The Craft of Intelligence

by Allen W. Dulles

  In modern terminology, the word “crypt,” meaning “something hidden,” conveniently gets around the distinction between codes and ciphers since it refers to all methods of transforming “plain text” or “clear text” into symbols. The over-all term for the whole field today is “cryptology.” Under this broad heading we have two distinct areas. Cryptography has to do with making, devising, inventing or protecting codes and ciphers for the use of one’s own government. Cryptanalysis, on the other hand, has to do with breaking codes and ciphers or “decrypting” them, with translating someone else’s intercepted messages into proper language. To put one’s own messages into a code or cipher is to “encrypt” them. However, when we translate our own messages back into plain language, we are “deciphering.”

  A cryptogram or cryptograph would be any message in code or cipher. “Communications intelligence” is information which has been gained through successful cryptanalysis of other people’s traffic. And now, having confused the reader completely, we can get to the gist of the matter.

  The diplomatic service, the armed services and the intelligence service of every country use secret codes and ciphers for classified and urgent long-distance communications. Transmission may be via commercial cable or radio or over special circuits set up by governments. Anyone can listen in to radio traffic. Also, governments, at least in times of crisis, can usually get copies of the encrypted messages that foreign diplomats stationed on their territory send home via commercial cable facilities. The problem is to break the codes and ciphers, to “decrypt” them.

  Certain codes and ciphers can be broken by mathematical analysis of intercepted traffic, i.e., cryptanalysis, or more dramatically and simply by obtaining copies of codes or code books or information on cipher machines being used by an opponent, or by a combination of these methods.

  In the earlier days of our diplomatic service, up to World War I, the matter of codes was sometimes treated more or less cavalierly, often with unfortunate results. I remember a story told me as a warning lesson when I was a young foreign service officer. In the quiet days of 1913, we had as our Minister in Rumania an estimable politician who had served his party well in the Midwest. His reward was to be sent as Minister to Bucharest. He was new to the game and codes and ciphers meant little to him. At that time our basic system was based on a book code, which I will call the Pink Code, although that was not the color we then chose for its name. I spent thousands of worried hours over this book, which I have not seen for over forty years, but to this day I can still remember that we had six or seven words for “period.” One was “PIVIR” and another was “NINUD.” The other four or five I do not recall. The theory than was—and it was a naïve one—that if we had six or seven words it would confuse the enemy as to where we began and ended our sentences.

  In any event, our Minister to Rumania started off from Washington with the Pink Code in a great, sealed envelope and it safely reached Bucharest. It was supposed to be lodged in the legation’s one safe. However, handling safe combinations was not the new Minister’s forte, and he soon found it more convenient to put the code under his mattress, where it rested happily for some months. One day it disappeared—the whole code book and the Minister’s only code book. It is believed that it found its way to Petrograd.

  The new Minister was in a great quandary, which, as a politician, he solved with considerable ingenuity. The coded cable traffic to Bucharest in those days was relatively light and mostly concerned the question of immigrants to the United States from Rumania and Bessarabia. So when the new Minister had collected a half-dozen coded messages, he would get on the train to Vienna, where he would quickly visit our Ambassador. In the course of conversation, the visitor from Bucharest would casually remark that just as he was leaving he had received some messages which he had not had time to decode and could he borrow the Ambassador’s Pink Code. (In those good old days, we sent the same code books to almost all of our diplomatic missions.) The Minister to Bucharest would then decipher his messages, prepare and code appropriate replies, take the train back to Bucharest and, at staged intervals, send off the coded replies. For a time everything went smoothly. The secret of the loss of the code book was protected until August, 1914, brought a flood of messages from Washington as the dramatic events leading up to World War I unrolled. The Minister’s predicament was tragic—trips to Vienna no longer sufficed. He admitted his dereliction and returned to American politics.

  The uncontrollable accidents and disasters of war sometimes expose to one opponent cryptographic materials used by the other. A headquarters or an outpost may be overrun and in the heat of retreat code books left behind. Many notable instances of this kind in World War I gave the British a lifesaving insight into the military and diplomatic intentions of the Germans. Early in the war the Russians sank the German cruiser Magdeburg and rescued from the arms of a drowned sailor the German naval code book, which was promptly turned over to their British allies. British salvage operations on sunken German submarines turned up similar findings. In 1917 two German dirigibles, returning from a raid over England, ran into a storm and were downed over France. Among the materials retrieved from them were coded maps and code books used by German U-boats in the Atlantic.

  An American naval exploit which took place toward the end of World War II has given us an even more thrilling story of the capture of enemy code and cipher material. This was the result of a carefully laid plan and not of a lucky accident. A German submarine, the U-505, was captured, intact, on June 4, 1944, off the coast of French West Africa by units of the United States Navy under the command of Rear Admiral Daniel V. Gallery.

  During World War II, Allied action resulted in the destruction of over seven hundred German U-boats. The U-505, which now reposes in the Museum of Science and Industry, Chicago, was the only one that was brought back afloat and in one piece. It had been the consistent practice of the German U-boat crews whose subs were forced to surface and surrender to insure that the submarine would sink as the crew abandoned ship. In this instance, however, as the result of skillful preparation, a boarding party from Admiral Gallery’s task force managed to get abroad the U-505 just as its own crew was abandoning it after having set its valves for scuttling. At the risk of their lives and not knowing how many seconds they had before the submarine would take its final plunge, some ten men from the American naval boarding crew charged down the hatch and closed the scuttling valves just in the nick of time. Their escape was later aided by a German sailor. He had jumped overboard and was swimming near the sinking German sub when a member of the boarding crew hauled him aboard again and got him to disclose the workings of a conning tower hatch which was on the escape route of the Americans who had gone below. As they threw him back into the water, it was with a heartfelt “Thanks, bud,” but rescue was at hand for him and the other German crew members.

  All the records and files and technical equipment aboard the sub, including its codes and ciphers, were rescued, and the submarine was safely towed to Bermuda.

  But this was not the end of the story. If the Nazis had learned that the submarine had not been scuttled or destroyed before capture, they would have been alerted to the probable seizure of the code and cipher material aboard and would never again have used them. Obviously several thousand American naval personnel, from the beginning to the end of the operation of capture and of towing, knew the facts, and for many this was their great story of the war. The problem of impressing upon all these sailors the importance of keeping the capture secret was a bigger task even than capturing the submarine itself. But this was done with success. The Germans believed that the submarine had gone to its watery grave, carrying with it the secrets which in fact proved very useful to us.2

  2An account of this naval exploit appears in Daniel V. Gallery, Twenty Billion Tons Under the Sea (Chicago: Henry Regnery Co., 1954).

  Military operations based on breaking of codes will often t
ip off the enemy, however. When, during World War I, the Germans noticed that their submarines were being cornered with startling frequency, it was not hard for them to guess that communications with their underwater fleet were being read. As a result, all codes were immediately changed. There is always the problem, then, of how to act on information derived in this manner. One can risk terminating the usefulness of the source in order to obtain an immediate military or diplomatic gain, or one can hold back and continue to accumulate an ever-broadening knowledge of the enemy’s movements and actions in order eventually to inflict the greatest possible damage.

  Actually, in either case, the attempt is usually made to protect the real source and keep it viable, by giving the enemy fake indications that some other kind of source was responsible for the information acquired. Sometimes an operation that could damage the adversary is not undertaken if it would alert the enemy to the fact that its origin was solely due to information obtained by reading his messages.

  During World War I, the first serious American cryptanalytic undertaking was launched under the aegis of the War Department. Officially known as Section 8 of Military Intelligence, it liked to call itself the “Black Chamber,” the name used for centuries by the secret organs of postal censorship of the major European nations. Working from scratch, a group of brilliant amateurs under the direction of Herbert Yardley, a former telegraph operator, had by 1918 become a first-rate professional outfit. One of its outstanding achievements after World War I was the breaking of the Japanese diplomatic codes. During negotiations at the Washington Disarmament Conference in 1921, the United States wanted very much to get Japanese agreement to a 10:6 naval ratio. The Japanese came to the conference with the stated intention of holding to a 10:7 ratio. In diplomacy, as in any kind of bargaining, you are at a tremendous advantage if you know your opponent is prepared to retreat to secondary positions if necessary. Decipherment of the Japanese diplomatic traffic between Washington and Tokyo by the Black Chamber revealed to our government that the Japanese were actually ready to back down to the desired ratio if we forced the issue. So we were able to force it without risking a breakup of the conference over the issue.

  The “Black Chamber” remained intact, serving chiefly the State Department, until 1929, when Secretary Stimson refused to let the department avail itself further of its services. McGeorge Bundy, Stimson’s biographer, provides this explanation:

  Stimson adopted as his guide in foreign policy a principle he always tried to follow in personal relations—the principle that the way to make men trustworthy is to trust them. In this spirit he made one decision for which he was later severely criticized: he closed down the so-called Black Chamber. . . . This act he never regretted . . . . Stimson, as Secretary of State, was dealing as a gentleman with the gentlemen sent as ambassadors and ministers from friendly nations.3

  3Henry L. Stimson and McGeorge Bundy, On Active Service in Peace and War (New York: Harper & Brothers, 1948).

  Our Army and Navy had, fortunately, continued to address themselves to the problems of cryptanalysis with particular emphasis on Japan, since American military thinking at that time foresaw Japan as the major potential foe of the United States in whatever war was to come next. By 1941, the year of Pearl Harbor, our cryptanalysts had broken most of the important Japanese naval and diplomatic codes and ciphers; and we were, as a result, frequently in possession of evidence of imminent Japanese action in the Pacific before it took place.

  The Battle of Midway in June, 1942, the turning point of the naval war in the Pacific, was an engagement we sought because we were able to learn from decrypted messages that a major task force of the Imperial Japanese Navy was gathering off Midway. This intelligence concerning strength and disposition of enemy forces gave our Navy the advantage of surprise.

  A special problem, in the years following Pearl Harbor, was how to keep secret the fact that we had broken the Japanese codes. Investigations, recriminations, the need to place the blame somewhere for the disheartening American losses threatened to throw this “Magic,” as it was called, into the lap of the public, and the Japanese. Until an adequate Navy could be put on the seas, the ability to read Japanese messages was one of the few advantages we had in the battle with Japan. There were occasional leaks but none evidently ever came to their attention.

  In 1944, Thomas E. Dewey, who was then running for President against President Roosevelt, had learned, as had many persons close to the federal government, about our successes with the Japanese code and our apparent failure before Pearl Harbor to make the best use of the information in our hands. It was feared that he might refer to this in his campaign. The mere possibility sent shivers down the spines of our Joint Chiefs of Staff. General Marshall himself then wrote a personal letter to Mr. Dewey, telling him that the Japanese still did not know we had broken their codes and that we were achieving military successes as a result of our interception and decoding of their messages. Mr. Dewey never mentioned our code successes. The secret was kept.

  One of the most spectacular of all coups in the field of communications intelligence was the British decipherment of the so-called Zimmermann telegram in January, 1917, when the United States was on the brink of World War I.4 The job was performed by the experts of “Room 40,” as British naval cryptanalytic headquarters were called. The message had originated with the German Foreign Secretary Zimmermann in Berlin and was addressed to the German Minister in Mexico City. It outlined the German plan for the resumption of unrestricted submarine warfare on February 1, 1917, stated the probability that this would bring the United States into war, and proposed that Mexico enter the war on Germany’s side and with victory regain its “lost territory in Texas, New Mexico, and Arizona.”

  4This story has been well told in Barbara Tuchman’s book. See Bibliography.

  Admiral Hall, the legendary Chief of British Naval Intelligence, had this message in his hands for over a month after its receipt. His problem was how to pass its decrypted contents to the Americans in a manner that would convince them of its authenticity yet would prevent the Germans from learning the British had broken their codes. Finally, the war situation caused Lord Balfour, the British Foreign Secretary, to communicate the Zimmermann message formally to the American Ambassador in London. The receipt of the message in Washington caused a sensation at the White House and State Department, and created serious problems for our government—how to verify beyond a doubt the validity of the message and how to make it public without letting it seem merely an Anglo-American ploy to get the United States into the War. My uncle, Robert Lansing, who was then Secretary of State, later told me about the dramatic events of the next few days which brought America close to war.

  The situation was complicated by the fact that the Germans had used American diplomatic cable facilities to transmit the message to their Ambassador in Washington, Count Bernstorff. He relayed it to his colleague in Mexico City. President Wilson had granted the Germans the privilege of utilizing our communication lines between Europe and America on the understanding that the messages would be related to peace proposals in which Wilson was interested.

  The President’s chagrin was therefore all the greater when he discovered to what end the Germans had been exploiting his good offices. However, this curious arrangement turned out to be of great advantage. First of all, it meant that the State Department had in its possession a copy of the encrypted Zimmermann telegram, which it had passed to Bernstorff, unaware, of course, of its inflammatory contents. Once the encrypted text was identified, it was forwarded to our embassy in London, where one of Admiral Hall’s men redecrypted it for us in the presence of an embassy representative, thus verifying beyond a doubt its true contents. Secondly, the fact that deciphered copies of the telegram had been seen by German diplomats in both Washington and Mexico City helped significantly to solve the all-important problem that had caused Admiral Hall so much worry, namely, how to fool
the Germans about the real source from which we had obtained the information. In the end the impression given the Germans was that the message had leaked as a result of some carelessness or theft in one of the German embassies or Mexican offices which had received copies of it. They continued using the same codes, thus displaying a remarkable but welcome lack of imagination. On March 1, 1917, the State Department released the contents of the telegram through the Associated Press. It hit the American public like a bombshell. In April we declared war on Germany.

  When one compares the cryptographic systems used today with those to which governments during World War I entrusted the passage of their most vital and sensitive secrets, the latter seem crude and amateurish, especially because of their recurring groups of symbols which tipped off the cryptanalyst that an important word or one in frequent usage must lie behind the symbols. When Admiral Hall’s cryptanalysts saw the combination “67893” in the Zimmermann telegram, they recognized it and knew that it meant “Mexico.” Under the German system it always meant that. Today such a cipher group would never stand for the same word twice.

  Today not only all official government messages but also the communications of espionage agents are cast in equally secure and complex cryptographic systems. Soviet agents, for example, in reporting information back to Moscow, use highly sophisticated cipher systems. Here as elsewhere, as defensive measures improve, countermeasures to pierce the new defenses also improve.

  6

  Planning and Guidance

  The matters that interest an intelligence service are so numerous and diverse that some order must be established in the process of collecting information. This is logically the responsibility of the intelligence headquarters. It alone has the world picture and knows what the requirements of our government are from day to day and month to month.