“Saipan was invaded in mid-1944 and, during the early stages of the battle, the Japanese captured two M-109 U.S. army cryptography documents and got the data back to Japan. These documents enabled the Japanese to read American military communications.”
“The navy ignored MacArthur and the impact of Japanese sailors looting all the crypto equipment from the American submarine Darter which was grounded on a shoal of the Filipino island of Palawan in late 1944. The Japanese destroyer that was pursuing Darter went aboard the sub after the crew abandoned it and found that all the Americans crypto codes and machines were intact. This material was taken and enabled the Japanese to read U.S. Navy encrypted traffic for the rest of the war. This was another reason for the relative success of Kamikaze aircraft to show up and attack when least expected.”
See:
Fortunately, ‘late 1944’ was too late for the Japanese to make a difference.
Huh, and I thought that this was going to be about the Code Talkers, where the US had a truly unbreakable code, and then proceeded to “improve” it to the point where it turned into a mere letter-substitution cipher, of the sort that hobbyists solve in a few minutes.
Not really. The code talkers were still talking Navajo (which was chosen because it had been established to a reasonable level of confidence that nobody in Japan or Germany knew the language) and then added substitutions (still in Navajo) as an extra layer of encryption.
The substitutions were not letter by letter but for certain code words. Instead of “submarine”, or “airplane”, or “munitions”, the code talkers would use standard Navajo words that substituted for those terms.
Except that they did use letter substitutions. Things like for D, they used the word “dog”, except then translated into Navajo. Which is a convoluted process for creating the cipher, except that the end result is still just a letter substitution cipher.