To coordinate IBM’s dozens of war projects for the U.S., Watson created his own corporate “Department of Logistics.” This company bureau helped IBM focus its resources on some of America’s most sensitive military projects. Watson appointed IBM General Manager Nichol head of this department. Ironically, Nichol was at the same time overseeing IBM’s foreign operations in Nazi Europe. Watson explained that Nichol “is eminently fitted for this important work by reason of his broad executive experience… combined with military knowledge.” He added, “He has intimate knowledge of foreign industrial methods and resources.”39
IBM and its technology were in fact involved in the Allies’ most top-secret operations. The Enigma code crackers at Bletchley Park in England used Hollerith machines supplied by IBM’s British licensee, the British Tabulating Machine Company. Hut 7 at Bletchley Park was known as the Tabulating Machine Section. As early as January 1941, the British Tabulating Machine Company was supplying machines and punch cards not only to Bletchley Park, but to British intelligence units in Singapore and Cairo as well.40
By May 1942, IBM employees had joined America’s own cryptographic service. A key man was Steve Dunwell, who left Endicott’s Commercial Research Department to join other code breakers in Washington, D.C. The group used a gamut of punch card machines made by IBM as well as Remington Rand to decipher intercepted Axis messages. Captured enemy code books were keyed into punch cards using overlapping strings of fifty digits. The punched cards were sorted. Each deciphered word was used to attack another word until a message’s context and meaning could laboriously be established. At one point, Dunwell needed a special machine with electro-mechanical relays that could calculate at high speed the collective probability of words that might appear in a theoretical message bit. Dunwell sought permission from Watson to ask that the device be assembled at IBM. Watson granted it. Later, Watson rewarded Dunwell for his service to the nation by allowing him to spend his honeymoon in Watson’s personal suite at the IBM Country Club.41
It was an irony of the war that IBM equipment was used to encode and decode for both sides of the conflict.42
IBM was there even when the Allies landed at Normandy on June 6, 1944. Hollerith machines were continuously used by the Weather Division of the Army Air Forces to monitor and predict the tempestuous storms afflicting the English Channel. When Allied troops finally landed at Normandy, MRUs went in soon after the beachhead was secured.43
War had always been good to IBM. In America, war income was without equal. Within ninety days of Pearl Harbor, Watson was able to inform the media that IBM had secured more than $150 million in munitions and other defense contracts. Total wartime sales and rentals tripled from approximately $46 million annually in 1940 to approximately $140 million annually by 1945.44
IBM machines were not just used to wage war. They were also used to track people. Holleriths organized millions for the draft. Allied soldiers missing in action, as well as captured Axis prisoners, were cataloged by IBM systems. The location of every serviceman anywhere in the world, from George S. Patton to the most anonymous buck private, could be determined by punching a request into a Hollerith. All military payments were automated and continuously distributed by IBM tabulators. A National Roster of Scientific and Specialized Personnel was assembled to aid in the war effort.45
A central reason IBM machines could yield such extraordinary people tracking capabilities in America arose from its extensive use in the 1940 census. This census asked a number of detailed, personal questions. A national campaign of gentle persuasion was launched to convince people to provide the answers. In one radio address, First Lady Eleanor Roosevelt promoted the 1940 census as “the greatest assemblage of facts ever collected by any people about the things that affect their welfare.” She acknowledged, “Much doubt has been raised as to the propriety of some of the questions.” But, she added, they were designed to yield “facts which will provide illuminating data on problems which have become particularly pressing.”46
On Sunday, December 7, 1941, Japan attacked Pearl Harbor. Within forty-eight hours, the Bureau of the Census published its first report on Japanese Americans entitled Japanese Population of the United States, Its Territories and Possessions. The next day it published Japanese Population by Nativity and Citizenship in Selected Cities of the United States. On December 10, it released a third re port, Japanese Population in the Pacific Coast States by Sex, Nativity and Citizenship, by Counties. Using IBM applications, the Census Bureau had tracked the racial ancestry of Japanese Americans based on their responses to the 1940 census.47
Census Director J. C. Capt confirmed, “we didn’t wait for the [American] declaration of war [which was proclaimed Monday afternoon, December 8]. On Monday morning, we put our people to work on the Japanese thing.” Since only 135,430 Japanese Americans lived in the United States, the results were tabulated quickly. A single sort was necessary: race.48
Divulging specific addresses was illegal. So the Census Bureau provided information that located Japanese-American concentration within specific census tracts. Census tracts were geographic areas generally yielding 4,000 to 8,000 citizens. When necessary, the Census Bureau could provide even finer detaiclass="underline" so-called “enumeration districts,” and in some cities “census blocks.” With this information, the American government could focus its search in select communities along the West Coast—even if it did not have the exact names and addresses.49
A Census Bureau official explained to a federal commission that he was happy to provide “a detailed cross-tabulation for even the most minute areas… for which data were collected. In other words, enumeration districts and in some instances cities by blocks…. Sheets of paper from the tabulation machines were sent out to WCCA [Wartime Civil Control Administration, which was responsible for the internment]… and became the basis for the WCCA statistical activities.”50
The maps displaying Japanese population density were marked with dots, one for each ten persons. American and Dutch census bureaus simultaneously used Hollerith systems in 1943 to create racial “dot maps” as a means of organizing transfers to concentration camps. Hollerith experts dedicated to such projects seemed to work according to an established protocol on either side of the Atlantic, almost as though they had the same consultant. Lentz and his colleagues published highly detailed articles describing their registration programs. These articles appeared not only in Dutch statistical journals, but were translated into German for the journal of the German Statistical Society, and then translated into English for the journal of the American Statistical Association. The Journal of the American Statistical Association was regularly read by everyone in the American statistical community as well as the engineers and consultants at IBM.51
If locating the Japanese by census block was insufficient, the Census Bureau was willing to take the next step to deliver actual names and addresses. “We’re by law required to keep confidential information by individuals,” Census Director Capt declared at the time. He added, “But in the end, [i]f the defense authorities found 200 Japs missing and they wanted the names of the Japs in that area, I would give them further means of checking individuals.”52
By February 19, 1942, President Roosevelt could confidently sign Executive Order 9066 authorizing the internment of Japanese Americans on the West Coast. On March 22, 1942, the evacuations began in Los Angles. The U.S. Supreme Court later upheld the legality of such a measure based on ancestral grounds alone.53