PART ONE OF EIGHT
Enigma
How Joe Desch, based in Dayton, beat the Nazi code and helped win the war
By Jim DeBrosse
Dayton Daily News
Debbie Anderson knew in the summer of 1986 that time was running out for her 79-year-old father. After a series of small strokes and then a broken hip, Joe Desch, a man of iron independence, was forced to recuperate in a Kettering nursing home. There he took with him the painful secret he had kept coiled inside for most of his life.
"He seemed still to be fighting battles that should have been settled years ago," Anderson said. "There were little things that would set him off on tirades."
Anderson knew only that her father's pain and anger were tied to the World War II codebreaking research he had done at Dayton's National Cash Register Co. — the work that had driven him to a nervous breakdown, but which he had never been able to talk about, not even to explain the Congressional Medal for Merit he had received for his war service.
But then, one morning that summer, Anderson thought she had found a way to help her father unburden his heart. There was an article in the paper about the late Capt. Joseph Rochefort, who was receiving a posthumous medal from President Ronald Reagan for cracking a World War II Japanese code and providing the key to one of the U.S. Navy's greatest victories at the Battle of Midway.
"I went out in the morning," Anderson said, "when he was brightest, and cheerfully handed him the clipping and asked, 'Dad, did you know this guy? See, they're giving him a medal like yours — but posthumously.' "
Desch burst into a stream of invectives "that could have been heard all over the state," she said. "He immediately tensed up and yelled, 'They've probably found out I'm in here and have this place bugged — afraid I'll spill the beans. Don't you ever dare bring that up again!' "
Anderson didn't. A year later, her father would be dead from a massive stroke.
FALL 1989
It was through her 10-year-old son that Debbie Desch Anderson first began to unlock her father's mystery.
Jesse, her younger boy and a fourth-grader at Holy Angels, had to write a family history for school. Jesse wanted to profile his grandfather, Joseph R. Desch, an electrical engineer and executive at NCR, who had died two years before. The boy was fascinated by his grandfather's personal things — scads of old photographs, papers and memorabilia boxed up in the basement of her Kettering home.
Anderson and her son started digging through her father's items. She happened upon two thick transcripts she had seen before but had never bothered to read. They were Desch's interviews with Henry Tropp, a historian from the Smithsonian Institution, dated Jan. 17 and 18, 1973.
This time Anderson began skimming the 300-plus pages of questions and answers in earnest, finding mostly technical stuff about Desch's role in the development of the modern computer. But beginning on page 111 of the first-day's interview, and continuing to page 119, the text had been slashed through with a felt-tipped pen. In the margin was written: "Delete from tape and manuscript."
The redacted words spoke of "an electronic cryptanalytic machine," of classified code names, British scientists and of top-secret equipment dumped and buried in the middle of the night.
Anderson devoured the words, hardly believing her eyes. Here at last was a glimmer of her father's top-secret work during WWII.
"He always said it had to do with codebreaking, and he wouldn't say anything more," she said.
As a teen-ager and young adult, Anderson never pressed him on it, either. She knew her father had gone on to head the military research division at NCR during the 1960s, and that's all she cared to know. "I was a true child of the '60s. I was embarrassed that my father was part of the military-industrial complex," she said.
Desch had burned the most crucial NCR war documents — but somehow had left the transcripts, and those eight pages intended to be stricken from history, at his home in Kettering. Now here they were, in Anderson's hands, like a thin ray of light pointing the way to treasure deep inside a cave.
For the next decade, Anderson would try to illuminate her father's mystery, often finding herself at odds with national intelligence officials and skeptical historians. But in doing so, she would discover the historic role that her father — as well as NCR and Dayton — played in shortening the world's most horrible war.
In 1942 and 1943, Desch had headed a top-secret program at NCR to develop a high-speed deciphering machine, called a Bombe, to crack the Nazi submarine code. The project was second in priority only to the Manhattan Project that built the Atom Bomb, and perhaps second only to the bomb in hastening the war to an end.
Joseph R. Desch died on Aug. 3, 1987, at age 80, with his story untold — until now.
With the help of Anderson and Colin Burke — a University of Maryland history professor who first unearthed the details of the NCR project — as well as newly released information from the NCR archives and dozens of interviews with the men and women who built the NCR machines, the Dayton Daily News will recreate those crucial months in Dayton leading up to the production of the NCR Bombe.
JUNE 5, 1943
It was a sunny morning with just a hint of sea haze on the horizon as Lt. "Goose" McAuslun and Lt. Richard S. Rogers flew their Navy warplanes over the vast, empty reaches of the mid-Atlantic, patrolling the vicinity of 35 degrees North, 45 degrees West, midway between Jacksonville, Fla., and Morocco.
McAuslun's Avenger dive bomber was armed with four depth bombs. Rogers' Wildcat fighter bristled with high-caliber machine guns. But in the blind-man's game of hunting enemy submarines, the pilots had something else just as important as their weapons — reliable intelligence information locating a line of 17 German U-boats in the area, lying in ambush for a westbound Allied convoy.
The pilots' mission: Find and kill the subs. The pair was on the last leg of a five-hour patrol, 70 miles from their escort carrier U.S.S. Bogue, when McAuslun spotted U-217 cruising placidly on the surface about seven miles to their right.
McAuslun signalled Rogers, who banked his Wildcat and dove, strafing the sub's deck. "He just tore off and left me in his smoke," McAuslun, now 81, recalled from his Henderson, N.C., home.
McAuslun followed in his tubby Avenger and released his four depth bombs at 75 feet, just as the sub began to dive.
But the crew of U-217 was too late. The four bombs exploded on either side of the sub's hull, lifting it out of the water and splitting the hull. The sub sank in just 33 seconds, with Rogers blasting at it one last time to send it on its way.
It was the first time the U.S. Navy had sunk a Nazi submarine in a purely offensive action. The day before, using the same intelligence, pilots from the Bogue had damaged and scattered three other U-boats. Naval historians say those two days marked a turning point in the Battle of the Atlantic.
The Navy pilots had no idea that the information guiding their mission had originated from Building 26 at NCR in Dayton, where an intercepted German message had been cracked just days before. Naval intelligence had relied on a high-speed decrypting machine called a “Bombe†— designed and manufactured in a top-secret program at NCR. So secret, in fact, that NCR's key role in the Allied intelligence effort known as Ultra would not come to light until half a century later, when a 1992 directive by then-President Bill Clinton released many classified documents.
Since the early 1970s, much attention has been given to the early Polish and British successes in cracking the Enigma code. The Poles called their first decoding machine a "Bomba," perhaps after the brand of ice cream cones favored by the codebreakers. Operating from the famed codebreaking school at Bletchley Park outside of London, the Brits refined and further mechanized the device, based on the theoretical work of mathematician Alan Turing. They dubbed their device a "bombe."
But by the spring of 1942, the German navy was again operating in total secrecy and with a vengeance, thanks to an upgrading of their Enigma machines. With the British Ultra effort stumped and the Germans dispatching ever more submarines to the Atlantic, the Allies feared they would lose ships to the wolfpacks faster than they could be replaced.
North Atlantic sinkings more than quadrupled in the last half of 1942 compared to the last half of 1941 — from 600,000 tons to 2.6 million tons. "And each of the nearly 500 ships sunk in those six months," wrote military historian David Kahn in Seizing the Enigma, "meant more freezing deaths in the middle of the ocean, more widows, more fatherless children, less food for some toddler, less ammunition for some soldier, less fuel for some plane — and the prospect of prolonging those miseries."
The Enigma was like a typewriter that encoded messages by scrambling each keystroke through a series of rotors. It could generate billions upon billions of possible letter combinations.
But unbeknownst to the Germans, the Poles and the Brits had been able to crack the three-rotor Enigma machine, relying in part on captured German documents. But when the German Navy added a fourth rotor on Feb. 1, 1942, the number of possible combinations for producing any one letter overwhelmed their decrypting abilities.
Under increasing pressure from the U.S. Navy, which had been kept in the dark while soaring numbers of its ships and sailors were lost to the wolfpacks, the British finally relinquished their control over Ultra and told the Americans to give it a go.
What was needed, and in a hurry, was a high-speed decoding machine that could run through all the possible Enigma combinations at heretofore unheard-of speeds — a machine that the British had been working on since late 1941 without success.
Navy theoreticians at the Massachusetts Institute of Technology envisioned an all-electronic machine many times faster than the British bombe. But only two companies in the United States had the technical capability at that time to produce such a marvel — IBM and Dayton's NCR. The obvious choice for the Navy was NCR, where chairman of the board Col. Edward A. Deeds had a long working relationship with MIT as well as the Navy's top brass. It also had idle capacity — NCR, unlike IBM, had been ordered by the War Production Board to stop making its major product, cash registers, for the duration of the war.
In the end, the weight of the Navy’s demands, and the nation's, would fall most heavily on one man’s shoulders — those of Joe Desch. A modest but brilliant engineer, at age 35 he headed NCR's electrical research laboratory. Desch had already contributed to the war effort, unknowingly at the time, by inventing an electronic counter capable of operating at 1 million counts per second — at least 100 times faster than anything achieved before — for use in developing the first atomic bomb.
Unlike the Navy's theoretical engineers, who were mostly graduate students and professors at MIT, Desch was a floor-trained industrial genius, as savvy about front-office politics as he was knowledgeable about state-of-the-art electronics.
The MIT engineers "had a very different orientation than a practical engineer like Joe, but he was able to handle them all and get the job done. He was adored by these guys," said Colin Burke, former historian-in-residence at the National Security Agency and author of Information and Secrecy, the first book to detail NCR's attack on the Enigma Code.
In fact, when Desch took his wife, Dorothy, to Boston to meet the Navy personnel at MIT, she was completely thrown when people at the welcoming cocktail party started greeting her husband as "Dr. Desch."
"Dad said that people who mattered knew better, but that it was incomprehensible to the graduate students there that a man of his importance wouldn't be a Ph.D.," Debbie Anderson said.
Born in 1907, four years after the Wright brothers' first flight, to a German immigrant mother and a long line of Dayton area wagonmakers, Desch capitalized on his keen intelligence, and his fascination from an early age with the budding technology of radio, to rise above his humble beginnings in Dayton’s Edgemont neighborhood. He worked his way through the University of Dayton and graduated with honors in electrical engineering.
Prior to the war, Desch had built a national reputation for his work in designing miniature, fast-firing gas tubes — the "microchips" of the 1940s and the basis for electronic calculators at the time. The Navy was betting that if anyone could build a new generation of super-fast deciphering machines, it was Joe Desch.
The Bombe project would not only prove to be the biggest technical challenge of Desch’s career, but an overwhelming emotional drain. For the next two years, it would mean working 14-hour days under mounting pressure from Navy officials. It would mean severing relations with his German immigrant relatives. It would mean being placed under 24-hour surveillance, with his supervising officer quartered in his own home.
For the duration of the war, his life would be pinned under a microscope. And before it was over, he would suffer a nervous breakdown.
"They kept the pressure on, kept the pressure on and poor Joe took it very seriously," said Bob Mumma, the NCR business manager who reported directly to Desch and took over for him after his breakdown. "It destroyed Joe's health."
Mumma, now 95 and a resident of the Otterbein Home in Lebanon, scoffed at the suggestion that he and others involved in the Bombe project were heroes who shaved months from the conflict and saved hundreds of thousands of lives.
"We were just trying to win the war," Mumma said.
SPRING 1942
For weeks that spring, Desch pored over blueprints for what promised to be a remarkable new code-breaking machine developed by Navy theoreticians. Even though he was overseeing 13 other military projects at NCR, this would be his ultimate challenge — a totally electronic deciphering machine 100 times faster than anything in use.
The Navy Bombe would rely on tens of thousands of the miniature, fast-pulsing tubes Desch himself had invented — pushed to the limit of their capabilities. After all, the foe they were attacking seemed almost invincible.
When the German military purchased all rights to the Enigma Cipher Machine first conceived by Dutchman Hugo Koch in 1919, and added several of their own innovations, they had every reason to think their coding system was impenetrable — even if the machine were captured.
In essence, the Enigma was an electrical typewriter that scrambled each keystroke through a series of alphabetized rotors, so that the text seemed generated at random. When a key was depressed, it tumbled the first rotor and, after a complete revolution, a second and a third rotor — just like the odometer on a car. But on the Enigma, each rotor scrambled the alphabet in a different way.
The operator could set each rotor at a different starting position as well, changing the line-up of the rotors. A notched ring on one rotor controlled the rotational behavior of the rotor to its left. As an extra security measure, the machine contained a plugboard that transposed individual letters (a for e, for instance, or h for m) to further confuse anyone trying to crack the code.
To unscramble the text, the receiving Enigma operator would be sent the original rotor positions, in code, as part of the message. He would also have a chart instructing him in that day's plugboard settings. Only an officer could change the line-up of the rotors inside the machine, based on frequent orders from command headquarters.
With all those variables in use, the possible number of Enigma encipherings for each character was staggering — 100,000,000,000,000,000,000,000 — or 10 to the 23rd power. And when the Germans introduced a four-rotor machine in early 1942, the possible combinations grew to 10 to the 26th power. (By comparison, the number of all atoms in the observable universe is 3 to the 80th power.)
Indeed, the Enigma might have been impregnable, except that even the Germans did not always follow proper communications security. Lapses and laziness, such as not resetting the wheel positions between dispatches and using predictable greetings and sign-offs, would create "cribs" (strings of known or suspected plain text) that would give Allied codebreakers the wedge they needed to decipher intercepts.
Nor did the Germans know that the British had secretly captured Enigma code charts and the machines themselves by raiding unwary German weather ships and boarding abandoned U-boats that couldn't be scuttled by their crews in time.
The NCR Bombe would basically work like an Enigma machine in reverse. By feeding it a "crib" of known or suspected plain text from an intercepted message, the Bombe would crunch through all the billions upon billions of possible rotor combinations on the Enigma machine until it arrived at the rotor sequence that had enciphered the text. Once the positions of the four rotors were known for any part of the cipher, the remainder of the scrambled text could be easily broken by running it in reverse through a captured Enigma machine.
But the more Desch went over the Navy plans, the more his practical side bumped up against a major stumbling block — an all-electronic machine would require far too many gas tubes (70,000 in all) and generate too much heat to operate reliably.
But what would Desch tell the Navy intelligence officers who had put so much faith in his engineering skills?
Monday: A complex man with an impossible job.
• Contact Jim DeBrosse at 225-2437 or by e-mail at jim_debrosse@coxohio.com.
Proud Daughter of Walter (Monday) Poniedzialek
540th Engineer Combat Regiment, 2833rd Bn, H&S Co, 4th Platoon
There's "No Bridge Too Far"