The Secret in Building 26

by Jim DeBrosse

  As luck would have it, an apparent mix-up in the mail delivery between OP20G and Bletchley Park soon brought the simmering distrust and jealousies between the two agencies flaring to the surface. Denniston’s early October dispatch—a bag of materials containing detailed answers to all but one of Driscoll’s questions—never reached OP20G, the Navy claimed. It didn’t take long for Safford, who feared the British were breaking their promises, to push Leigh Noyes into firing off a series of complaints to the British. Through November and December 1941, angry memos and accusations flew across the Atlantic. Noyes didn’t mince his words: Britain had broken its promise to OP20G; America had no use for the Bombe; and if GCCS cooperated, Driscoll could have her method working on real problems. *18

  On December 5, Denniston telegraphed Captain Edward Has-tings, the liaison for British intelligence in Washington: “I still cannot understand what Noyes wants and am disappointed at my apparent failure as far as the Navy Department is concerned. . . . About October 1st I sent out material, answers to Mrs. Driscoll’s queries, technical details and a covering letter to Safford. Only one parcel has been acknowledged and there has been no reply to the questions in my letter. This may be one cause of this misunderstanding.”

  More puzzling still is why neither Driscoll nor Safford launched a search for the materials. Safford had received Denniston’s cover letter, listing the contents of what had been sent to OP20G and asking for answers to three questions about Driscoll’s method. She never responded.

  Noyes, who was unaware of the complexities of the mail mix-up, continued to fire off angry memos to the British, some of them clearly threatening. The U.S. Navy, he said, had never agreed to confine itself to Enigma research. It had always intended to be “operational”—that is, intercepting and decoding messages on its own. He told Hastings that all the Navy wanted from the British was the information on the Enigma and the codebooks and Enigma machine that Safford and Driscoll had requested.

  Then, belying later histories of GCCS and OP20G relations, Noyes apologized to the British, twice. On December 10 and again on the twelfth, he declared that British explanations and actions since his outbursts had satisfied him and “everyone” at OP20G. The missing package, of course, was found. On December 13, Bletchley received a cryptic yet pointed message from someone in the U.S. Navy Department: “Luke Chapter 15, v 9: And she found it. She calleth together her friends and neighbours saying: Rejoice with me for I have found the piece which we lost.”

  Noyes’s repentance did not mean all was well between the two crypto agencies, nor that Driscoll had abandoned her stubborn cause. Tensions continued as the Navy once again requested a copy of the Enigma machine. And the Americans and the British drifted even further apart in early 1942 as both OP20G and GCCS dealt with their own internal shake-ups.

  That changed in mid-1942, however, when the bureaucratic upheaval settled and new leadership under Joseph Wenger took control at OP20G, in part because President Roosevelt insisted on forging closer ties with the British. The Americans admitted to themselves that Driscoll’s catalog would not find all the Enigma settings needed to read messages. Her technique eventually found an important niche in OP20G’s codebreaking operations, but it proved useful only after the harder work had been done by the Bombes.

  DRISCOLL’S REFUSAL OF British help was one of the biggest Allied intelligence mistakes of World War II—second perhaps only to the British refusal to believe for several years, until mid-1943, that the Germans were reading their convoy Cypher Number 3, which Safford had repeatedly criticized as a “third-rate communications system” endangering Allied ships in the Atlantic. And yet Driscoll cannot carry the blame alone: the roots of the error were deep in the soil of America’s intelligence history.

  With its attention before the war focused on the threat from Japan, the United States had been late in realizing the importance of the European theater and even tardier in recognizing the importance of attacking the German encryption machines.

  But if “Miss Aggie” had possessed a more open attitude and perhaps a little less self-confidence, the outcome of Denniston’s visit could have been quite different, and the Americans could have saved at least six months in the search for their own Bombe—six months that would have been crucial in turning the Battle of the Atlantic in the Allies’ favor sooner.

  Denniston left Washington for England in August 1941 with deep concerns about OP20G and its leadership. He concluded that Driscoll was America’s version of Bletchley’s Dillwyn Knox, the crusty old English codebreaker (and Greek scholar by training) who had no use for technology or mathematicians.

  At least some of the Navy’s radio-intelligence officers would have agreed with Denniston. Long before they could gain control of OP20G, Joseph Wenger and his mentor, Admiral Stanford Caldwell Hooper, had been trying to move the Navy into a new era of automated codebreaking and advanced mathematical approaches. On the day Driscoll turned down Britain’s experience with its Bombes, the new guard was poised to drag American cryptanalysis into the electronic age.

  4

  Toward an American Bletchley Park

  November 3, 1941—Washington, D.C.

  THE MEETING AT OP20G that day must have begun uneasily, perhaps even more so than Alastair Denniston’s encounter with Agnes Driscoll two and a half months before. The group of three young engineering graduate students, recruited from MIT, came to offer their expertise to Driscoll, the Navy’s top cryptanalyst, and her supervisor at OP20G, Laurance Safford. The MIT students, viewed by the Navy as prestigious academics, didn’t object when Safford introduced each of them as “doctor” or “professor.” And Driscoll, who had spurned Denniston’s earlier offer of help, likely had a more open mind after months of grappling in the dark with the Enigma.

  Here was a face-to-face encounter between two generations of codebreakers. The emerging engineering and math experts who placed their faith in automation and electronics were reaching out to the old guard, who believed there was no viable substitute for brainpower, the traditional techniques of pencil and paper, and a good dose of luck. While Driscoll and Safford knew and sometimes even used the newer statistical techniques, such as the Index of Coincidence, their years of experience taught them that cracking a code or cipher usually depended on the lightning strike of human intuition or some unpredictable entry into the system—the theft or capture of machines and instruction books, for example, or an enemy slipup, such as transmitting the same message in plain and cipher texts.

  The MIT students spent the day at OP20G’s offices on Constitution Avenue delivering the same message to anyone who would listen: tell us what you need to do your jobs better and faster, and we can build the machines to help you. (Driscoll and Safford must have wondered if the hands reaching out to help weren’t also intent on dislodging them.) The new guard and its sponsors—most notably Stanford Hooper and Joseph Wenger—believed that the demands of cryptanalysis had changed forever. The latest wave of electromechanical ciphering devices, such as the Enigma, were forcing codebreakers to become mathematicians, statisticians, and engineers—not just linguists with a bent for puzzle solving. Without machines of their own, they believed that cryptanalysts would have to perform seemingly impossible feats of calculation to penetrate the new machine-enciphered messages. The only hope, the new guard felt, lay in the potential of high-speed calculators and the emerging use of photoelectric sensors and microfilm to detect patterns of text.

  The three MIT students—John Howard, Lawrence Steinhardt, and John Coombs, all in their twenties—went on to form a key part of OP20G’s new M section, a math research group established in 1942 that eventually included eighteen mostly engineering and math types who were given a broad mandate to do scientific research for OP20G. Much like the young British mathematicians who were then working wonders at Bletchley Park, the new guard at OP20G had been trained at MIT and other prestigious universities in the country. It included Yale graduates Howard Engstrom, Marshall Hall, and James Wakelin; Ha
rvard graduates Willard Quine and Andrew Gleason; Columbia’s James Pendergrass; and Princeton’s Donald Menzel. Although East Coast schools dominated the roster, the group also included Al Clifford from Cal Tech, and graduates garnered from some of the top midwestern universities, including Michigan (Charles B. Tompkins), Illinois (Louis Tordella), and Northwestern (Howard Campaigne).

  HOOPER, WENGER, AND even Safford had done all they could to recruit such men in the 1930s. They had scoured the university ROTCs, set up faculty scouts in math departments, and persuaded as many college men as possible to take OP20G’s cryptologic course by mail. At least six of the eighteen recruits had entered the Navy through ROTC and reserve training or by taking the correspondence courses.

  Though criticized by the British for being largely inexperienced in codebreaking, the men in the M section had backgrounds that in many ways mirrored that of Bletchley’s driving force, Alan Turing: they were young, creative, and fascinated by numbers. The youngest of M’s college recruits was twenty-year-old Gleason; the old-timer of the bunch was its leader, thirty-nine-year-old Captain Howard T. Eng-strom.

  Engstrom had the perfect credentials for leading M, although it hadn’t been an easy climb for the first-generation Bostonian. Born of a Swedish-immigrant fisherman and a Finnish nanny, Engstrom was only ten when his father died. His mother took in laundry in order to keep her talented boy in high school. According to his daughter, Eng-strom never forgot his sense of shame as he transported home his teachers’ dirty clothes from school each afternoon and brought them back clean the next morning.

  Engstrom climbed into the top academic echelons through hard work and smarts. He was only twenty when in 1922 he earned his undergraduate degree in chemical engineering from Northeastern University, the old YMCA college intended to educate boys from poorer families. He spent several years at Western Electric, acquiring a practical background in communications. After getting his master’s degree from the University of Maine in 1925, he was accepted at Yale, where he earned his doctorate in mathematics, and then at Cal Tech, where he was one of the National Research Council Fellows.

  Fluent in at least four languages, including German, Engstrom was admitted to the most prestigious mathematical center in the world, Göttingen, where he found himself among the great figures of twentieth-century mathematics and physics, spending his days in lectures and cutting-edge tutorials and his nights in Ratskellers and Biergartens. He had returned to Yale as a junior mathematics professor when he joined the Naval Reserve and began taking OP20G’s correspondence course.

  Another of the more senior members of the new guard—and one of the few with a Navy background—was thirty-four-year-old Robert Beelville Ely III. Ely had found his way into the Ivy League by way of ability as much as wealth. His father, Robert E. Ely II, was a successful dye merchant who had sent his son to prep school, Penn Charter in Philadelphia, and then off to Princeton. There Ely had taken honors in mathematics while compiling a record of extracurricular achievements worthy of fictional Ivy League hero Frank Merriwell. Athletic, handsome, and well-spoken, Ely was on the crew and de-bating teams all four years at Princeton and was twice named class orator.

  Perhaps because of his strength as a speaker, he turned from math after his undergraduate years to the study of law, and after taking his law degree from the University of Pennsylvania he founded an organization to serve young up-and-comers like himself: the Philadelphia Junior Bar Association. Ely entered the Naval Reserve in 1933 by way of the supply corps but soon grew discontented with such mundane matters as writing procurement contracts. He, too, signed up for OP20G’s correspondence course and qualified for a transfer there, where his abilities in math landed him in the M section.

  Not present at that November 3 meeting but certainly there in spirit was Joseph Wenger, who had been assigned to the Office of Naval Communications in July 1941, after a tour of sea duty. Perhaps more than anyone else in Navy communications intelligence at the time, Wenger was determined to crystallize OP20G around the new scientific types. He and Stanford Hooper had to fight a near constant battle from the early 1930s through the 1950s to gain the support they needed to bring naval communications into the electronic age.

  Hooper had been the first to create a Navy section in advanced code and signal research in the 1930s. He soon entrusted it to Wenger, then a thirty-year-old Annapolis graduate and one of Agnes Driscoll’s first students at OP20G. Wenger also had practical experience as a seagoing communications officer, having served on the cruisers Pittsburgh and Milwaukee. He became the driving force behind what by the late 1940s would become the world’s most technically advanced codebreaking agency.

  Born in 1901 in the bayou country of Patterson, Louisiana, Wenger had the mind of a scientist but the soul of an artist, according to his son, Jeffrey. When Joseph Wenger was four, his father—a Swiss immigrant with a doctorate in philosophy—moved the family to Washington, D.C., to take a job as an economic analyst with the State Department. It was in those early years in Washington, as well as the summers on his mother’s family plantation in Louisiana, that Wenger developed his talent for painting and sketching. But by the time he was seventeen, Wenger’s more practical side took over: he realized that art would be a risky way to make a living and opted instead to apply to the U.S. Naval Academy, where he could get a free college education. Even at Annapolis, Wenger appeared to suffer second thoughts about his career choice: a haunting 1922 self-portrait in academy uniform shows a wary, serious young man with penetrating brown eyes.

  Quiet and reserved, Wenger frowned on smoking, never swore or lost his temper, and never took a drink until he suffered a nervous breakdown in the fall of 1943, following the worst of OP20G’s wartime struggle, and only then occasionally imbibed in alcohol per his doctor’s orders. To a man, those who worked in the Navy with Wenger during and after the war say they never saw him laugh or joke. “He was all business—a very dedicated man—and he had his mind on his work,” said George M. Robb, who worked under Wenger at the Naval Security Annex in Washington after the war. Joe Eachus, a member of the M section, concurred that Wenger appeared to be utterly humorless but that his grasp of codebreaking techniques and the latest communications technologies “made everybody he commanded respect him.”

  Those who met Wenger rarely forgot him. He was six feet tall and exceedingly thin, with hawklike, studious features, which explains why his nicknames at Annapolis included “Skinny” and “Buzzard.” Stomach problems that plagued him most of his adult life later gave him a frail, hollow-cheeked appearance that made him seem taller than he actually was.

  Following graduation from Annapolis in 1923, Wenger began his Navy career in the Bureau of Ordnance but by 1929 had found his way into communications intelligence and the tutelage of both Agnes Driscoll and Hooper, who became his lifelong mentor. Wenger was hospitalized for his stomach problem in 1934, but by 1935 he had bounced back and was made head of OP20G’s new research desk, called Y. The new section, also created at Hooper’s urging, was devoted to the application of science to cryptanalysis. But Wenger was soon rotated out to sea again, from 1938 to 1941, and the Y section languished without him, leaving OP20G’s commander, Laurance Safford, to do it all. The Navy’s rotation policy, which dictated a tour at sea every few years for its officers, was a source of practical experience for men like Wenger, but it was clearly a hindrance to advanced research and other Navy programs that demanded specialized personnel.

  The chain of influence from Hooper to Wenger and, later, to Eng-strom ultimately led back to one of the most respected and controversial scientific figures in America at the time, MIT’s Vannevar Bush. Bush was not only one of the nation’s outstanding academic engineers but one of the most powerful scientist-politicians in the nation’s history—a mover and shaker in the rise of government-driven Big Science before and after World War II. Bush believed in close ties between the military and research engineers, as long as the engineers were given enough money and freedom from bureaucr
acy to accomplish their best work. As the first man to conceive the idea of a desktop personal computer (though he was far from designing a successful working one), Bush launched a decades-long quest to build an information machine that would aid both scientists and codebreakers.

  In 1931, Bush and his students at MIT achieved national and international fame by building the world’s largest and most powerful calculating machine—a room full of gears and rods and shafts and motors that took mechanical analog computing to its limits but looked as though it were inspired by Rube Goldberg. Bush called it a differential analyzer, but the popular press was quick to dub the sprawling device a “giant brain.” Although international visitors flocked to MIT and clones of the analyzer were built in Europe and America—including one by General Electric and another at the U.S. Army’s Aberdeen Proving Ground—the invention didn’t bring in the flood of money that Bush had expected for MIT.

  The analyzer also overshadowed other projects at MIT, ones with far greater potential for the future of calculation because they were based on electronics and photoelectricity—technologies that promised to deliver results at close to the speed of light. To support his students and to continue his research, Bush began a search for ways to raise funds. Over the years, he refined his grantsmanship skills to a high art, wedding the hottest technologies of the time to the desires of funding agencies. To provide resources and jobs for his “boys” at MIT, he also developed a network of contacts among the executives of the nation’s leading companies, including NCR.