A surviving encrypted dispatch tells us what someone feared: an intercepted courier, a rival court or an enemy listening station. The protective method depended on how messages traveled and how quickly they needed to be read. A difficult cipher that a clerk could not use reliably might be less useful than a modest one embedded in a carefully controlled process.
Conceal the message or conceal its presence?
Classical accounts describe both substitution and hidden delivery. These address different problems. Cryptography obscures a message’s meaning; steganography attempts to hide the existence of communication itself. A concealed letter can still be readable when found, while an obvious ciphertext can remain unintelligible without its key. Historical operations often combined the two.
Caesar’s reported substitution is a compact example of the first approach. Accounts of concealed writing supply examples of the second, though ancient anecdotes deserve source criticism. A vivid story about a clever messenger is evidence for a literary account, not automatic proof that a method was widespread or routinely successful.
Courts, clerks and the growth of keys
Medieval and early modern diplomacy created sustained demand for secret correspondence. Nomenclators mixed letter substitutions with code symbols for important people and concepts. Their security depended partly on safeguarding key sheets, updating conventions and managing messengers. Large symbol lists could complicate an interceptor’s work but also made copying errors easier.
Polyalphabetic schemes responded to vulnerabilities in fixed substitution. Leon Battista Alberti’s fifteenth-century cipher disk is an important development in using changing alphabets. Later proposals by Bellaso and Vigenère explored other constructions. These systems did not replace every older method at once; practical habits and institutional resources varied.
Assume the system is known
In 1883, Auguste Kerckhoffs articulated principles for military cryptography, including the idea that a system should remain secure even if the enemy knows its operation. The secret should be concentrated in a manageable key rather than in the entire design. This distinction remains central to understanding why publicly described algorithms can protect information.
It also changes how we read earlier failures. A captured manual, compromised clerk or intercepted key sheet is not the same as a mathematical attack against every possible key. Operational security and cryptographic strength interact, but describing the specific failure matters. “The code was broken” can conceal several very different events.
Machines and organized cryptanalysis
Radio transformed interception: messages no longer needed to be physically seized from a courier. Rotor machines increased the complexity of encryption while intelligence organizations developed industrial-scale listening, indexing and analysis. Traffic patterns could reveal relationships and activity even when message content remained unreadable.
The Polish attack on Enigma in the 1930s and the later Allied work during the Second World War illustrate collaboration across mathematics, linguistics, engineering and administration. British bombe machines helped search possible Enigma settings under specified constraints. Colossus served a different cryptanalytic problem involving Lorenz traffic; it was not an Enigma-decoding machine.
The public-key turning point
Whitfield Diffie and Martin Hellman’s 1976 publication, building on a wider research setting that included Ralph Merkle, brought public-key ideas into open academic discussion. The RSA construction followed in 1977 through Ronald Rivest, Adi Shamir and Leonard Adleman. Earlier related work at Britain’s GCHQ became publicly known later.
These developments changed key distribution and authentication possibilities. They did not make shared-secret encryption obsolete. Modern protocols commonly combine public-key techniques with symmetric encryption because the techniques serve different roles. Explaining those roles is more useful than treating “public key” as a synonym for universal or effortless security.
Standards, implementations and new assumptions
The Advanced Encryption Standard, published by NIST in 2001, selected the Rijndael design for a standardized symmetric block cipher. Secure use still depends on suitable modes, key management and correct implementation. An algorithm name alone cannot establish that an application or communication channel is safe.
In 2024, NIST published its first three finalized post-quantum cryptography standards: FIPS 203, FIPS 204 and FIPS 205. These address key establishment and digital signatures under different mathematical assumptions from widely deployed older public-key systems. They are not evidence that every existing encrypted message has already become readable.
The historical lesson is therefore a disciplined one. Ask what a system promises, under which assumptions, and against which adversary. Studying Caesar or Enigma develops that habit of analysis. Protecting a current account, however, requires current service guidance and established security practices—not a homemade adaptation of a historical cipher.
Standards and institutional sources
- Bletchley Park — wartime systems, people and machinery.
- NIST FIPS 197 — the Advanced Encryption Standard.
- NIST post-quantum cryptography project — standards and current project context.