Cryptology Idea and History of Cylinder Ciphers

a7Hzq .#5r< kÜ\as TâÆK$ ûj(Ö2 ñw%h: Úk{4R f~`z8 ¤˜Æ+Ô „&¢Dø

Motivation

• Cipher cylinders provide polyalphabetic substitution with independent alphabets combined with easy, error-tolerant handling.
• Cipher disks (as well as equivalent slides) are handy but provide only the »secondary alphabets« of the primary alphabet.
• Tables are unhandy in production and require elevated concentration during use.

The key of a cylinder cipher is the order of the disks. Therefore the key space consists of all permutations of the disks, sometimes even of a choice of disks from a larger collection.

This makes the key space quite large.

The single alphabets on the disks should not be considered as part of the key.

• The enemy might get hold of them by capturing a device.
• The alphabets are not easily changed, because this means producing and distributing new sets of disks.

Historical Data

Cipher cylinders are roughly an invention of the 19th Century. They were in use in 20th Century by the military of many countries up to World War II.

Forgotten precursors:

• Combination locks with numbers (such as for a contemporal dispatch case or suitcase) or letters or other symbols were in use in Europe from around 1400 on, in China even before, and very popular in the 18^th Century. They might even be the origin of the idea of a »key« in cryptography.
• Francis BACON (1561–1626)? The NSA Cryptologic Museum exhibits a cipher cylinder, and the accompanying text contains the passage: »Similar systems have been described by writers, including Francis Bacon in 1605 …«
• Fredrik GRIEPENSTIERNA (1728–1804) in 1786 constructed a cipher cylinder with 57 disks for King Gustav III of Sweden. (This works somewhat different from the systems we considered here. In particular its settings could not be changed, or in other words, it used a fixed key. The picture was taken from the replica at HNF.) Griepenstierna probably had the idea from his grandfather Christopher POLHEM, an engineer who constructed clocks and similar devices.
• Thomas JEFFERSON (1743–1826)—from 1801 to 1809 third president of the United States—in 1795 invented a device with 36 disks, that apparently never was used. It used wooden disks. JEFFERSON's description was detected only in 1922.
• Charles BABBAGE gave a description in a 1854 Journal article.

The idea was taken up again after KASISKI had shown how easily periodic polyalphabetic ciphers with short periods can be broken, and KERCKHOFFS had demonstrated the vulnerability of dependent alphabets.

More facts:

• Étienne BAZERIES (1846–1931) in 1891 constructed a device with 20 disks and proposed it for use in the French Army. Because de VIARIS broke it shortly after, it was rejected.
• Arthur HERMANN in 1893 proposed the use of (equivalent) strip ciphers. The US Army used them from World War I on.
• In 1922 the US Army introduced the device M-94 with 25 disks, following a proposal by FRIEDMAN. This was used in several enhanced variants up to the end of World War II (for example the M-138-A used 30 strips from a collection of 100). The Germans regularly broke it (ROHRBACH).

For Further Reading

• Cipher Clerk (Simulation of cylinder ciphers and others, by Wilhelm Plotz).
• Louis Kruh: The Genesis of the Jefferson/Bazeries Cipher Device. Cryptologia 5 (1981), 193–208.