|
|
Oldest known cipher device, found in a home near Jefferson's |
Thomas Jefferson (yes, our second president) invented his "wheel cypher" in the 1790s and a deceptively innovative little
device it is. In fact, this machine was so far ahead of its time, it was still in active use in the US military 150 years later
at the beginning of WW2. You can see an example from my collection below (the US military version, not Jefferson's).
The Jefferson cipher had 36 wheels arranged around an axle with a different random alphabet printed on the outside of each wheel.
This is a
polyalphabetic cipher and at first pass it may seem to be similar in function to 36 Vigenère disks in series, but it is not.
The key is not a repeating keyword but the order of the 36 wheels on the axle. There are 36 X 35 X 34 X...X 2 X 1 (=36! or 36
factorial) ways to arrange 36 wheels on an axle, which is 3.72 X 10**41. Jefferson calculated this number exactly,
calling it "372 with 39 cyphers [zeros] added to it". In addition, each of the wheels has a different
substitution alphabet, unlike using the same Vigenère wheel 36 times. You could keep up a secure correspondence with many
others by having a different wheel arrangement with each person.
The sender of a cipher message using the Jefferson cipher wheel would arrange her disks in the agreed upon order and then simply spin
each disk to spell out the first 36 letters of her message, using any of the other 25 lines as the cipher. There were no numbers or
punctuation or spaces. The receiver of the
message would arrange his cipher wheels in the prescribed order and then spell out the enciphered message across one horizonal line.
By scanning across the other 25 horizonal lines, the one with the message will clearly stand out from the rest of the gibberish.
|
|
Portrait of Thomas Jefferson in 1791 |
Jefferson's cipher wheel was beyond doubt the most advanced, secure and user friendly cipher system of its time. He seems to have
invented it out of the clear blue sky, as his existing papers and letters do not show a study of the science of cryptology. In the
early 1790s, Jefferson was the first Secretary of State and used a cipher system called nomenclator, which was the cipher system in
vogue for over 450 years until around 1850. Nomenclator is a combination of cipher and codes. As President, he selected the Vigenère
cipher as the official cipher for the Lewis and Clark expedition.
Dr. Robert Patterson, a mathematics professor and President of the Philosophical Society, sent Jefferson a letter in 1802 recommending
a cipher system which is called columnar transposition, with nulls at the beginnings of the columns. Jefferson responded back to
Dr. Patterson, "I have thoroughly considered your cypher, and find it so much more convenient in practice than my wheel cypher, that I
am proposing it to the Secretary of state for use in his office." Later, he followed up in another letter, "We are introducing
your cypher into our foreign correspondences."
That Jefferson did not recognize his wheel cypher to be far superior to Patterson's cipher does not speak too highly of his cryptologic
understanding. If he would have recommended his own cipher, the US would have benefitted from a much more secure system well into
the 20th century. Instead, his cipher was buried in history until it was rediscovered among his papers in the Library of Congress in
1922.
Jefferson's wheel cipher was to be reinvented at least twice. Etienne Bazeries, a French military cryptanalyst, invented his Bazeries
cylinder in 1891 but it was never adopted by the military. Then Captain Parket Hitt of the US Army invented it
in strip form in 1913. The strip form was made into a cipher called the M-138A. In 1915, Major Joseph Mauborgne redesigned it into
the 25 wheels of the M-94, which became the main battlefield cipher for the US military until 1942.
Mauborgne would go on to cryptologic fame with several breakthroughs to his credit. The first involved the Playfair cipher, used
by the British as their field cipher. It was the first digraphic cipher, meaning that the cipher was based on pairs of letters in
plaintext being enciphered into another pair. Two pairs of plaintext letters with letters in common could produce ciphered text with
no letter in common. Mauborgne had the first recorded solution to this cipher in 1914, authoring the first work on cryptology
published by the US government. He developed the first theoretically unbreakable cipher, the one-time pad. Later, as a two-star
general and Chief Signal Officer, he built and directed the team, led by William Friedman, that solved the Japanese Purple cipher
in 1940. He retired a few months before Pearl Harbor.
The US M-94, except for the number of wheels, is an exact replica of Jefferson's cipher wheel. Jefferson's invention of this 120
years earlier, while being somewhat preoccupied with the founding of a new country, is testament to his extraordinary genius. The
concept of a rotor device with interchangeable wheels was the precursor to the various rotor-based cipher machines, such as the
Enigma and Hagelin machines, which were developed in the early 1900s.
Pictures from My Cipher Collection
