May 14, 2013

Finest Hour 149, Winter 2010-11

Page 31

Churchill and Intelligence – Sigsaly : Beginning the Digital Revolution

“Today digital technology is the backbone of the information industry….But the pioneering work for many of these capabilities was performed early in World War II.”
—United States National Security Agency

2024 International Churchill Conference

Join us for the 41st International Churchill Conference. London | October 2024

More

By Christopher H. Sterling

Professor Sterling ([email protected]) teaches media law and policy at The George Washington University, Washington, D.C. His articles in Finest Hour include “Churchill Afloat: The Liners He Rode” (FH 121); “Face-Off: Churchill, Reith and the BBC” (FH 128); and “Getting There: Churchill’s Wartime Journeys” (last issue). He also reviews books for FH and the Washington Society for Churchill, which he heads.

The hundreds of wartime messages between Churchill and Roosevelt are well known and documented. Not as well appreciated is that these text messages, sent by secure teletype, diplomatic cable, or courier, represent only a part of top-level wartime communication. For the two leaders there was another, faster way to communicate across the Atlantic: the telephone. Protecting the security of those messages gave rise to the SIGSALY system—regarded as the “pioneering work” in digital technology by the U.S. National Security Agency.¹

In the Beginning

When the British General Post Office and American Telephone & Telegraph opened the first commercial radio telephone links in 1927, charges for their use were very high: $75 or £15 for three minutes, about $900 in today’s values. Trained operators were required to make the complex connections. And, because they relied upon radio transmission (the first telephone cables entered service only in 1956), security was a serious problem.

Any radio transmission can be intercepted and its coded signals broken, as the British learned early in World War II from Bletchley Park’s “Ultra” codebreaking effort. Security considerations greatly limited the use of transatlantic telephone calls in the war’s first two years, requiring close monitoring to ensure that vital information did not reach enemy ears.

Nevertheless, the telephone was literally the only way to communicate over vast distances. Churchill particularly depended on it, often to the despair of people he was talking to (see “Joys of the Scrambler” overleaf). At first the Allies tried to protect essential voice traffic with the A-3 voice-scrambling system developed for its commercial telephone users by AT&T. Based on 1920s analog technology, it was not nearly secure enough for communicating military secrets.²

Well aware of A-3’s shortcomings, American and British signals experts insisted that the voice link only be used with carefully controlled code words, so as not to reveal pending plans and operations to enemy listeners. Many senior commanders who telephoned often forgot the warning, only to be reminded by a censor interrupting their call. In a March 1942 memo to Churchill’s senior staff, the Cabinet Secretary, Sir Edward Bridges, warned of the dangers of unguarded reliance on transatlantic radiotelephone service.³

The fame of Bletchley Park obscures the fact that the enemy was listening, too, and also had its technical wizards. Germany’s Deutsche Reichpost had established a listening station on the Dutch coast by early 1941. German Intelliegence developed technical means to detect Allied transatlantic signals, and eventually could decode their content as they were being sent. Captured records at the end of the war showed just how good the enemy listening activity was: The Germans had recorded complete transcripts of many calls between senior Allied military figures, and even some between Churchill and Roosevelt.⁴

The telephone nevertheless remained in use throughout the war because it was simply indispensable, being available at all hours. Censors closely monitored whatever anyone, Churchill included, might say on the line.

AT&T, understanding the limitations of its A-3 scrambler, began seeking a better system as early as 1936. Though many schemes had been patented, none provided absolute security. In New Jersey, the Bell Laboratories research team, headed by A.B. Clark, focused on developing a way to turn voice signals into digital data—this at a time when digital technology was more theoretical than practical.

Homer Dudley, a Bell research physicist, eventually created a “Vocoder” (voice coder) device to convert analog voice sounds into digital signals while preserving “some” voice quality. An early prototype was demonstrated at the 1939 World’s Fair. Research on a viable transmission system was initiated as “Project X” in 1940.⁵

Shortly after the Japanese attack on Pearl Harbor and America’s entry into the war, Bell technicians demonstrated their system to the U.S. Army Signal Corps. The Army issued a contract to build the two trial devices. The first prototype was completed in August 1942.⁶ Its manufacturing priority was high and security was extremely tight. Indeed, only after high-level negotiations was Bletchley Park computer authority Dr. Alan Turing allowed even to see the device, and report his evaluation to Churchill in London.⁷ As confidence in its ability grew, six terminals had been ordered by late 1943.

The Sigsaly System

To operate this complex system, the U.S. Army 805th Signal Service Company was formed and intensively trained by AT&T personnel. The 805th comprised eighty-one officers, mostly lieutenants and captains, and 275 technical and master sergeants. Eventually, teams of five officers and ten enlisted men (many of whom had worked for the Bell System before the war) were assigned to keep each of the system terminals operating “full-time” (usually eight hours a day, since the devices required rigid maintenance and testing the rest of the time). Security rules stated that only American personnel have access to the equipment—a qualification that gave the British pause, but didn’t interfere with operations.⁸

Some time in late 1942, the code name SIGSALY was attached to the project. Not an acronym, though it resembled one, it was simply a cover, the SIG being common in Army Signal Corps terminology. Operators called prototypes “Green Hornets” (after a popular American radio melodrama), for the buzzing sound heard by anyone attempting to eavesdrop on the conversation. After the war it was learned that the Germans had recorded SIGSALY transmissions, but were unable to decode the signals.

As fully developed, the analog vacuum-tube powered SIGSALY equipment was huge—some forty racks of vacuum tube-powered electrical equipment weighing about fifty-five tons, taking up 2500 square feet and requiring 30,000 watts of power. Because of the many heat-producing tubes and circuits, each terminal was air conditioned—a rare service in World War II.⁹

SIGSALY equipment for London was shipped on HMT Queen Elizabeth in May 1943. It was far too bulky to be squeezed into the offices housing Churchill’s Downing Street Annexe above the Cabinet War Rooms, let alone the cramped bunker below ground. Space was found about a mile away in the basement of Selfridge’s Annexe on Oxford Street, where the Americans had set up a military communications center. An underground cable linked the two sites.

General Sir Hastings Ismay, the PM’s chief of staff, kept Churchill informed on the system’s progress, including several early test failures as operators came up to speed with the complex device. It was highly secret. Fewer than twenty-five of the most senior British civilian and military leaders were cleared to use the London terminal.¹⁰ Installation and testing of SIGSALY in Washington, and at Eisenhower’s headquarters in Algiers, also began in the spring of 1943.

SIGSALY entered service in mid-July 1943 with a military teleconference between London and Washington.¹¹ The latter had extensions to the White House (Roosevelt had decided against installing the main terminal there, knowing Churchill’s working hours) and the Navy building on Constitution Avenue. Heavy traffic finally required a second SIGSALY terminal at the Pentagon. As the Allies gained ground against Germany, SIGSALY terminals were also placed in Paris, Frankfurt and, after V-E day, Berlin.

In the Pacific, the system was used by 1944 with terminals in Oakland, Honolulu and Brisbane (Gen. MacArthur’s initial headquarters in Australia). After their recapture, systems were set up in Guam (to control B-29 raids, including the atomic bomb missions), Manila (MacArthur’s final headquarters), and finally Tokyo. One terminal was even aboard a ship, following MacArthur’s island-hopping campaign. All told, a dozen SIGSALY sites supported 3000 top-secret teleconferences, chiefly among military commanders.

Churchill himself appears not to have used SIGSALY until April 1944, and thereafter only very occasionally, tending to prefer the old-fashioned scrambler, with all its potential risk.¹² Ruth Ive, one of the high-level telephone monitors who recently published a book on her experiences, suggests several reasons for this:

Aside from the need to enter the War Rooms in the afternoon to use the dedicated telephone room (there was no link to Downing Street) Ive believes that Roosevelt did not want to be placed “on the spot,” with no time to consider issues or problems. She also reports that other than Churchill and Eden, no civilian War Cabinet members or Ministry personnel were authorized. Churchill did use SIGSALY to place his first call to Truman after Roosevelt’s death in April 1945. Two weeks later, they had a two-hour discussion concerning tentative German surrender offers— the longest call ever made over SIGSALY equipment.¹³

Aftermath

As the Allies neared victory, the heavy SIGSALY terminals were gradually removed from service and returned to the U.S., starting with the Algiers equipment in 1944, as Eisenhower had shifted his headquarters to Britain. The London facility was “recovered” on 31 October 1945.¹⁴ Since the technology was still unknown to any other nations, the system remained highly secret for three more decades (two years, indeed, longer than Bletchley Park’s codebreaking role).

SIGSALY was first made public in news accounts in mid-1976, when more than thirty Bell Labs patents (many applied for between 1941 and 1945) were finally granted.¹⁵ The system is recognized today as a technical pioneer for its initial use of numerous techniques which are widely used in telecommunications.¹⁶ When the Cabinet War Rooms were being prepared for public tours in the 1980s, Bell Labs provided detailed advice on the SIGSALY telephone set that appears today in the underground room Churchill used in his occasional trans-Atlantic talks with Roosevelt.¹⁷

Despite continuing secrecy, Bell Labs won several contemporary awards for its work, including “Best Signal Processing Technology” in 1946. Since SIGSALY was still classified, attendees at the awards ceremony simply had to accept the verdict of the judges that this was something of crucial importance. A.B. Clark, a key player with Bell Labs and by then Director of Research and Development at the National Security Agency, delivered his acceptance speech over a coded phone line: “Phrt fdygui jfsowria meeqm wuiosn jxolwps fuekswusjnvkci! Thank you!”

Endnotes

1. J. V. Boone and R. R. Peterson. SIGSALY—The Start of the Digital Revolution (Ft. Meade, Maryland: NSA/CSA Center for Cryptologic History, 2000). Refer to www. nsa/gov.

2. “Secure Speech Transmission,” in M. D. Fagen, ed., A History of Engineering and Science in the Bell System: National Service in War and Peace 1925-1975 (Chicago: Bell Telephone Laboratories, 1978), 292.

3. Donald Mehl, SIGSALY: The Green Hornet—The World War II Unbreakable Code for Secret High-Level Telephone Conferences (Kansas City: privately published, 1997), 17. Mehl was one of the Signal Corps officers who worked with SIGSALY during the war.

4. Mehl, 15-16. See also Ruth Ive, The Woman Who Censored Churchill (Stroud: History Press, 2008), 93-100; and “Ruth Ive’s Book,” Finest Hour 141, Winter 2008-09, 7.

5. Fagen, 291-312.

6. Fagen, 310.

7. Mehl, 66-71, quotes many of the relevant letters in this controversy. Ive, 109, describes the Turing visit, but incorrectly identifies him as the “director” of Bletchley Park, which he never was.

8. Patrick Weadon, “The SIGSALY Story,” information sheet (Ft. Meade, Maryland: National Cryptologic Museum, n.d.)

9. “Signal Corps Fixed Communications in World War II: Special Assignments and Techniques. Washington: Signal Corps Historical Section (Project E-10), December 1945,” 33. A copy of this once-secret study is in the AT&T/Bell Labs Archives. Thanks to Dr. James Spurlock for this source.

10. The location, access, and use issues were all controversial: see Mehl, 72-77.

11. Boone and Petrson, op. cit. See note 1.

12. Peter Simpkins, Cabinet War Rooms (London: Imperial War Museum, 1983), 58.

13. Ive, 114, note 4.

14. “Signal Corps Fixed…” Appendix A.

15. “Patents issued on WW II Speech Encoding Technique,” in Bell Labs News, 16:28:1, 12 July 1976. A list of the patents involved is in Fagen, 297.

16. Mehl, 51-53 cites several other sources. See also the list of firsts in Boone & Patterson, 3-4.

17. T. W. Thatcher, “Notes: Description of SIGSALY Telephone Set and Its Connections,” 28 February 1984. Copy in AT&T/Bell Labs Archives. Thanks again to Dr. James Spurlock.

18. Paul D. Lehrman, “Award Winners from the Dawn of Modern Audio” in Mix, Professional Audio and Music Production, 1 August 2006 (http:// mixonline.com).