IBM and the Holocaust (85 page)

For three weeks in 1943, Robert Carmille actually worked at the CEC agency in Lyons. Of the several dozen CEC staffers he remembers seeing, at least ten were assigned to marketing because IBM continually tried to increase its share of France's Nazi-era punch card business.
²²

Carmille then displayed posthumous commendations for valor and bravery bestowed on his father by both the Allies and the French government. And he showed us an IBM newsletter published during Nazi occupation for employees of the French subsidiary. The strictly business publication featured a photo of an IBM commemorative medal depicting Thomas J. Watson's face set against a regal laurel wreath.
²³

Later, a contemporary of the younger Carmille's sent my French publisher a CD-ROM filled with photographs of heretofore unknown documents. They indicate that the elder Carmille was in direct contact with officials of the
Maschinelles Berichtwesen (MB),
the Nazi government punch card agency in Berlin. Lieutenant von Passow of the
MB
exhorted the elder Carmille to replace the French demographic service's Bull machines with Dehomag equipment, even as Passow lamented that the German machines were dependent upon "American money and technology." The documents also support Robert Carmille's chronology that his father's service took pains not to extend the professional census to the occupied zone, thereby denying the Reich information it needed to complete its plan to organize slave labor in France. This instruction was reversed once the elder Carmille was discovered and taken to Dachau. But by then, it was too late to materially undo Carmille's sabotage.
²⁴

Much more research has yet to be undertaken in France. Several French men involved in the statistical service have kept their stories quiet for decades but are now ready to offer the documents and testimony—long kept secret—to chronicle exactly what was and was not done with punch cards during the war.

POLAND

The largest cache of discoveries involved Poland. In 1939, after the German invasion, IBM divided up occupied Poland into two commercial territories. The first centered in Upper Silesia, in land annexed by Germany and serviced by Dehomag. The second was conducted in the remainder of occupied Poland, the so-called General Government that encompassed cities such as Krakow and Warsaw. The General Government territory was to be serviced by a newly incorporated IBM Polish subsidiary, known as Watson Buromaschinen GmbH, directly controlled by IBM NY. Polish survivors and new documents have shed great light on the Hollerith presence there.

Historical journalist Christian Habbe of
Der Spiegel
first sent me information regarding the
Abteilung Hollerith
at the Stutthof concentration camp. The information has been quietly residing on the Polish government memorial's recently established Polish-language website at
www.kki.net.pl/~museum
. Stutthof was Hollerith-coded 13. In recent years, Stutthof archivist and historian Marek Orski has documented more information about Hollerith than any other camp historian, including those at Auschwitz.

Stutthof 's
Abteilung Hollerith
was organized in early August 1944 as deportations intensified and the camp's population suddenly grew to 50,000. SS Rottenfuhrer Werner Reiss was ordered to undergo training at a
Zentral
Institut
seminar at the Storkow concentration camp, which maintained some two dozen IBM machines. On August 4, 1944, Reiss took the 6:20 p.m. train to Danzig where he connected to the Berlin express at 11:20 p.m., and then shuttled to Storkow. The next day, his training began. After Storkow, he attended additional seminars for "Hollerith file experts" at the
Zentral Institut
main office at Block F, 129 Friedrichstrasse in Berlin. Thereafter, Reiss re ported to the SS Economics Administration, Office D II in Oranienburg, the agency overseeing slave workers and the Extermination by Labor campaign.
²⁵

When Reiss returned to Stutthof, he established the camp's
Abteilung
Hollerith,
assigning a group of Polish prisoners, including Leszek Zdrojewski, Bronis
/
law Pep
/
lonski, Julian Krawczyk, and Krzysztof Dunin-Wa"Hollerith Kartei" (even though they were paper, not punch cards) are preserved in Stutthof 's Hollerith archival files. The highest number belonged to Prisoner 99044, who entered the camp on October 27, 1944. The punch cards identified nameless prisoners by their Hollerith codes for scores of needed job skills. Duplicates of the professional file were maintained in Berlin.
²⁶

An extraordinary eyewitness is Leszek Zdrojewski, who headed up Stutthof's
Arbeitseinsatz
and interfaced with the camp's
Abteilung Hollerith.
In Fall 1944, he was sent to
Zentral Institut
in Berlin for Hollerith training. Accompanied by Reiss, Zdrojewski spent two to three days at
Zentral Institut,
where he saw some twenty sorters and tabulators served by a staff of more than 100 clerks, frantically feeding machines and searching for specific professions among the camps' populations.

During my presentation at the Jewish Historical Institute in Warsaw, I unexpectedly encountered Krzysztof Dunin-Wasowicz, one of the prisoners forced to work in Stutthof ’s
Abteilung Hollerith
. Dunin-Wasowicz is now a retired historian, who for some time was associated with the Polish Academy of Science and understands the uses of Hollerith. During a twenty-minute presentation, he outlined the history of IBM’s technology at Stutthof. An electrifying moment came when I held up an enlargement of the secret Hollerith Camp Codes. I pointed to code 6,
Sonderbehandlung
, that is "extermination." Dunin-Wasowicz acknowledged the codes were the exact ones the Nazis used at Stutthof, including code 6 for extermination in the camp’s small but active gas chamber, which murdered some one thousand people.
²⁷

Other Polish eyewitnesses also came forward. The newspaper
Slowo Polskie
located Leon Krzemieniecki, probably the only man still living who worked in the Hollerith Department of the railroad office that kept tabs on all trains in the General Government, including those that sent Jews to their death in Treblinka and Auschwitz. It must be emphasized that Krzemieniecki did not understand any of the details of the genocidal train destinations. Indeed his duties required tabulating information on all trains, from ordinary passenger to freight trains. Krzemieniecki's interview in the newspaper, and subsequent extensive oral history with me, revealed that the railway's Hollerith Department in Krakow required a five-room office on Pawia Street equipped with fifteen punchers, two sorters, and a tabulator that he recalls was "bigger than a sofa." The high-security office was guarded by armed railway police.
²⁸

Fifteen Polish women were employed just to punch the cards and load the sorters. Three German nationals supervised the railway office, undertaking the final tabulations and summary statistics in great secrecy. Handfuls of printouts were reduced to a small envelope of summary data, which was then delivered to a secret destination. Truckloads of the preliminary printouts, which created the secret summaries, were then regularly burned, along with the spent cards, Krzemieniecki recalls.

As a forced laborer, Krzemieniecki was compelled to work as a "sorter and tabulator" ten hours per day for two years. He never realized in any way that his work involved the transporation of Jews to gas chambers. "I only know that this very modern equipment made possible the control of all the railway traffic in the General Government," he told the newspaper. Only after
IBM and the Holocaust
was released did he begin to recall the significance of his work.
²⁹

In 1944, as the Russians advanced, his group loaded the machines onto trucks, which moved the equipment to Dresden. "I think they vanished without a trace," he added.
³⁰

Krzemieniecki's relatively small group punched in only a limited amount of information on the ten-column cards, mainly the number of the train, whether it was a cargo train, whether express or regular, and the distance traveled. For example, a cargo designation was coded 8. Five-digit kilometer records were punched in as well. No alphabetical machines were used, hence all codes had to be memorized. An "outside technician," who spoke German and Polish and did not work for the railroad, was almost constantly on site to keep the machines running. The technician generally undertook major maintenance on the machines approximately once each month.
³¹

The Pawia Street operation undoubtedly interfaced with a much larger and robust Hollerith operation at the Group IV Transportation Office elsewhere in Krakow. This office continuously tabulated details about the length of train lines, number and availability of locomotives and freight cars, as well as the amount and type of cargo, and "the number of persons transported."
³²

Railway management was among IBM's most diverse and highly developed applications. Typically, destinations were specifically preprinted on the IBM cards in order to locate and route boxcars and engines. Accounting cards organized and itemized the freight billing. Locomotive efficiency studies constantly sought to maximize fuel use and typically tabulated the exact amount of coal used to haul specific types of freight in the boxcars. The tailored railroad management programs, the custom-designed punch cards printed at IBM's Rymarska Street print shop across from the Warsaw Ghetto, and the leased machines utilized by railways in Poland were not under the German subsidiary, but the New York-controlled subsidiary in Warsaw, Watson Buromaschinen GmbH.

"I knew they were not German machines," recalled Krzemieniecki in the newspaper interview, and in our later discussion. "The labels were in English. . . . The person maintaining and repairing the machines spread the diagrams out sometimes. The language of the diagrams of those machines was only in English."
³³

I asked Krzemieniecki if the machine logos were in German, Polish, or English. He answered "English. It said 'Business Machines,' " I asked, "Do you mean '
International
Business Machines'?" Krzemieniecki replied, "No, '
Watson
Business Machines.' "
³⁴
That was the correct answer. In Poland, IBM NY's new subsidiary operated under the German legal name: Watson Buromaschinen. But the Polish machines proudly bore logo tags with the subsidiary's name in English: Watson Business Machines.

Among the most dramatic post-book revelations was the discovery of the massive Hollerith statistical center in Krakow known as the
Hollerith
Gruppe,
staffed by more than 500 punching and tabulating employees and dozens of machines. Research discovered a previously unknown Berlin agency called the Central Office for Foreign Statistics and Foreign Country Research, which continuously received detailed data from the Statistics Office of the State Secretariat in Krakow. Nazi Hollerith expert Richard Muller headed the operation in Krakow.
³⁵
Discoveries about this office answered questions about where much of the data for all of Poland was processed.

A variety of Hollerith-equipped Nazi offices had been operating across occupied Poland from the day of invasion on September 1, 1939. America was not in the war; Watson had not yet returned his medal; and hence he maintained his complete commercial support for the Hitler regime throughout the initial rape of Poland. As part of this strategic commercial support, IBM NY agreed to a vital installation of machines so massive it was not called a Hollerith Department, but a
Hollerith Gruppe.
This vital installation would permit the Nazis to organize the systematic looting and subjugation of Poland, as well as implement other plans for its citizens.

Just after invasion, Hitler's General Government in Warsaw asked its Regional Planning Department to establish a Central Statistics Office at the shuttered Jagiellonian University in Krakow. By April 1940, the Nazis formed a working group comprised of a single German statistics expert assisted by former employees of the Polish Statistical Service and other Polish civil servants. This group sifted through some 60,000 printed volumes of raw information in the Polish Statistical Service library in Warsaw, preparatory to conversion to Hollerith data. The selected information, along with all previously supplied Hollerith machines and staff, were relocated from Warsaw to the new Nazi agency in Krakow. So large was the enterprise, the library of raw intelligence to be punched filled two halls.
³⁶

By September 1940, the Reich issued a Decree for Statistics in Poland, creating the new expanded "Statistics Office." Within a few months, the Krakow Statistics Office at 24 Nurnerstrasse subsumed most other statistical operations in Poland. By late 1941, the Statistics Office employed 420 persons including 16 Germans in 6 distinct groups—Group I: Administration; Group II: Population and Culture; Group III: Food and Agriculture; Group IV: Economic Trade and Transportation; Group V: Social Statistics; and Group VI: Finance and Tax. A November 30, 1941, Statistics Office report explains, "The
Hollerith Gruppe
area of operation stretches across all subject areas," adding that a major expansion plan would see staffing rise to 500 persons within a month.
³⁷

The expansion was dependent on more leased machines, spare parts, company technicians, and a continued, guaranteed supply of millions of additional IBM cards. Because backlogged orders for Hollerith machines required a year or more to deliver, IBM's long-term supply commitment almost certainly dated to the first days of World War II. Indeed, the Statistics Office report was written just weeks after IBM's European general manager Werner Lier visited Berlin to oversee IBM NY's deployment of machines in Poland and other countries. The Krakow Statistics Office's November 30, 1941 report assured Berlin, "the installation of the equipment necessary for the work of the
Hollerith Gruppe
has commenced. It is estimated that the equipment will be ready for use by the end of the year, and that training of the prospective employees can begin. The employee designated as the leader of this group now participates in a seminar in Berlin to train for this subject area. . . . Survey material is already in preparation."
³⁸