Page 126 - The Secret War in the Italian front in WWI (1915-1918)

THE SECRET WAR ON THE ITALIAN FRONT IN WWI (1915 – 1918)

In code systems, code groups made of letter or numbers replace words, syllables, or entire sentences.
For this purpose, vocabularies or code books are generated with varying dimensions, ranging from
a few dozen to ten thousand terms or more, depending on the code application.
Codes can be one-part codes or two-part codes which are also called inverted codes. One-part
codes have not only plaintext elements listed alphabetically, but also code groups in ascending
order when made of numbers or in alphabetic order when made of letters. Therefore, a single
codebook is enough for both coding and decoding. In two-part codes, code groups are chosen
randomly, requiring two codebooks: one for coding and another for decoding where the code
groups are shown in ascending or alphabetic order.
A double ciphering or overencoding can be implemented by further ciphering the code groups
obtained with the first coding process, by means of a substitution or transposition method.
The different cryptographic methods will be discussed in more detail further on, along with the
variety of versions used by the Italian and the Austrian armies during the World War I.

cryPTanalysIs MeThods
The methods, tricks, and intuitions that help understand the enemy’s encoded messages are so
numerous and diverse as to make it impossible to classify them, let alone describe them in a short
paragraph. According to the American cryptologist William Friedman, the method to decrypting
a message would consist, in brief, of identifying in succession: language, cipher and key, though
in practice this logical sequence is not systematically followed.
With the aim of identifying the language, the analysis of some statistical parameters characteristics
for each language is required, such as the occurrence of alphabet letters, bigrams (couple of letters) or
trigrams, representing specific features that are lost only with some coding methods. This frequency
analysis proves particularly useful also to solve cryptograms based on mono-alphabetic substitution.
In fact, in that case, a diagram of single letters occurrence maintains - even after the encoding process
- the same structure and it only shifts with respect to that of the original language .
29
The solution of cryptograms achieved by poly-alphabetic substitution systems relies on their
reduction to several mono-alphabetic systems which in turn are solvable by the frequency analysis.
According to this principle, since the 19 century, some decrypting procedures known as Babbage-
th
Kasinski and Kerckhoffs methods have been designed for poly-alphabetic systems of the Vigenére
type.
To break some transposition ciphers, it may be convenient to analyse the sequence of some letters,
called pilot letters: for instance, in Italian the letter Q is always followed by letter U, while in
German the letter C is followed by H in 90% of cases or by K. In some circumstances, it can be
helpful to analyse the sequences of bigrams and trigrams. Another approach known since 19
th
century to solve transposition systems is the multiple anagrams method which however needs the
availability of several cryptograms not only employing the same cipher and key, but also with the
same length.
All the methodologies mentioned above will be more thoroughly discussed further on.
Of course, the easiest and most effective method to break ciphers and, above all, codes rely on
having plaintexts corresponding to encoded messages or even the whole code book, achieved by
stealing it from the enemy or taking advantage of the various opportunities provided by Human
Intelligence.

29 The use of homophones - that is different symbols corresponding to the most frequent letters and used alternatively - allows
reducing the occurrence frequency of the latter, which changes the spectrum characteristics.

124