How to make a codebook versatile. The example of the ASLET code

Abstract

A brief study of a Spanish military code used in Spain in the late 1940s and early 1950s based on unpublished archival sources.

Keywords:

• ASLET code
• codebook
• cryptography
• Spain
• Spanish military code
• Spanish military cryptography

Acknowledgments

We would like to thank Mr. Álvaro and Mr. Rafael Sarmiento, both now deceased, for honoring me with their friendship and invaluable help; Mr. Francisco Javier López-Brea, for his incalculable help and the documentation he very kindly shared with us; and, lastly, Ms. Catherine Dexter, for translating to English and editing this article, and the editors for their assistance.

About the authors

José Ramón Soler Fuensanta earned his undergraduate degree in science from the University of Barcelona. He received his PhD in engineering from the National Distance University (UNED). He is currently a teacher at the School of Tourism of the Island Council of Ibiza (Spain). He developed an interest in cryptography many years ago and has been an avid collector of studies on the subject ever since. His main interest is the area of Spanish cryptology, especially that of the 20th century.

Héctor Soler Bonet is currently in the fourth and final year of his Bachelor’s Degree in History at the University of Alicante. His main field of interest is Spanish military history, mainly military intelligence and the information services.

Diego Navarro Bonilla, PhD is an Associate Professor of Library and Information Science and Archival Organization. A former director of the Institute of Research on Intelligence for Security and Defense (Carlos III University of Madrid).

Notes

1 Among others, that of destroying the plaintext message once encrypted and the original ciphertext message once decrypted. Fortunately for historians, this was not always followed, and thus we have been able to obtain codes and ciphers by having decrypted the message against the encrypted original. See, for example, the encrypted letters between Ferdinand the Catholic and his father, King John II of Aragon available at the National Library of Spain with the references MSS/20211/64, MSS/20211/73 and MSS/20211/94, among others.

2 Among others, we can cite Professor Iñiguez, who broke the code of the Republican Air Force during the civil war. Another leading figure was no doubt Juan Solabre Lazcano (Soler Fuensanta, López-Brea Espiau, and Navarro Bonilla 2012), a collector of all types of telegraphic codes which he later used for his work as a cryptanalyst. An expert on diplomatic codes in the Spanish High Command during the postwar era, he broke some thirty of them from approximately twenty countries, including the Vatican, Italy, Ireland, Chile, Venezuela, Canada and Great Britain.

3 The cryptographer who created it thought about crucial details such as as making sure it could be used with just one hand, the left, to facilitate the coder’s work.

4 (Friedman 1976a, 62).

5 Rodolphe Lemoine, alias REX, although his real name was Rudolf Stahlmann, worked for the French information services before the Second World War. Famous for his participation in the French part of the cryptanalysis of Enigma, he was an expert on obtaining copies of diplomatic codes. For further information on this character, see Paillole (1985).

6 The former was an accomplishment of Donovan’s OSS and is described in detail in Persico (2002, 197–8). The latter is described by Desmond Bristow in his memoirs (Bristow and Bristow 1993, 164–5).

7 The Spanish code secretaries of the 16th century, in reality cryptanalysis experts in addition to being responsible for encrypting messages, memorized the nomenclatures they used. To get an idea of the difficulty of this, the reader can look at any of the General Ciphers of Philip II published in Devos (1950).

8 For an overview of the forms of cryptanalysis of the codes, see the cryptanalysis book by Friedman (1976c) and the Éléments by Baudouin (1939).

9 (Friedman 1976a).

10 The Perea code, in its different versions, was the code used by the Spanish Navy for encrypting its communications from the end of the 19th century to the middle of the 20th. It was tested for the first time in 1893, and its first definitive version was published in 1894 under the title “Código de señales para uso de la Marina de guerra” [Signal codes for use by the Navy]. Its creator was Antonio de Perea y Orive, who was a vice-admiral in the Navy and the Marquis of Arellano. Three additional versions were published subsequently, in 1914, 1925 and 1934. The Perea code used five digits. The first three represented the page number and the remaining two the line number. Once the text of the message had been encrypted, the groups were super-encrypted using a booklet of single-use keys that were generally valid for one month; they were referred to using an identifying code name: “Jefe,” “Mando,” “Fuerzas Sutiles,” etc. There was only a brief period, a few months at the beginning of the Spanish Civil War, during which the super-encryption operation was not carried out, because it had no utility as both sides shared the same keys.

11 Re-ciphering, also known as super-encryption, consists of carrying out a supplementary transformation on the encrypted text to hamper cryptanalysis.

12 For a more detailed view of super-encryption methods for codes, see Friedman (1976b).

13 This method, generally in the form of non-carrying addition, was widely used in the Spanish diplomatic world. It consists of adding a random number, typically specified in a second book used for this purpose, to the encryption result. For example, if a particular word were represented by the number 4578 and the random number were 5236, the result would be: $$\begin{array}{cccc}4& 5& 7& 8\\ 5& 2& 3& 6\\ 9& 7& 0& 4\end{array}$$ As we can see, 8 plus 6 and 7 plus 3 do not produce carrying to the sum of the next two digits. The decryption process would be symmetrical. We would first use non-carrying subtraction to obtain: $$\begin{array}{cccc}9& 7& 0& 4\\ 5& 2& 3& 6\\ 4& 5& 7& 8\end{array}$$ With this result, we would look up the equivalent value in the code book to obtain the cleartext.

14 Some texts on the subject include: (Devos 1950), (Galende Díaz 1995) and (Soler Fuensanta and López-Brea Espiau 2016).

15 We know from a letter to his uncle, the Emperor Ferdinand, on May 24, 1556, in which he says that he has decided to change the cipher used by Charles V to communicate with his ministers in Italy and elsewhere because he believes it to be known to his enemies.

16 France had François Viète; England had Thomas Phelippes; Holland had Felipe von Marnix; and Spain had Jerónimo González, Juan Vázquez de Zamora, Gaspar de Soto and Luis Valle de la Cerda, among others. In reality, all of the cipher secretaries of the viceroys and ambassadors were also expert cryptographers and cryptanalysts. For more information on the subject, see chapter 4 of Soler Fuensanta and López-Brea Espiau (2016).

17 For more information on the subject, see chapter 4 of Soler Fuensanta and López-Brea Espiau (2016).

18 This is a type of one-part code that was used very widely in Spain for centuries. Basically, it is a combination formed by a one-part code and a traditional cipher, generally a homophone table. The purpose of the latter was to facilitate the encryption of terms not included in the code. Theoretically this enables the use of a mixed code, in which certain words are encoded and the cipher is used for the rest of the text. One of the most famous of these was the one used in the correspondence between Mary Stuart and Babington, whose decryption literally cost Thomas Phelippes, Lord Walsingham’s cipher secretary, his head.

19 The code is kept in the Archivo General de Simancas. Patronato Real, Legajo 52, Doc. 33. The reference was obtained from Soler Fuensanta and López-Brea Espiau (2016, 84).

20 Soler Fuensanta and López-Brea Espiau (2016, 85).

21 In 1788, we see another one, much more complex and with a schema very similar to those of today, used by Miguel Gálvez Gallardo during his time as a minister plenipotentiary in Prussia and Russia (1788). This a code with some three million groups of one, two, three and four numeric digits, without re-ciphering (Soler Fuensanta and López-Brea Espiau 2016, 165).

22 Archivo General de Simancas, Cámara de Castilla, Diversos, Leg. 34, Doc. 48. The reference was obtained from (Soler Fuensanta and López-Brea Espiau 2016, 165).

23 DEI was the abbreviation for Germany, Spain, Italy in Spanish (Deutschland, España, Italia). It was a specialized trilingual trench codebook for the Navy. Around 180 copies were printed. Unfortunately, we have not been able to locate any of them.

24 On August 30, 1939, following its announcement in Official State Gazette number 243, the High Command is created with the functions of organization and preparation of the armed forces for joint action in the event of war. It is divided into three sections:

• First Section. Military

• Second Section. Economic.

• Third Section. Information.

Within this Third Section are the listening (telephone and radio), cryptography and keys services, which are grouped in the Fourth Division. A month later the Statistics Commission is created, with a somewhat vague function, which brings together remnants of the secret services and decryption groups from the Spanish Civil War. During the world war, they monitored the diplomatic services of the nations in the conflict very closely, thanks in particular to interception of the telephone lines of the main embassies in Madrid with the help of technicians from the Abwehr, the German military information organization. The keys of the legations of the US, Japan, Portugal and Vichy, among others, were deciphered (Ros Agudo 2002).

25 The members of the Italian decryption group in Madrid worked very closely with their Spanish counterparts. The two groups set up operations in twin apartments in the same building on Lagasca street in Madrid. The Spanish group occupied the second-floor apartment, with some six or seven members. Prominent among them were Fernando Solabre, who was an expert on diplomatic codes, Rafael Fariñas and Olegario Riande, with the latter moving to the Spanish embassy in Paris a couple of years later. The Italian group occupied the third-floor apartment on the left side. It was made up of five members, led first by Colonel Francesco Dragone, who died in 1940, and later by Lieutenant Colonel Damiani, who was also a member of the group. The other members included Commander Nacki, Ensign Vuolo and Lieutenant Rossi (Soler Fuensanta and López-Brea Espiau 2010).

26 The Abwehr headquarters in Spain was located in an annex to the German Embassy, at Paseo de la Castellana 4, in Madrid. It also had a series of stations spread out over different capitals and coastal areas which reported the movements of merchant ships to the central office in Madrid, to Berlin or to submarines operating in the area. The amount of information disseminated must have been very great, as there were 34 radio operators in the Madrid headquarters alone and 10 female assistants in the decryption offices (Ramírez Copeiro del Villar 1996).

27 European Axis Signal Intelligence in World War II, as revealed in the “TICOM” (Target Intelligence Committee) Investigations and other prisoner of war interrogations and captured material, principally German.

28 Lt. Col. Mettig was the second-in-command at OKW/Chi from December 1943 to the end of the war. He was the chief of Division A (Cryptography).

29 His job was to jam the emissions of the radio stations broadcasting propaganda critical of the Regime from outside the country. Among them, and perhaps the most famous, was Radio España Independiente, a communist radio station better known as “La Pirenaica,” which started broadcasting from Moscow in 1941. To do so, it had several facilities, including one in Madrid, near the airfield, and one in Málaga [Note from the head of Air Force Transmissions to Colonel Sarmiento dated 10 April 1951. Archives of the authors].

30 There are various studies on the relations between Franco’s Spain, Mussolini’s Italy and Nazi Germany. Among others, we can cite (Heiberg and Ros Agudo 2006), (Heiberg 2006) and (Ros Agudo 2002).

31 What did occur was an attempted invasion from France, through the Pyrenees, by between 4,000 and 7,000 well-equipped guerillas. The operation dubbed “Reconquest of Spain” was a failure, because the Franco Regime, faced with the threat of a possible Allied intervention through the Pyrenees, had stationed some 50,000 men on the French-Spanish border to defend it.

32 Soler Fuensanta, López-Brea Espiau and Weierud (2010).

33 The Swiss were informed in 1941 by the Polish services in exile that the security of the Enigma machine had been compromised by the French (Lastours 2001, 137). Various German citizens who had fled their country resided in Spain or were simply stationed there in connection with the Information Services. It is not known whether Goering’s AF knew how to break the commercial Enigma machine. It is not unreasonable to think that, once Germany was defeated, some warning arrived indicating that the machine was not as secure as one might hope.

34 We know of the case of Fernando Baringué Millat, who, having returned to Spain from France, after seeing the dramatic situation of Spaniards in the refugee camps in that country, was approached about entering the Cryptanalysis Section, an offer he rejected. (Telephone interview with Mr. Baringué a few months before his death.)

35 Baltasar Nicolau Bordoy was born in Felanitx, Mallorca on January 19, 1903. He enlisted in the Army on April 15, 1922. After the war broke out, Nicolau joined the Nationalist side, the winning side in Mallorca, where he was stationed. At the beginning of the war, he combined his purely military work with intelligence work in the Listening and Decryption Section of the General Staff of Palma de Mallorca, a function for which he was especially gifted. He rose to the rank of infantry captain that same year, assuming responsibility for the leadership of the encryption, listening and cryptanalysis services on the island of Mallorca. In that year alone, 1936, he managed to break eight Republican codes and eighty-four keys, earning him congratulations from General Franco. His cryptanalytical ability was exceptional: over the course of the war, he managed to discover twenty-six codes and one hundred forty-four keys. People who knew him say that he was a person with remarkable intuition in this area.

Once the war ended, and by direct order of General Franco, Nicolau stayed on, providing his services in the Balearic General Command but reporting directly to the High Command. In 1941, he was promoted to commander and moved to the High Command’s statistics service, a euphemism for the cryptanalysis section. After the war was over, he was a professor of cryptography and cryptanalysis in the courses organized by the High Command for Army, Navy and Air Force officials.

36 The Spanish Volunteer Division, more commonly known as the Blue Division, was a German division (the 250th Infantry) made up of Spanish volunteers to fight alongside the Germans in the Soviet Union. They were assigned to the North Army Group in charge of the siege of Leningrad. For more information on the subject, see Kleinfeld and Tambs (1983) or the more recent (Caballero Jurado 2019).

37 Radio discipline at the time was apparently excellent, as indicated by the Wireless Intelligence Section of Gibraltar in 1942: “the discipline is surprisingly good and talk of an official nature is practically nonexistent” (García Mostazo 2003, 215).

38 Report on the ASLET code, dated April 6, 1951. Archive of the authors.

39 In a homophone table, each element has more than one representation, and therefore each character in cleartext can be represented by several different groups (Kahn 1968, xiv).

40 The process of encryption using this key is very simple. To encode a letter, for example H, we choose any two-digit number in the key, for example 88, and append the number of the column in which it is found, in this case 5. The encrypted term will be 885. To decrypt this, we simply take the last number of the three-digit group, which will tell us the column that contains the remaining two-digit groups. Once this has been found, we take the letter corresponding to the row. In Table 1, we have indicated this in bold to make it clearer (Soler Fuensanta and López-Brea Espiau 2016, 434).

41 Soler Fuensanta and López-Brea Espiau (2016, 382-387).

42 Photographs of both are available at http://www.criptohistoria.es/utilizados-en-espana.html

43 This was contrary to the transmission rules, as Colonel José Valento of the Air Force Transmission Headquarters acknowledged in a report on the ASLET code dated April 6, 1951. Archives of the authors.

44 Various notes from Leoncio Lacaci of the High Command to Colonel Sarmiento dated April 7, 1948 and September 13, 1949. Archives of the authors.

45 Note from the printing headquarters on February 7, 1949 to Colonel Sarmiento. Archives of the authors.

46 Refers to the Balearic and Canary islands.

47 Spanish West Africa (SWA) was the name given to the group of Spanish colonies in northwest Africa. It included the territories of Ifni, the Spanish Sahara, Cabo Juby, Sidi Ifini, and the northern zone with Tétouan and Melilla.

48 Refers to Laâyoune and Tan-Tan in what was then the Spanish Sahara.

49 Refers to the Army Transmission Center. The highest-level authority in the area of transmissions in the Spanish Army of the period.

50 This is a three-letter code with some 450 groups, not in order, for liaison between air and naval forces. It is dated 1947 and has the peculiarity, from which the code derives its name, that the first letter of the cipher group is always an M.

51 High Command. Interministerial Transmission Joint Board. Combined liaison and transmission exercise between the Army, Navy and Air Force. May 1950. Archives of the authors.

52 Letter from Colonel José Vento, Head of Transmissions of the Air Force General Staff, to Antonio Sarmiento León-Troyano, Head of the Transmission Center, dated April 6, 1951. Archives of the authors.

53 Copies 001 to 300 were deposited at the Transmission Center. Copies 301 to 518 were sent to the High Command, while the copies numbered 519 to 523, were kept at the printing headquarters. Delivery Note of the headquarters of the Army Geographic Service Printing Facilities, dated September 21, 1949. Archives of the authors.

54 Internal note regarding rules for formalizing delivery of the ASLET codebooks. Dated October 1,1951. Archives of the authors.

55 Page XVI of the codebook.

56 The number of columns in the table is specified by the number of letters in the key. The number of rows is easily obtained by dividing the number of groups in the message (excluding the control endings) by the numbers of letters in the re-ciphering key. If the remainder is zero, the quotient will be equal to the number of rows. Otherwise, we must add one to the quotient. If R is the remainder and Q the quotient, there are R columns with Q + 1 elements and the rest with Q elements. Once we know the size of the key, we arrange the groups in columns, starting from the left, and, when all the groups are in place, we obtain the message by reading by rows.

57 Only messages about combat or maneuvers in some of the combined transmission exercises of the Army, Navy and Air Force were exempt from this second, re-ciphering, step. Governing Commission of the Interministerial Board of Transmissions of the High Command. May 1950. Personal archives of the authors.

58 The Variable pagination is a random alternate pagination from 00 to 49. The Special pagination is the result from subtracting 100 from the pagination number. For example, the word “adversario” (adversary) in the normal pagination will be 0223 and in the Special pagination 9823 (See Figure 11 to see how to obtain the Variable pagination). The Special pagination is the result from the non-carrying substraction 02 – 100 = 98.

59 Acknowledgement of receipt in Q code.

60 One of the cases of closer collaboration occurred in 1956, with the announcement of the return to Spain of the Spanish residents in the USSR, exiled during the Spanish Civil War, and prisoners of the Blue Division, the volunteer unit Franco sent to Russia to fight alongside the Germans during World War II. In a series of eight expeditions, 1,919 people came to Spain, of which 1,439 remained in the country and 446 returned to Russia or left for other countries. Interrogation of the returnees was led by Lieutenant Colonel Arozarena, of the High Command, with the assistance of the police and the CIA. At the time it was feared that this was an attempt to recruit people to spy for the Soviet Union, which, based on the results, was true (Bardavío, Cernuda and Jáuregui 2000, 100–14).

61 An inventory of 1975, a document released by our good friend Javier López-Brea Espiau regarding cipher material for the cryptographic section of the High Command, lists several machines with a collection or acquisition date of 1958. Among them, we can see three Hagelin M-209 and one C-443 (which really should be C-433) machines oddly labeled as “requisitioned.” In this same document, we see the acquisition date of the CX-52 as 1962.

62 Division dedicated to listening and cryptanalysis.