Richard The Breaking of the Japanese Army’s Codes in Cryptologia 28(4), October 2004, 289–308. Jack Brown (see immediately above) comments on this matter on page 16. In fact, the IJN used standard Morse for transmitting its main operational series of codes known as JN-25.
A navy adopting additive cipher systems for the first time could have gained extra security by adopting an unusual patternless allocation of dots and dashes to the ten digits. It would be necessary to prevent trainees from sending practice messages over the air waves. If some such person transmitted 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 then anyone intercepting the signals would pick up how the digits are represented. Compare the second maxim of Sect. 8.19.
- 11.
The Bletchley Park terminology was ‘dot’ and ‘cross’, with the former corresponding to the binary digit 0 and the latter to 1. Thus ‘cross’ + ‘cross’ = ‘dot’, etc.
- 12.
In fact the first quintuple, 0-0-0-0-0, was not used in either letter mode or shift mode. This would have been convenient in separating different (unenciphered) messages punched out on the same piece of paper tape. The code set out in the main text is not the original version invented by Baudot but rather a more recent modification. (Some use has been made here of the Wikipedia free encyclopaedia and in particular of the entries for Gilbert Vernam, XOR, stream cipher and one-time pad.)
- 13.
A good comprehensive account of ‘secret writing’ over the centuries is given by David Kahn, The Codebreakers, Weidenfeld and Nicholson, London, revised second edition 1996.
- 14.
For example, the British interception stations at Hong Kong and Singapore in the 1930s had great difficulty picking up signals emanating from the Japanese Mandated Islands in the Pacific.
- 15.
Encryption and regular changes of call signs could only enhance security: this was done to some extent by the Germans in the European war.
- 16.
The importance of traffic analysis as a source of signals intelligence was recognised in WW1. The very distinguished American cryptanalyst William Friedman (see Chap. 4) served as an officer in the US Army intelligence unit on the Western front late in WW1. He subsequently wrote ‘Its utility in deriving intelligence about enemy intentions from a mere study of the ebb and flow of enemy traffic, without having to solve the traffic, was of unquestionable value’. (See The Friedman Legacy: A Tribute to William and Elizebeth Friedman, NSA Sources in Cryptologic History, Number 3, page 109.)
As an example of its use in WW2, Hut 6 at Bletchley Park had a TA unit called Hut Sixta. James Thirsk’s book Bletchley Park: An Inmate’s Story (Galago, 2008) explains the development of the Fusion Room which merged information from Hut Six and Hut Sixta. So-called ‘operator chat’ was also useful. Another such book is Joss Pearson’s Cribs for Victory (2011), which is an edited and augmented version of a text written earlier by her father, Neil Webster.
- 17.
In the paper mentioned in Note 9, Gray describes the large map used in a dedicated room at Bangalore to plot ‘fixes’ from DF stations.
Some account of the various projections used for constructing on flat paper maps of portions of the surface of the planet may be found in Wikipedia and other appropriate websites. Of these the most familiar is the standard Mercator projection, which has the undoubted merit of representing meridians (great circles through the poles) by vertical lines and parallels (small circles parallel to the equator) by horizontal lines. It has the much more subtle property of being conformal, that is representing angles correctly. But it distorts the regions near either pole, misrepresenting the areas of Canada and other such countries. And, for the purpose of direction finding, it represents great circles by excessively complicated curves. Hence it is quite inappropriate for locating the source of a radio transmission from the directions observed at two or more stations.
The gnomonic (also called gnomic) projection is constructed by projecting from the centre of the planet on to the tangent plane at a fixed point on its surface. This ties in better with conventional latitude and longitude if the fixed point is taken to be on the equator. For the Pacific one may as well take the point 0∘N, 180∘W or perhaps 0∘N, 150∘E. This projection, which can handle only one half of the planet at once—scarcely a problem—represents the meridians by vertical lines and the parallels by arcs of hyperbolas. It has the practical merit of representing great circles by straight lines.
In practice a gnomic chart would have just selected meridians and parallels (perhaps every second degree of arc) marked, together with dots for the direction finding stations. The coastlines of islands and continents would be harder to draw and not particularly useful. The approximate location of a radio source would be identified on the gnomic chart, its latitude and longitude read off, and the work would then be transferred to a conventional Mercator map.
- 18.
Gannon, Inside Room 40, page 168.
- 19.
One suitable reference is Chap. 3 of the book by Frederick Brooks and Kenneth Iverson: Automatic Data Processing, John Wiley and Sons, New York, 1963.
- 20.
An interesting letter on this matter may be found in General Blamey’s papers at the Australian War Memorial. Copies of some of Blamey’s Sigint material are to be found in RG457 in NARA. See also Chap. 19, notes 9 and 20.
- 21.
Stephen Budiansky has written a useful paper Codebreaking with IBM Machines in World War II, which is published in Cryptologia, 25(4), October 2001, 241–255.
- 22.
The GCCS had to find secure methods of disposing of 2 million used cards per week from Bletchley Park! Secure acquisition was also a problem. In a report on a visit to Bletchley Park, Friedman praised the productivity of the machine unit.
- 23.
Although electricity moved at the speed of light, the cards did not. They were transported by rollers and other devices that were continually stopping and starting. The IBM company had developed methods of doing this at commendable speed. Rewiring an existing device was relatively easy but getting totally new card processing technology under way would have taken more time than was available.
Note 46 of Chap. 9 is not easy reading but it gives a rare contemporary description of one of these special purpose electro-mechanical machines.
- 24.
For example, section 20 of Whelan’s report is headed Positional Double Repeat Search. It describes the processes used ‘to find messages in common with others by virtue of each having a pair of cypher groups (GATs) repeated in another and where the left to right separation of one of these groups from the other was the same in a given pair of messages.’ The American jargon was ‘double hits’ rather than ‘double repeats’. As these ‘cypher groups’ were of five digits, this is almost certainly about processing JN-25 intercepts. In the next section Whelan deals with the mechanical production of difference tables, of which more later.
Edward Simpson’s Bayes at Bletchley Park (Significance, 7(2) June 2010 76–80) confirms that tabulating equipment was used at Bletchley Park in processing JN-25 intercepts.
- 25.
The Government Code and Cipher School (GCCS) was renamed the Government Communications Headquarters (GCHQ) soon after the end of WW2 and is still called GCHQ.
- 26.
NARA RG457 Box 1126.
- 27.
Mackenzie King kept a very extensive typed diary over many years. This is now on the website of Library and Archives Canada. The entry of 26 May 1942 includes the following text: ‘I opened one marked “For immediate delivery: most secret” from the Naval department. The message was from the US liaison officer of the Pacific Command. It indicates that the US Intelligence believes there will be an attack against Midway, Hawaii and the Aleutians probably during the period 20 May to 20 June.’ Mackenzie King was very concerned about any possible Japanese attack on Alaska. He discussed the matter with his War Cabinet.
- 28.
A memorable device used in the European war was to use Spitfires without guns to fly at great heights taking photographs of the ground below. As otherwise the camera lenses would be covered in condensation from the cold air around the plane, photographs were taken through the exhaust stream of the engine. The aircraft used were said to be ‘cottonised’ in view of the role of the redoubtable Australian aviator Sidney Cotton in getting photographic intelligence going. In the Pacific War the distances to enemy sites were often too great for photographic intelligence to be feasible until planes from carriers were put to this use.
- 29.
Attention has to be drawn to the obituary of Andrew Gleason published in the Notices of the American Mathematical Society, 56(10) November 2009 1236–1267. Gleason worked in WW2 for Op-20-G and achieved much in decrypting Seahorse Enigma, used for communications between Berlin and the German Naval Attaché in Tokyo. This involved the exploitation of redundant encryption, which is discussed in Chap. 14.
Chris Christensen’s U.S. Naval Cryptologic Mathematicians during World War II, in Cryptologia 35(3) (2011) 267–276 is a useful general reference here.
- 30.
The original source is TNA HW 14/35 and it is cited in Paul Gannon’s Colossus, page 365.
- 31.
William Friedman (see Note 16 above and Chap. 4), is known to have had at least one breakdown caused by stress and overwork in WW2. Athanasius Treweek of Frumel, the naval Sigint base in Melbourne, gave a detailed interview on his WW2 experiences to his son-in-law Alan Roberts on 2 September 1985, with the following being recorded:
-
Treweek: ‘Well, they tried to get me to stay on, but I said the mental strain of this was so great—it affected my digestion for the rest of my life. The mental strain of doing this work was so great that I couldn’t have faced it.’
-
Roberts: ‘What was the strain?’
-
Treweek: ‘Trying to solve a problem against time—feeling if you could go on for another half hour you’d have the answer, and you’re getting more and more tired. That’s what it is. It took a great effort to say “I have to stop working on this now.” People may be dying or something as a result.’
Essentially the same point was made by Captain Arthur McCollum, USN, in a hearing of the Congressional Investigation into the Pearl Harbor Attack. The reference is volume 8, page 3400 of the transcript:
‘This type of work is one of the most trying mental exercises you can have. We have had a number of our officers and a number of our civil people break down rather badly under continual punching on this sort of thing and it is a continual concern of officers who handle these people to keep them from coming to a mental breakdown on this type of work.’
-
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Donovan, P., Mack, J. (2014). Communications and Sigint. In: Code Breaking in the Pacific. Springer, Cham. https://doi.org/10.1007/978-3-319-08278-3_1
Download citation
DOI: https://doi.org/10.1007/978-3-319-08278-3_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08277-6
Online ISBN: 978-3-319-08278-3
eBook Packages: Computer ScienceComputer Science (R0)