When a new message arrived, Dot looked for stereotypes, which were words that occurred frequently in the same place. “Maru” was a common one, but there were others as well, depending on the origin and the goods being transported. For example, one station transmitting from Singapore—the #3 Sen San Yusoo—sent a regular report on the shipping of oil to Hiroshima, Manila, and Tokyo. Stereotyped words might include ship names and numbers; the number of kiloliters of light oil, crude oil, heavy oil, aviation gasoline, or other gasoline aboard; how many trips each ship would take, and when. Another Singapore station transmitted to Hiroshima, Tokyo, and Moji a report of ships leaving for Palembang. Stereotypes might include the ship number or name, the date and hour of departure, the speed, the course, and the date and hour of scheduled arrival at the mouth of the Musi River.
Another transmitted a daily weather report with data including wind velocity and direction, temperature, and condition of the surface of the Andaman Sea, the South China Sea, the Yellow Sea, and other faraway bodies of water. Dot handled a lot of weather reports. Sitting at her table in Arlington, Virginia, Dot was amused at how many bits and pieces of information she knew about what the weather was like eight thousand miles away.
Another station transmitted a report on small boats available for supply services. Mentioned might be steel barges, wooden barges, special boats, small boats, twenty-metric-ton boats, plywood barges, and cargo submarines.
A station at Surabaya originated a report on the departure of ships escorted by a single Navy plane, including the names and types of ships (motor, sail, fishing), number of barges being towed, date of departure and destination, speed, scheduled date of arrival, route, and daily position of ship on consecutive days at given hours. A report from Shanghai about a supply ship might include a message spelling out that the probable route was “from Shanghai along the coast to the Yangtze River up the Yangtze River to WUU (Buko) down the river to Nanking and finally across the East China Sea to Moji.”
Here is how Dot did her work: Let’s say she knew that the code group for “arriving” was 6286 and she knew where this word was likely to appear. She would find that place in the message and look at the GAT before her. Books at Arlington Hall listed common code words as well as possible enciphered versions. She would look for a match, or she could do the math in her head and strip out the additive herself. Sometimes—when they were desperate—the code breakers would take the code groups and encipher them with every possible additive. A smattering of 2468 code groups included:
4333 hassoo—to send things
4362 jinin—personnel
4400 kaisi—beginning, commencing
4277 kookoo—navigate, to sail
4237 toochaku yotei—scheduled to arrive
4273 hatsu yotei—scheduled to leave
There were vocabulary words associated with sailing schedules. According to training materials compiled at Arlington Hall, atesaki was “destination” or “address”; chaku was “arriving”; dai ichi was “first”; honjitsu was “today.” Maru was “commercial ship”; sempakutu was “ship”; sempakutai was “convoy unit”; teihaku was “anchoring”; yori was “from”; yotei was “schedule.” Gunkan was “warship.” Chu was “now.” Hatsusen was “ship leaving.” Hi was “day”; hongetsu was “this month”; senghu was “onboard ship”; shuzensen was “ship being repaired”; tosai sen was “ship loading.”
Dot’s workday consisted of messages that, once deciphered, said things like “PALAU DENDAI/ 2/ 43/ T.B./ TRANSPORT/ 918/ (/878/)/ 20th/ 18/ JI/ CHAKU/ ATESAKI/ DAVAO/ SEMPAKUTAI/ 4/ CEBU/ E.T./”
If it sounds hard and exhausting, it was. The administrators at Arlington Hall concluded that Section B-II—the Japanese Army section—had the most complex mission in the place. This was owing to the intricacy of the Japanese Army’s cryptologic system and periodic major changes in how the systems worked.
At first, the 2468 system was enciphered by false math, but in February 1944, a few months after both Dot and Crow arrived, the Japanese began using squares. As the Japanese coped with their changing island situation, the ex-schoolteachers had to cope with Japanese cryptanalytic changes. During 1944, there were thirty thousand water-transport-code messages received each month. This meant breaking a thousand messages a day. In August 1944, the Japanese began using a new additive, new codebooks, a new square, and new indicator patterns. Staying abreast of these was Crow’s department. Crow with her math skills had been assigned to the “research unit,” which did the ongoing analysis that enabled Dot to do the active processing work. Dot didn’t know that, nor did Crow.
The schoolteachers working 2468 got a bit of specialized training. The course was developed by an all-female committee—Evelyn Akeley from Skidmore, Alice Beardwood from Bryn Mawr, Elizabeth Hudson, Juanita Schroeder, Mildred Lawrence, and Olivia Fulghum. The instructors were likewise female. Among them were Lois Harer, Lenore Franklin, Margaret Ludwig, Margaret Calhoun, and Alice Goodson.
The team of women came up with a ten-day course explaining controls and indicators, decipherment using squares and charts, traffic procedures, preambles, and message analysis. Students studied the mathematical recovery of additives, the patterns of Japanese texts, the reconstruction of squares, and the recovery of indicator keys. Women learned to compare messages such as
4 Oct 1944: 8537 1129 0316 0680 1548 2933 4860 9258 4075 4062 0465
6 Feb 1945: 5960 1129 1718 6546 1548 3171 0889 9258 4075 4062 0465
6 Mar 1945: 7332 1129 1718 3115 1548 8897 7404 9258 4075 4062 0519
and to see that these messages were relayed about the same day every month and that certain code groups recurred in the same place and probably represented a stereotype word.
The workers breaking 2468 received training in basic Japanese vocabulary, focusing on words found in shipping reports. The course concentrated on kana, which are syllables that amount to a phonetic rendering. They learned that the “typical Japanese syllable consists of one consonant followed by one vowel, eg HI-RO-HI-TO. YO-KO-HA-MA. The verb is at the end. Nouns have no singular or plural.” The women were given vocational aptitude tests and rated as “clerical,” “technical,” or “analytic.” Analytic work was the most difficult, and it was the category both Dot and Crow had been chosen for.