meet an agent. The initial message would be: Gus traveling to Camp 2
(Los Alamos) to meet probationer (agent). The code clerk would then
take the code book and find the five-digit code group for the words. In
the case of proper names or locations, the word was spelled out; the
code book had a five-digit code group for each letter.Gus traveling Camp 2 meet probationer
45211 14402 34500 14521 22305To ensure the security of the message, the clerk would then take a one-
time pad, a list of random number groups, and would add a five-digit
group to each of the code groups from the book. (Numbers were not car-
ried as in “espionage arithmetic.”) The clerk ended the message with
five-digit code groups from the one-time pad, which would show the re-
cipient what numbers had been used in enciphering the message. Thus:45217 14402 34500 14521 22305 Code book
12345 32503 13542 33454 61234 Added numbers
57552 46905 47042 47975 83549As a last step to ensure the security of the message, the numbers would
be translated into letters, using an established code (0 O, 1 I, 2 U,
3 Z, 4 T, 5 R, 6 E, 7 W, 8 A, 9 P.) The final message
to Moscow would read:RWRRU TEPOR TWOTU TWPWR AZRTP.Even if an adversary had access to the Soviet code book, the system
was secure if the code clerks used the one-time pads only once. During
the first two years of the war, however, the NKVD and GRU code clerks
went through existing one-time pads and began to use the numbered
groups repeatedly. This allowed American and British cryptographers to
see a pattern in the messages; over years of hard work, thousands of
messages were decrypted.
Two excellent sources for the layperson on the Venona process are
John Earl Haynes and Harvey Klehr, Venona: Decoding Soviet Espi-
onage in America(New Haven, Conn.: Yale University Press, 1999), and
Robert Louis Benson and Michael Werner, Venona: Soviet EspionageVENONA CODE NAMES AND ENCRYPTION• 31906-313 z07 App G.qxd 7/27/06 8:04 AM Page 319