by Jim DeBrosse
Contents
Imprint
Title Page
Dedication
Acknowledgments
Half Title Page
Prologue
1.
Building the Perfect Machine
2.
Guesswork, Moxie, and Just Plain Luck
3.
Miss Aggie’s Big Blunder
4.
Toward an American Bletchley Park
5.
A Giant Leap . . . and a Step Backward
6.
The Turing Memo
7.
Troubles with Adam and Eve
8.
U-boats on the Run
9.
The WAVES Come Aboard
10.
A Well-Oiled Machine
11.
An Enemy Within?
12.
Triumph!
13.
New Challenges . . . and Breakdown
Epilogue: Burying the Past
Endnotes
Notes
Bibliography
Index
About the Authors
Also by Jim DeBrosse
Also by Colin Burke
Copyright
For Ira and Charline
—JIM DEBROSSE
To my wife, Rose, who gave me my son, Andy,
my very special gift in life
—COLIN BURKE
Acknowledgments
A NONFICTION BOOK jointly authored by a journalist and an historian would appear to be a natural partnership: both professions dig as deeply as possible into the truth of a story. But while the journalist often validates with telling details and insightful quotes, the historian insists on the bedrock of written documentation and unbiased data. Coauthors Jim DeBrosse, a reporter for the Dayton Daily News, and Colin Burke, a leading authority on American intelligence agencies and the emergence of the computer, found theirs an uneasy alliance at times—much like the British and American codebreakers featured in this book. But they hope that, together, they have made accessible a highly arcane but significant piece of modern American history to a wider public.
The collaboration grew out of an eight-part series that DeBrosse wrote for the Dayton Daily News, “NCR and World War II: The Untold Story.” Burke was the expert source for the articles. Soon af-ter the series appeared in the paper in spring 2001, Burke called DeBrosse and announced that he had gathered further information on the U.S. Bombe program, including a dramatic espionage attempt.
The two agreed that DeBrosse would research the human side of the story, with as many interviews with survivors and their families as possible, and Burke would supply the historical spine of the book with archival documents—a process he had begun more than a decade before. DeBrosse was responsible for the writing of the text and for creating a narrative; Burke outlined the book’s historical themes, established the accuracy of its facts, and helped his coauthor grasp the more daunting technical aspects of the German Enigma machines and their electromechanical foes, the Bombes. Together, the two aimed for a balance—between the claims of the British and the claims of the Americans, between a technical treatise for codebreaking aficionados and a compelling narrative for the uninitiated, between a grand reinterpretation of history and a complacent acceptance of past writings.
Little information was available when, in the late 1980s, Burke began exploring the origins of the modern computer, in which the NCR Bombes played a part. The National Security Agency, the Navy, and the Army had kept the military’s 1930s and 1940s computer research under Top Secret and Ultra protections. Histories of the American intelligence agencies released to the National Archives at the time were vague and typically censored. Many surviving participants in America could not talk about their work, and British documents were hopelessly locked away. Even F. H. Hinsley’s authoritative series of books on British intelligence in World War II contained little useful information.
Burke’s first book, Information and Secrecy: Vannevar Bush, Ultra and the Other Memex, had to rely heavily on open primary sources, especially those related to the history of computers and computing, hoping that enough bits of evidence could be gathered and pieced together to yield some answers. He spent years with the Hagley Museum’s magnificent and well-managed collection of computer industry documents, with deep gratitude to its helpful staff.
The Hagley’s collections provided enough hints to direct Burke’s research into primary sources at the Smithsonian Institution, the Massachusetts Institute of Technology, the Rockefeller Archives, Dartmouth College, the U.S. Patent Office, the Library of Congress, the Washington and Suitland branches of the National Archives, the Navy’s legal branch, and the Naval Historical Center.
Along the way, many other institutions were contacted for help, including the hundreds of libraries badgered by Burke’s home university’s interlibrary loan department. Especially tolerant were the archivists and historians at the NCR Archive, the Charles Babbage Institute, the National Security Agency’s history office, IBM, and the Eastman Kodak Company. Professor Brian Randall of the University of Newcastle upon Tyne in England also was very gracious.
Although Burke’s training as an historian made him skeptical of oral history and after-the-fact reminiscences, several former intelligence officials provided helpful information, including two men who were willing to have their names cited, Waldron S. MacDonald and Joseph Eachus, the latter generously consenting to interviews by both authors. MacDonald and Eachus were considerate and forthright while still protecting the secret aspects of their work.
By the late 1980s, Burke had gathered enough from open sources to persuade the government to release classified documents related to America’s codebreaking machines during the 1930s and 1940s. Letters citing the Freedom of Information Act were sent to the National Security Agency, and a few years later, a very large box of documents appeared at Burke’s door, which, for convenience, was called the RAM (Rapid Analytical Machines) File.
Unfortunately, the FOIA documents were very heavily censored, with some of the most important facts covered with impenetrable black lines. Burke spent several years trying to turn the unreadable into useful information while also pursuing other leads. He traveled to libraries, historical societies, and government offices in large and small towns in the Midwest and border South. Special thanks must go to the historians who kept the records of the Miami Valley Chatauqua and to Brian Hackett, executive director of the Montgomery County Historical Society in Dayton.
More FOIA requests were sent out to the Department of Justice, the FBI, the Bureau of Prisons, the Selective Service, the CIA, the Army record centers, and the Social Security Administration.
By 1992 Burke felt he had exhausted the primary sources and knew enough to draw some valid conclusions about the Navy’s early computer efforts. Soon after the publication of his book, he was contacted by Joseph Desch’s daughter, Debbie Desch Anderson. After a decade-long quest of her own to uncover the truth of her father’s secret labors during the war, she was organizing a reunion of the Navy WAVES who had worked on the NCR Bombes. Fortunately for all, the reunion in late 1995 came just as the federal government began to relax some of the restrictions on the Bombe project, allowing many participants to grant interviews.
Anderson, who became a friend of both Burke and DeBrosse, first brought them together. She, along with historians at the Montgomery County Historical Society, who are the guardians of the massive, four-million-item NCR Archive, approached DeBrosse in the fall of 2000 about writing a newspaper story on Desch’s and NCR’s unheralded role in Ultra. Anderson, in turn, recommended Burke as an expert source.
Ande
rson deserves much credit for bringing the U.S. Bombe proj-ect out of the technical history books and into the public eye. Her quest to unravel her father’s secret work laid the human foundation for this book and brought credit not only to her father and the other NCR engineers on the Bombe project but to hundreds of Navy WAVES who were never told the purpose of their work.
A year or so after the WAVE reunion, the government decided to release more intelligence agency documents. By 1998 the new RG457 Historic Cryptologic Collection was available at the National Archives. As before, John Taylor lived up to his legend: he was a masterful guide through the series. Then, to the surprise of many historians (and those in the intelligence community), another release soon followed—the RG38 “Crane” materials, which had been held for decades in a secret Navy depository in Crane, Indiana. The Crane collection contained some startling new documents, and Barry Zerby provided much help as he and other NARA archivists struggled to handle a flood of requests.
More surprising than the Crane release was a change in British declassification policies. In the late 1990s, a significant number of documents from the World War II British Ultra projects began to appear at the Public Record Office, outside London. Although difficult to find, there were many significant and, at times, astonishing pieces of evidence.
At the same time, a group of volunteers began to rescue the famed Bletchley Park. They established a museum in the mansion and operational huts and began to publish new information on the Bombes and the other codebreaking machines built in England during the war. Tony Sale and his wife and many others at Bletchley Park provided Burke with tender care on his visits to England as he tried to keep in touch with the new document releases.
Burke also thanks the staffs at the PRO, the National Archives, the Naval Historical Center, NSA’s new Cryptologic Museum, NSA’s history office, and scholars and researchers Ralph Erskine, Jim Reeds, Frode Weierud, Phil Marks, David Hatch, David Kahn, Robert Hanyok, Rebecca Ratcliff, Edward J. Drea, Lee A. Gladwin, Lou Holland, John Traegesor, Jeffery Wenger, and Henry Schorreck.
Three years ago, DeBrosse began conducting interviews with scores of people involved in the NCR Bombe project and their family members, as well as pilots and sailors who fought in the Battle of the Atlantic. Anderson withstood hours of personal interviews and patiently fielded hundreds of phone calls to clarify details, without complaint and always with the offer of a cold beer for her visitor. The staff of the Montgomery County Historical Society and the NCR Archive never denied a request for help or an unannounced visit. Particularly helpful were Bill West, Brian Hackett, Claudia Watson, Curt Dalton, and Mary Oliver.
Other important sources for DeBrosse were Phil Bochicchio and Gilman McDonald, both of whom know the workings of the NCR Bombes, inside and out, literally; NCR engineers Lou Sandor, Robert Mumma, Don Lowden and Carl Rench, who helped piece together the Bombe project and bring to life the man chiefly responsible for its design, Joe Desch; Peg Fiehtner of the Naval Security Group Command and the volunteer staff of the Wenger Command Display of the Naval Cryptologic Veterans Association, all of whom are dedicated to honoring the proud work of Navy codebreakers; and more than a dozen Navy WAVES who shared their memories of those exciting years, especially Catherine Racz, Evelyn Hodges Vogel, Evelyn Urich Einfeldt, Susan Unger Eskey, and Veronica M. Hulick.
DeBrosse would like to thank the editors at the Dayton Daily News, Ron Rollins and John Erickson, who were instrumental in shaping the original series, as well as publisher Brad Tillson for allowing him a six-month leave to write the first draft of the book. DeBrosse also received much encouragement from his colleagues at the paper, especially Mary McCarty, Tom Beyerlein, Steve Bennish, Ken Palen, and Melanie Raley. Kathleen Schamel provided invaluable assistance as a weekend babysitter so that her brother could find time to write the book. Tom DeBrosse, as always, was a bedrock of brotherly advice, and Alvin and Jane Sanoff graciously opened their home, and their hearts, whenever Jim Debrosse needed to conduct research in the Washington, D.C., area.
Finally, the authors express their deep gratitude to literary agent Fran Collin, who never gave up on the idea of placing their book; Random House’s Benjamin Dreyer and Timothy Mennel, whose sometimes painful cuts and suggestions made this book a much better read, as well as Dennis Ambrose, who gave it his unflagging attention to detail; and, of course, editor Bob Loomis, whose gentle hand and unerring advice miraculously brought forth a coherent tale.
THE SECRET
IN BUILDING 26
Prologue
Building 26—NCR Campus—Dayton, Ohio
THE SECRET IS hidden in plain sight: a nondescript glass-and-panel box of a building standing four stories high, flanked by parking lots and set back far enough from the busy intersection of South Patterson Boulevard and West Stewart Street that few motorists even notice it, much less wonder what might have transpired there some sixty years ago. Inside the now empty structure is another empty building of tan brick and glazed cinder block, a far more interesting art-deco design that has been nearly encapsulated by the newer addition and can now be seen only from the rear.
A granite boulder emblazoned with a small bronze plaque is all that testifies to the building’s proud history. It sits along a nearby sidewalk that few people traverse even by day. The plaque reads:
In 1942, the United States Navy joined with the National Cash Register Company to design and manufacture a series of code-breaking machines. This project was located at the U.S. Naval Computing Machine Laboratory in Building 26, near this site. The machines built here, including the American “Bombes,” incorporated advanced electronics and significantly influenced the course of World War II.
October 2001
Institute of Electrical and Electronics Engineers
Nearly one thousand people worked on the Navy’s top secret proj-ect in Building 26 between 1942 and 1945. They were sworn never to divulge the nature of their work.
The full story of their struggle, sacrifice, and ultimate triumph has never been told—until now.
March 10, 1943—North Atlantic
ON A MOONLESS night of squalling snow and frothing waves, U.S. Seaman First Class Edward P. Rego kept watch for U-boats from one of the two four-inch gun platforms on the stern of the SS William C. Gorgas—unaware, as was most of its crew, that the merchant vessel, part of Allied Convoy HX228, was carrying nine hundred tons of TNT for delivery to Liverpool. Rego, a seventeen-year-old gunner from Massachusetts, was on his first transatlantic crossing. He was also cold and bored. There was so much shifting ocean out there and so little he could see.
But the enemy was closing in for the kill.
Earlier in the day, U-336 had spotted the sixty-ship convoy and had radioed its location to U-boat Control. By midnight, a wolf pack of six submarines had been assembled and was preparing to strike at close quarters.
The first to fire a spread of torpedoes was U-221, whose ambitious young skipper, Kapitänleutnant Hans Trojer, was out to prove himself. A torpedo soon found the 5,400-ton British Tucurinca, causing the vessel to explode with such force that “hundreds of steel plates flew like sheets of paper through the air,” as Trojer noted afterward in his log. A second hit the 6,600-ton American Andrea F. Luckenbach and sent “heavy debris [crashing] against my periscope, which now became difficult to turn. The whole boat re-echoed with bangs and crashes.” A third damaged the 7,200-ton American liberty ship Lawton B. Edwards.
Rego was far enough away that the explosions sounded like distant thunder, but within the next two hours he had his first taste of death from below. Just before midnight, two torpedoes from U-444 ripped into the engine room of the 7,200-ton Gorgas, instantly killing the crew there but miraculously missing the tons of explosives amidships, in the hold.
The fully loaded ship went dead in the water, and the waves started crashing over its deck. Fearing the Gorgas was a sitting target for another torpedo attack, sixteen of the twenty aft-gun crew jumped over the fantail, never to be seen ag
ain. But Rego had been grabbed in time by the ship’s cook, Manuel Moralis, who urged him to move forward, along the side of the sinking ship, to where the two men were able to don survival suits and launch a lifeboat. Over the next two to three hours, Rego and Moralis pulled other survivors from the thirty-seven-degree waters, until the British escort Harvester came to their rescue.
In all, fifty-one survivors of the Gorgas were retrieved by the Harvester, but their ordeal that day was far from over.
Like sharks trailing the scent of blood, the U-boats pressed their attack against the crippled convoy late into the morning. The Harvester, already limping from a ramming incident with U-444, was given the coup de grace at 10:00 A.M. with two torpedoes from U-432. This time, Rego had no time to reach a lifeboat; Moralis, who had saved him during the first attack, did not survive the second. Rego jumped overboard and began swimming through the icy waves. He soon spotted a life raft, but like a mirage it disappeared and reappeared from his shifting line of sight and seemed never to come nearer, no matter how hard he struggled. After nearly an hour of battling the water, he decided to free himself from his thick rubber survival suit and, with only his own adrenaline to protect him from the cold, at last reached the raft. Four hours later, a British corvette spotted the raft and retrieved Rego and four other survivors—so weakened by the cold that they had to be carried onto the ship. Of the seventy original crew aboard the Gorgas, only twelve, including young Rego, had earned the unenviable distinction of surviving two torpedo attacks in nearly the same day.
