Top NSA Banner

Articles:

1. The American Revolution's One-Man National Security Agency
2. William F. Friedman
3. General Patton and COMINT
4. Three African-American Cryptologic Pioneers

The American Revolution's One-Man National Security Agency

There was an American who seems to have been the American Revolution's one-man National Security Agency, for he was the one and only cryptologic expert Congress had, and, it is claimed, he managed to decipher nearly all, if not all, of the British code messages obtained in one way or another by the Americans. Of course, the primary way in which enemy messages could be obtained in those days was to capture couriers, knock them out or knock them off, and take the messages from them. This was very rough stuff, compared to getting the material by radio intercept, as we do nowadays.

This one-man NSA was James Lovell, and besides being a self-trained cryptologist, he was also a member of the Continental Congress. With his cipher designs, Lovell became America's first cryptographic tutor. Unfortunately, his students, the American ministers abroad, though brilliant and talented in political matters, found his systems confusing and frustrating.

James Lovell, born in 1737, studied at Harvard, taught in his father's school in Boston, and became a famous orator. Arrested by the British after the battle of Breed's Hill, he was sent as a prisoner to Halifax in 1776, but soon thereafter he was exchanged and he returned to Boston. Chosen as a delegate to the Continental Congress, he attended the sessions of the Congress beginning in February 1777 and served continuously until the end of January 1782 when he took his only leave. In May 1777 he was appointed to the Committee for Foreign Affairs, where, among other responsibilities, he deciphered dispatches. He became the Committee's most indefatigable member, indeed, sometimes its only active member. Other members arrived and departed, but Lovell stayed on and for five years never visited his wife and children. Before he left Congress in 1782, Lovell had left his mark on American foreign relations and particularly on cryptography.

James Lovell enjoyed the challenge of making and breaking cipher systems. Unfortunately, even learned diplomats of his time had great difficulty understanding his cipher forms completely. For John Adams, the Lovell ciphers caused boundless confusion. As Adams confided in a letter to Francis Dana in Paris in March 1781: "I have letters from the president and from Lovell, the last unintelligible, in ciphers, but inexplicable by his own cipher." In short, Adams could not read Lovell's enciphered dispatches. However, John Adams was not the only diplomat troubled by Lovell's ciphers. In February 1780 Lovell wrote to Benjamin Franklin that the Chevalier de la Luzerne, who had become French minister to the United States the previous year, was anxious because Lovell and Franklin were not corresponding in cipher. Lovell had sent a cipher earlier, but Franklin ignored it. Lovell tried again. In March 1781 Franklin wrote to Francis Dana enclosing a copy of Lovell's new cipher and a paragraph of Lovell's letter in which the cipher was used. Somewhat bewildered, Franklin, accustomed to a simpler cipher, commented: "If you can find the key & decypher it, I shall be glad, having myself try'd in vain."

Lovell's considerable talents for breaking ciphers rewarded Nathaniel Greene and George Washington when enciphered dispatches from the British commander, Lord Cornwallis, were intercepted in 1780 and 1781. Lovell wrote to Washington that he believed the British ciphers were quite widely used among their leaders and urged the general to have his secretary make a copy of the cipher key that he was transmitting to Greene. Interestingly enough, Lovell had discovered a curious weakness in the British cryptographic system: "the Enemy make only such changes in their Cypher, when they meet with misfortunes, as makes a difference of Position only to the same Alphabet." What Lovell meant was that the same mixed cipher alphabet was merely shifted to another juxtaposition with the plain alphabet.

Lovell got his opportunity to break a critical British dispatch through good fortune. Sir Henry Clinton, commander of the British forces in America, sent an enciphered dispatch via a special courier to Cornwallis. The dispatch explained Clinton's inability to assist Cornwallis with a fleet at Yorktown until a specific day and urged him to hold out. Beached near Egg Harbor, the crew and courier were captured and brought to Philadelphia. It was learned that the courier had hidden the confidential dispatch under a large stone near the shore. Recovered, the dispatch was found to be written in three systems. It took Lovell two days to solve and read the dispatch. The original letter was then sent on to Cornwallis to enable the Americans to use their secret knowledge of the British plans and to counteract them.

James Lovell's secret ciphers produced more confusion than security for American diplomats during the revolution. Only gradually in the years after 1775 did American officials become sophisticated about cryptographic systems. Because of the frustration with ciphers, American statesmen began to rely more heavily upon codes rather than ciphers for secret foreign communications. All of the confusion over the Lovell ciphers provides a remarkable lesson for cipher inventors. Lovell tried to force his system on the best minds of the country--even they didn't understand it, and the system failed.

(This article was adapted from Ralph Weber, _Masked Dispatches: Cryptograms and Cryptology in American History, 1775-1900_, and _The Friedman Legacy: A Tribute to William and Elizebeth Friedman_. Both publications are available from the Center for Cryptologic History)

Back to Top

---

William F. Friedman

Once, when asked how he happened to become the father of American cryptology, William F. Friedman smiled and said, "I was seduced." This dapper little man with the mustache was a writer, teacher, inventor, and above all a cryptologist. Known as Mr. Friedman to his face, "Uncle Willie" behind his back, William Friedman was a leader in American cryptology for half a century and still significantly shapes the business today, over thirty years after his death.

Wolfe Frederick Friedman was born on 24 September 1891 in Kishinev, then part of imperial Russia, now Chisinau, capital of Moldova. His father, an interpreter for the Czar's postal service, emigrated to the United States the following year to escape increasing anti-Semitic regulations; the family joined him in Pittsburgh in 1893. Three years after that, when the elder Friedman became a U.S. citizen, Wolfe's name was changed to William.

After receiving a B.S. and doing some graduate work in genetics at Cornell University, William Friedman was recruited by "Colonel" George Fabyan for work in his department of genetics at Riverbank Laboratories, what would today be termed a "think tank," outside Chicago. Fabyan was an "honorary colonel" and a millionaire who could indulge his interest in science and industry.

In addition to the work being done in genetics and other industrial and agricultural topics, there was a cipher department at Riverbank studying the "Baconian Cipher." (Colonel Fabyan was convinced that Francis Bacon, rather than William Shakespeare, had written the plays credited to the latter.) Friedman became interested in the study of codes and ciphers, thanks to his concurrent interest in Elizebeth (sic) Smith, who was doing some of the "Baconian Cipher" research. When Riverbank was asked to train the military in the use of codes, Friedman was assigned as the principal instructor. As preparation, he made a thorough study of historical and present-day cryptology. Friedman served as a lieutenant in G6A2, the crypt unit of the American Expeditionary Forces (AEF) during World War I.

He returned to Riverbank until 1920, when he went to Washington, DC, to work in the Office of the Chief Signal Officer where, in 1922, he was made head of the Codes and Compilation Section. In 1929 he was selected to be the head of the newly organized Signal Intelligence Service (SIS). Throughout the early and middle 1930s, he created the organizational foundations of a cryptologic structure which evolved into the Army Security Agency (ASA) in World War II. In the process, he led the transition from paper and pencil cryptology into the modern era characterized by the application of machines to both cryptography and cryptanalysis.

The SIS/ASA is primarily famous as the group that "broke PURPLE"-- PURPLE was the American code name for a Japanese diplomatic cryptosystem. A distinguished team trained and led by Friedman was responsible for this, one of the major cryptologic breakthroughs of World War II, commensurate with the Polish solution of the German Enigma cipher. However, they didn't concentrate solely on breaking cryptosystems. Friedman and the members of the team also cooperated with Navy cryptographers to develop the most secure cipher machine of the World War, SIGABA. The major players on the team-- Rowlett, Sinkov, Kullback, Snyder, and Rosen--went on to be major influences in the development of the Armed Forces Security Agency (AFSA) and NSA.

After the war, Friedman served as director, Communications Research (1947-1949); cryptologic consultant, Armed Forces Security Agency (1949-1951); research consultant, NSA (1952-1954); special assistant to the director, NSA (1954-1955); member of the NSA Scientific Advisory Board (1954-1960); and special consultant to NSA (1955-1969).

His services did not go unrecognized. He received the War Department Medal for Exceptional Civilian Service in 1944, the Presidential Medal for Merit in 1946, the Presidential National Security Medal in 1955, and a congressional award of $100,000 for inventions and patents in the field of cryptology held secret by the government.

Perhaps Friedman's greatest achievements were introducing mathematical and scientific method into cryptology and producing training materials used by several generations of pupils. His work affected for the better both signals intelligence and information systems security, and much of what is done today at NSA may be traced to William Friedman's pioneering efforts.

William F. Friedman died in November 1969 of a heart attack. He is buried at Arlington National Cemetery.

Back to Top

---

General Patton and COMINT

General George S. Patton was an enigmatic and fascinating figure, one of the great commanders of World War II. His character and leadership remain the stuff of legends, still the subject of controversy. Many of his remarks, whether sanitized or in their original salty version, are quoted even today. The motion picture based on his life, flawed but reasonably accurate for a Hollywood production, continues as a popular feature in video.

It is interesting but not surprising to find that Patton became an astute consumer of communications intelligence. It is unclear whether he had much knowledge of COMINT or exposure to it during the North African or Sicilian campaigns, but Patton learned its worth in the drive across Western Europe after the D-Day landings and, once having learned COMINT's value, used it well.

In mid-August 1944, Major Warrack Wallace was assigned to Patton's Third U.S. Army as an assistant ULTRA recipient. ULTRA was the codeword assigned to COMINT derived from decryption of high-level German ciphers, principally from messages enciphered on the Enigma machine. Cryptanalysis was done at British code and cipher headquarters at Bletchley Park, outside London; items of importance were disseminated to "Special Liaison Units" (SLUs) in the field. The SLUs gave the material to G-2 (intelligence) ULTRA recipients, such as Major Wallace, who worked in effect as the COMINT G-2.

When Major Wallace arrived as assistant to a Major Helfers, the Third Army was "under canvas" about ten miles north of the French city of Le Mans. The ULTRA recipients had a tent which served as living quarters and as an office; there were no electric lights, no telephone, and no organic transportation. The ULTRA recipients would get collateral information from the G-2 and then "receipt" ULTRA data from the SLU. (ULTRA material was held no longer than twenty-four hours; it was turned back to the SLU "under counter-receipt.")

Every morning at 0900, General Patton and about forty officers would attend a meeting in the War Tent; this consisted of regular briefings by the G-2 and G-3 (operations), situation reports, plus a news report from radio broadcasts. At the conclusion of this meeting, the chief of staff would excuse all but seven officers, who were to remain for a "special briefing." Then either Major Helfers or Major Wallace would spread the ULTRA map over the regular war map and brief Patton and his senior staff on the enemy situation as seen in COMINT.

COMINT began to prove out for the Third Army. When ULTRA and ULTRA alone correctly predicted a drive by five German Panzer divisions against the Third Army at Avaranches, Third Army officers began to pay more attention to this source of intelligence. In addition, the Third Army G-2, Colonel Koch, arranged for the ULTRA recipients to get electric light and telephone installations at each new Army headquarters, and he arranged transportation for them at each move. When priority ULTRA messages arrived, Colonel Koch saw to it that Major Helfers or Major Wallace took it directly to the officer most directly involved, sometimes to General Patton himself. Whenever circumstances required Patton to be absent from the special briefing, he arranged for a special briefing in his caravan by the ULTRA officers later in the day.

When Third Army headquarters moved near the French city of Chalons an ULTRA message arrived at 0100 showing the German order for an attack at 0300. Patton had described the U.S. troops in the attack areas as spread out as "thin as the skin on an egg." He found means to alert the defending divisions without jeopardizing the security of ULTRA, and the German attack was successfully repulsed.

ULTRA provided extremely accurate order of battle information, often having exact figures down to the man and the gun for German units facing the Third Army. On two occasions, the regular G-2 staff placed German divisions on the line which actually were in Italy. In both cases, there had been no ULTRA messages indicating any move from Italy; in both cases, in a week or two, the G-2 corrected its mistakes by noting that the information came from POWs who had strayed to France from their former units in Italy.

There were, however, some special hazards for ULTRA officers serving with the Third Army. Major Wallace commented that General Patton's famous bull terrier, Willie, once paused in his operations against French dogs to show his contempt for intelligence by raising his leg against one of the ULTRA officer's best maps.

At the time, many believed ULTRA was applicable primarily in strategic operations and could be useful tactically only in a static situation. Major Wallace called this ridiculous. He found that Third Army armor and infantry elements were spread out over a wide area of France even as the main body of the Army was driving eastward. He wrote: "An army has never moved as fast and as far as the Third Army in its drive across France, and ULTRA was invaluable every mile of the way."

(Based upon a contemporary report written by Major Warrack Wallace)

Back to Top

---

Three African-American Cryptologic Pioneers

From 1939 to 1956 African-Americans employed by NSA and its predecessors were segregated into primarily support elements, a reflection of U.S. Army policies and the social tenor of the period. Even though President Truman issued two executive orders in 1948 setting out a policy of nondiscrimination in both the armed services and the federal government, several areas, such as the Russian plaintext traffic processing unit, remained all black into the 1950s. In addition, African-Americans for the most part held lower grade jobs in the machine division of that unit and others.

There were occasional exceptions to this state of affairs. In 1948 the Research and Development organization at Arlington Hall Station hired the agency's first black engineer, Carroll Robinson. Mr. Robinson was involved in the design and development of the agency's first in-house-developed digital computer, Abner I.

Two other African-Americans were hired by the Research and Development organization. Mitchell Brown and Charles Matthews were graduates of Hilltop Radio-Electronics Institute, a black-owned school in Washington, DC. They were brought on as engineering technicians and worked alongside their white counterparts.

Apparently the environment for African-Americans in the R&D organization was relatively positive and conducive to professional growth. Carroll Robinson became NSA's first African-American senior executive, eventually reaching the position of office chief. Brown and Matthews also gained positions of significant responsibility: Mr. Brown eventually became test director of the Digital Voice Processor Consortium Test program, which led to the selection of equipment for the secure telephone unit (STU) II. Mr. Matthews served as a project engineer on Abner I and later on Solo, the Agency's first transistorized computer. He also held several supervisory and middle-management positions before he retired in 1988.

All this is not to say the R&D organization was without its racial problems. Mitchell Brown pointed out in an interview that white engineering technicians with less experience or fewer qualifications were hired at higher grades. Moreover, Carroll Robinson noted that for many years African-Americans were not sent on overseas assignments because of the view that they would not be welcome in the host country.

Nevertheless, the R&D organization in the late 1940s was a bright spot in the employment of African-Americans in the cryptologic business.

(For the full story of the African-American experience in cryptology, read The Invisible Cryptologists: African-Americans, WWII to 1956, published in 2001 by the Center for Cryptologic History.)

Back to Top