Page 311 - The Secret War in the Italian front in WWI (1915-1918)

CHAPTER THIRTEEN

be entirely replaced by over-encoding thanks to tables different from those shown in the previous
picture.
The successes achieved by the Italians in breaking this new type of codes will be described in the
next Chapter.
Meanwhile, also the design of codes for small radio stations - previously left to the initiative of
individual divisions or lower-level units - was also regulated, by the Central Cryptologic Unit of
the Evidenzbureau who cross-checked every new system proposed by the units or by the Military
Telegraphs Command, as was the case among the Italians .
47

coMMenTs on The TransITIon froM cIPhers To codes

The need to frequently replace the ciphers required all the belligerent Armies to research systems
that imposed a high toll of effort on the enemy before it could break them, without at the same
time making the encoding and decoding work any tougher.
In the evolution of encoding methods, one can identify similar trends among the most important
Armies in the field. French cryptologist Giviérge recognized this phenomenon and wondered
whether it could come from “ideas that are in the air” or more likely from information contained
in documents mutually seized during the fighting .
48
Cryptanalysts who tried to penetrate enemy secrets may have contributed to the trend mentioned
above by employing the achieved knowledge also to improve their own systems. This process was
eventually helped by the concentration in a single organisation of code designers and decrypting
specialists, based on the widespread belief that “only experienced cryptanalysts can judge the
security of a cryptographic system,” as the experience of the World War I had shown so far.
49
Giviérge himself and then Kahn observed a tendency common to many armies, between the second
half of 1917 and the beginning of 1918, to abandon albeit not entirely the transposition techniques
and their numerous variations, adopted since the early years of war for radio station service and
for communication of subordinate units. These ciphers were replaced by short regular codes with
over-encoding or by two-part codes with - and more often without - over-encoding .
50
Sacco also recalled this trend as well, asserting with reference to all combat fronts: “near the end
of the war, in the large land units, two-part codes of 1,000-4,000 code groups were adopted. Of
these, books without over-encoding were changed every two or three weeks, while over-coded
ones every two or three months” .
51
As shown above, a similar transition from ciphers to codes occurred, at different times, also in
the Italian and Austrian armies. Moreover, it may be noticed that the already illustrated solution
to protect those codes with various drawing by lots methods, adopted by Sacco since September
1916, took place also within German and Austrian armies in the course of 1918.

47 J. Prikowitsch, op. cit., p 431.
48 Marcel Giviérge, Questions de Chiffre, op.cit., N° 33, March 1926, N° 34, May 1916, translated in English by the USA
War Department and published in the Signal Corp Bulletin (Cryptography and Cryptanalysis reprinted from the Signal Corp
Bulletin, p. 15 -16).
49 F.L. Bauer, op. cit. p. 207, Maxim N°2 of Cryptology.
50 M. Giviérge, Questions de Chiffre, op.cit., p. 15; D. Kahn, The Codebreakers, op. cit., p. 314 -315.
51 L. Sacco, Manuale, op. cit., p.285.

309