In chapter 1 of The Code Book, Singh wrote that “Cryptanalysis could not be invented until a civilization had reached a sufficiently sophisticated level of scholarship in several disciplines, including mathematics, statistics, and linguistics.” It is a reasonable comment for the reasons as below.
First of all, only if a civilization has developed mathematical methods can its code makers encrypt codes by using several mathematical algorithm like transposition or substitution. Equally, only if a civilization has developed data analysis can its cryptanalysts know how to use statistics to break codes easier. For example, to collect the frequency of each letter or symbol in a long text can lead to possible correspondence of plains and ciphers. The commonsense for an amateur cryptanalysts today is that the most frequent letter or symbol in a text may link to the character E. Some combination of letters are also critical for statistic analysis like THE or ED. Thus, frequency analysis is one of the most important methods for decryption. Finally, linguistics is also vital for a civilization to invent cryptanalysis because codes and ciphers are not only in English. Many other languages could written by letters in Latin or Spanish. Let alone that some languages can be written by their own characters but their pronunciation can be expressed by letters.
Those knowledges, in the past, are rarely related to civilian so that a cryptanalyst needed training to break codes. However, in the modern educational system, those subjects are parts of the general education that every individual has the chance to learn about some critical methods of the cryptography. Even without specific training of breaking code, amateurs can use their basic knowledge of math, stats and linguistics, like frequency analysis, to find their “own ways” to solving some of the ciphers.