The Ravensbruck cards Peet and others sorted were regularly couriered to a central processing bank for concentration camp labor, the Zentral Institut, on Friedrichstrasse in Berlin. Ironically, after the war, Peet settled in a quiet Berlin residential neighborhood on Friedrichstrasse. It was only after this book's research revealed the location of the processing center that Peet realized her pleasant apartment was in the very same building complex that housed the Zentral Institut.
In Dachau, the Hollerith installation was far larger than first realized. IBM and the Holocaust has documented that in most camps the Hollerith Department, or Abteilung Hollerith, was ensconced in the Arbeitseinsatz, that is, the Labor Assignment Office, generally just an ordinary barracks. But after publication I learned that at Dachau, two dozen Hollerith machines were kept in a special two-story reinforced-concrete structure known as "the Hollerith Bunker." Dachau memorial archives have discovered a photo of the fortified edifice. The archives also include a 1948 report by former Dachau survivors reconstructing various work detachments and the names of their supervisors at the camp. The report listed fifty leading slave labor battalions by the approximate number of prisoners, as well as the kapos and German officers in charge. The list includes the Political Department: 10 prisoners; Arbeitseinsatz: 10 prisoners; Sewing Department: 180 prisoners; but the Hollerith Bunker utilized 300 prisoners. The 1948 numbers are a personal estimate by the survivor group. However, by any measure, the Hollerith operation would have been one of the largest strictly administrative groups in Dachau.3
Unlike many structures, the pillbox-like Hollerith building survived the Allied bombardment of Germany, and still stands today just outside what remains of the main gate of the camp. Local police continue to refer to the concrete cube by its original name, the Hollerith Bunker.4 With so many machines needing monthly repair and ongoing maintenance, and with millions of job-specific punch cards required, Dehomag technicians were undoubtedly stationed in Dachau almost continuously.
In Buchenwald, where reinforced bunkers had not yet been built, the inherent nature of information technology allowed the Hollerith system to survive Allied bombardment. No one is sure when Buchenwald's Hollerith Department was inaugurated, but the camp's Hollerith group was dubbed Kommando 68.5
Allied bombing raids in August 1944 destroyed several buildings at Buchenwald, including the provisional Hollerith office. On September 23, Nazi bureaucrats, probably in Berlin, wrote to camp officers asking when the Hollerith files would be restored. Buchenwald's labor placement leader, SS Captain Albert Schwartz, reassured that the card files could be reproduced quickly. "[T]he Hollerith file burned completely," Schwartz answered, "… But because prisoners' cards for Buchenwald concentration camp had already been created, a copy is available in the file of the Arbeitseinsatz fuhrer [labor placement leader]. We are currently investigating just how many cards are missing… about 8,500 cards will have to be tediously reconstructed."6
Captain Schwartz added, "As labor assignment leader, I myself have the greatest interest in the reconstitution of the card file as soon as possible. In the meantime, I have found a provisional solution using data gathered solely for labor assignments, for which an auxiliary card file was created. However, this contains only number, name and occupation as well as the current labor brigade assignment. This auxiliary card file is also constantly needed in order to enter occupations and changes." To safeguard future operations, Captain Schwartz explained, "It is planned that the Hollerith card department be given a separate barracks as a workspace. However, the promised building supplies have not yet arrived."7
Nonetheless, Buchenwald's Hollerith operation continued at quite a pace, according to documents just discovered in the camp's archives by Buchenwald's tenacious historian Harry Stein. These documents reveal the almost daily fluctuations in the number of inmates working in the Hollerith unit. On August 18, 1944, just as the Allied bombing raids started, thirty-three prisoners worked in Abteilung Hollerith; their names were handwritten in a two-column roster with Prisoner Fanczak 55/55999 being the last entry. Beneath Fanczak's name, Prisoners 34/21813 and 56/42723 were listed as too sick for work.8
Five days later, on August 23, the Abteilung Hollerith workforce in creased. Scribbled tallies began with fifty prisoners, then subtracted two un avail able for work, creating a new total of forty-eight. But later three new prisoners were added to the detachment-Inmates Susic 40/44416, Muller 34/21756, and Cielecki 15/3988-these marked on scraps of departmental notes. Their last names and Hollerith prisoner numbers were also the final entries on yet another handwritten two-column roster. By August 27, the Hollerith Department's roster of fifty had been reduced by half when twenty-five prisoners from Buchenwald's notorious Little Camp, probably Jewish, had become unavailable, most likely due to illness or death. By the end of the month, a restored total of forty-eight was again reduced by illness or death in the Little Camp to a new work contingent of forty prisoners.9
On November 17, 1944, the Abteilung Hollerith at Ravensbruck concentration camp sent the Abteilung Hollerith at Buchenwald a group of Hollerith-codable paper prisoner identification forms and a Hollerith transfer list concerning five women transferred to Work Camp Taucha on November 6. The cover letter also complained that Buchenwald's Hollerith cards had not yet been received for hundreds of prisoners transferred to Ravensbruck in prior weeks. The list also mentioned 244 prisoners moved from Work Camp Torgau on October 5; plus 169 prisoners moved from Work Camp Leipzig, 128 prisoners from Work Camp Altenburg, and 64 prisoners from Work Camp Taucha-all on October 13; as well as six prisoners from Work Camp Leipzig on October 30; and finally, five from Work Camp Taucha on November 5. "We again request that these file cards be sent as quickly as possible," chastised Ravensbruck, "since they are urgently needed for new transfers."10
On January 25, 1945, the SS Economics Administration-the Oranienburg-based agency that operated all camps-dispatched a ten-page Hollerith list of slave laborers deployed in mines at Buchenwald, asking the camp to annotate each miner's current location and status.11
Prisoners servicing the Abteilung Hollerith at Buchenwald regularly came and went, but the card sorting continued almost until liberation. In fact, fourteen prisoners were still laboring in the department on February 2, 1945. By then, Prisoner Muller 34/21756, who was added less than six months earlier, had "exited." Thirteen men still remained hard at work on February 13, 1945.12 On April 11, 1945, Buchenwald was overrun by Allied troops. The clock above the concentration camp gate was permanently stopped at 3:15 p.m., the moment of liberation. By then, no more card sorting was possible.
New documents and accounts have surfaced at other concentration camp memorial archives, some of them obtained from previously unexploited records in recently opened Russian archives. For example, weekly "Change Reports" were newly discovered in Russia by Sachsenhausen archivist Winfried Meyer. These included 1944 Change Reports from Gross Rosen, Hollerith-coded 5, and Sachsenhausen, Hollerith-coded 11. These weekly lists of camp-by-camp work assignment changes were typed or handwritten on large codable paper forms, each column of the form pre printed with its corresponding Hollerith column number. Inmate Hollerith numbers were to be punched into column 22, work assignments in column 23, birth dates into column 5, gender into column 6. Other columns re corded work skills, such as carpentry, mechanics, or unskilled labor. Two narrow columns on the form confirmed card issuance upon entering the camp and again when transferring out. Each paper form featured a box at the bottom preprinted on the left mit H-Liste vgl, indicating "compared to Hollerith List," and gepruft for "punch card verified," and both were initialed by the Hollerith operator.13