by DAVID KAHN
TO GILBERT [codename for Trepper] from Director. Check whether Guderian [German panzer general] really at Eastern front. Are the 2d and 3d Armies under his command? …
TO GILBERT from Director. Report about 26 armored divisions being formed in France.
The intelligence came from informants throughout the entire Nazi regime. Schulze-Boysen himself was pivotally located in the Forschungsamt. Harnack held a high post in the Economics Ministry as an expert on the Soviet Union. The Rote Kapelle had sources in—among others—the Foreign Office, the Luftwaffe’s counterintelligence, the labor and propaganda ministries, and, in the person of young Horst Heilmann, an apostate from Naziism, in the cryptanalytic office of the Army. The arrhythmic monotone of its radio transmitters made beautiful music to Moscow. They sang to the Russians of German plans to encircle Leningrad instead of occupying it, of exact times of German parachute raids, of monthly aircraft production, of a Soviet code found at Petsamo in Finland, Luftwaffe losses, Luftwaffe production, capabilities of a new Messerschmidt fighter, production of synthetic fuel, foreign policy developments, political opposition to Naziism, troop movements along the Dnieper. The Russian bear performed to these tunes, knowing just where and how to claw and slash the Nazi forces.
The Funkabwehr monitors listened as this symphony reverberated through the ether. To them, it was cacophony. The cryptograms remained impervious. But the transmitters could be tracked down, and on June 30, 1942, another Belgian group, headed by a veteran Communist agent, Johann Wenzel, whose thorough knowledge of radio techniques had earned him the sobriquet “The Professor,” was raided in Brussels; Wenzel was nabbed in front of his set. The Gestapo took over, and what the most energetic mental thumping of an impersonal string of numbers had not done, a moderate physical truncheoning of human flesh did. The Professor’s wide acquaintance with Soviet spy communications soon had the Funkabwehr translating its file of back intercepts. In one of them, almost a year old, they read the true addresses of Schulze-Boysen and Harnack….
Of all Soviet networks during the war, by far the most important was the Swiss. It owed its supremacy in part to its location in neutral Switzerland, where it operated for a long time out of reach of the German Abwehr, and in part to having in the network the agent codenamed LUCY, whom many regarded as the greatest spy of the war. This was Rudolf Rössler, a small, quiet, bespectacled German publisher of leftist Catholic books whose codename came from his residence in Lucerne. His sources appear to have been ten World War I companions, all German officers, five of whom became generals and served at least part of the time in the Oberkommando der Wehrmacht. One was General Fritz Thiele, who had the O.K.W.’s Chif-frierabteilung under him and who, as No. 2 man in the O.K.W. signal organization, used its facilities to radio messages to Rössler. Thus he procured intelligence of exalted importance and precise accuracy with dazzling speed from the very heart of the German High Command itself.
Head of the ring was Alexander Rado, a cartographer whose maps appeared daily in the Swiss newspapers. A Hungarian Communist, he had been sent to Switzerland to set up a ring in 1936. Second in command and chief radioman was Alexander Foote, a bearlike, unperturbable Englishman in his mid-thirties who pretended to be living in Switzerland on independent funds to escape military service. Early in 1941 he installed in his flat at 2 Chemin de Longeraie, Lausanne, the radio station that was to be known as JIM. On March 12, after a thousand tappings of the call-sign FRX, he heard the Moscow Centre calling him through the crackle of static and the background noise of other transmissions: NDA NDA OK QRK 5, the latter meaning in the radio Q code that his signals were being heard very strongly.
The Swiss group produced results quickly. One night in the middle of June, Foote sent off to Moscow a brief but vital message:
DORA [codename for Rado] to Director, via Taylor [a courier]. Hitler has definitely set June 22 for attack on Russia.
This had no more effect on Stalin than had Sorge’s information, for Stalin evidently regarded the obvious community of interest between Hitler and himself in subduing England and dismembering her empire as outweighing information from two individuals. The incident spotlights one of the most difficult problems in the assessment of intelligence: credibility.
At first Foote contacted Moscow only twice a week. But when the Russo-German war broke out, he was informed that the Centre would be listening to him round the clock. Priorities were established: VYRDO for exceedingly urgent messages, RDO for urgent, and MSG for routine. Moscow was always in a hurry for information, and Foote, who operated as a sort of loner, transmitted most of the LUCY material and received the VYRDO messages pertaining to it. For two years his life fell into a routine that indicates the unglamorous ways of a spy. After a night of enciphering and radioing, he rose about 10, spent the morning maintaining his pose of emigré Englishman and the afternoon journeying to meet a cutout, or courier, at some unobserved place. “Having returned,” he wrote, “I usually had a long evening’s ciphering before me. According to the rules, all ciphering should have been done after dark and behind locked doors. But needs must when the Centre drove and in the more hectic times I was enciphering in all my spare moments.” During his active period, Foote sent 2,000 messages, or about six a day. They averaged 100 words each.
Contact with the Centre was made on a fixed wavelength. The Centre would reply on its fixed length. Both would then switch to another wavelength with a different call-sign for the evening’s work. “My transmission time was usually about one in the morning,” Foote wrote, “If conditions were good and the message short I was through in about a couple of hours. If, as frequently happened, I had long messages to send and atmospherics were bad I had to fight my way through and send when and as conditions allowed. Often on such occasions I was still at the transmitter at six and once or twice I ‘signed off’ at nine in the morning…. To be on the air for that length of time broke all the normal precautions against radio monitoring. But it was a chance which had to be taken if the intelligence was to be passed over, a risk which the Centre took despite frequent admonitions by Rado and me.”
As the Germans approached Moscow, communications became increasingly difficult, and suddenly, at twelve hours’ notice to the senior staff and none at all to its radio agents, the Centre was wrenched from its office and shifted 550 miles southeast to Kuibyshev. The move very nearly wrecked the Swiss ring. “On October 19,” Foote wrote, “Moscow went off the air in the middle of a message. Night after night Rado and I called, and night after night there was no reply. Rado was in despair and talked of going over to the British…. Suddenly one night at the scheduled time—and six weeks after the break—the Centre piped up. As if nothing had happened, they finished the message that they had cut off halfway through, a month and a half before.”
The information that the Centre’s receivers pulled in from the Swiss network was valuable in the extreme. For Rössler provided the Russian general staff with nothing less than the day-to-day German order of battle. This told the Russians just which forces were opposing them. How heavily they relied on it can be demonstrated by a negative case in which Lucy’s information was erroneous or falsified (just how this happened is not known). It dealt with troop dispositions, and, the Director told Foote after the war, it “cost us a hundred thousand men at Kharkov and resulted in the Germans reaching Stalingrad.” Such total dependence suggests that many of the Russian war victories owed their success to Rössler’s intelligence. Foote, in fact, believes that “Moscow very largely fought the war on Lucy’s messages.”
As with the Sorge network and the Rote Kapelle, the cipher in which this information was encapsulated could not be broken. The Funkabwehr and the Swiss police, the Bupo, intercepted hundreds of messages and read none. The Funkabwehr found that the transmissions were coming from Switzerland, where it had no power to make arrests, and the Bupo, which did have the power, at first was disinclined to do so in the case of an anti-Nazi group. For more than a year they left the ring alone, but Ge
rman pressure finally compelled them to act. In October, 1943, two of the Swiss transmitters were raided by the Bupo, and at 1:15 a.m. on November 20, as Foote was taking down a long message from Moscow, “there was a splintering crash and my room was filled with police…. I was arrested and the last link between the Centre and Switzerland was broken.” But its work was completed. Though another year and a half was to elapse before Germany surrendered, the issue was no longer in doubt; in the future shone the ultimate victory.
Russia’s wartime allies had never ceased to be her espionage targets, and peace enabled her to concentrate on them again. Soviet espionage scored most spectacularly with the atomic spies Klaus Fuchs and Allan Nunn May, but it did not neglect lesser fry. As the Iron Curtain clanged down and the Cold War grew gelid, secret agents were planted here and there throughout the free and uncommitted worlds. The Soviet spy net covered the globe. To direct it, to protect it, and to harvest its catch, an elaborate system of secret communications was required. Often the node of a spy ring was the Russian embassy, and here security began with physical safeguards.
In Canada, the cryptographic keys used by cipher clerk Igor Gouzenko were kept in a sealed bag that was placed each night inside a steel safe that was within an eight-room suite, closed by double steel doors and with iron bars and steel shutters on the white-opaqued windows, that was on the second floor of a separate wing of the brick embassy building, which was surrounded by a fence. In Australia, where Vladimir Petrov handled the espionage cryptography, the key to the safe in which his cipher documents were stored was kept in an envelope sealed with wax and a signet and locked inside the general embassy safe. The cipher suite here consisted of four rooms, two outer ones, which served for general embassy purposes, and two inner sanctums, a kind of holy of holies, in which the clerks enciphered the espionage messages. In the desk of the chief cipher clerk in an outer room, Petrov saw four revolvers in a drawer. Both embassies had stoves in which to burn cryptographic worksheets and other secret documents. In the 1960s, the Washington embassy kept chemicals at hand that could eat through a thick stack of paper in seconds; in an emergency, this is much faster than trying to stuff wads into the incinerator. Just how seriously the Russians take their security is illustrated by the fact that on New Year’s Day, 1956, they preferred to let flames gut the embassy in Ottawa rather than admit Canadian firemen to the grounds and, officials said, risk foreigners’ seeing their codes and ciphers.
For transmission, documents are photographed and the films—undeveloped, so that a thief’s illegitimate opening of them in the light would ruin them—sent in the diplomatic pouch. This procedure is followed for messages both to and from the embassy. When the 35-millimeter film arrives from Moscow, clerks develop it, make a single enlarged print from each frame, and destroy the negative. When Moscow acknowledges receipt of an embassy film, the embassy burns the original documents. The film for the secret police may come in a packet marked P.M.V., the initial letters of the Russian office of weights and measures. Late in the 1950s, the Russians began taking the precaution of using a locked container that automatically spilled acid onto the undeveloped film if anyone tampered with it. New cipher keys are sent by diplomatic bag; they are in an envelope, addressed to the secret police official, which is sealed inside an outer envelope addressed to the ambassador himself.
Those keys are one-time pads—in the old-fashioned manual form. Although the several branches of a Soviet mission—diplomatic, secret police, military, commercial, and political (Communist party)—all have their own keys and (in the larger embassies) cipher clerks, all probably use this system. All cables coming into the legation look alike: simple groups of five digits. The chief cipher clerk applies a key to the last group; it might decipher out as 66666, which on one day might mean that the message belonged to G.R.U., another day, K.G.B., another day, the trade section.
The one-time pads also play a role in the photographed letters. These are composed in plain Russian using the semisecret jargon of espionage—PACKING to mean ciphering, OPEN PACKING to mean plaintext, BANK for hiding place. In addition, specifically assigned codenames are used to cover real identities. For example, in Canada, Colonel Zabotin, the military attache, was GRANT, and ALEK stood for Allan Nunn May. (How successfully such codenames work may be seen in Canada’s Report of the Royal Commission on the Soviet spy ring, in which the commissioners conceded that “we have been unable to identify the following persons named under ‘cover-names’ in the documents and there definitely stated to have been members of Zabotin’s ring: GALYA, GINI, GOLIA, GREEN, SURENSEN.”) After codenames have been inserted, a code clerk copies the document, removing all sensitive terms and replacing the first with “No. 1,” the second with “No. 2,” and so on. The letter is photographed in this form. The terms themselves, with their numerical equivalents, are enciphered by the one-time pads. The numerical ciphertext is enclosed with the films in the diplomatic pouch on an ordinary sheet of paper.
Thus, when Vladimir Petrov developed a filmed letter from Moscow dated 25 November 1952, he found, in part:
We request you to report to us by the next luggage all the information known to you concerning No. 42, who figures in the departmental files in connection with her No. 43, and about her No. 44 in Sparta….
Depending on the availability of full particulars concerning No. 42 and her No. 44 in Sparta, we shall weigh the question of No. 45 to Sudania one of our planners along No. 46 Novators, under the guise No. 44 of No. 42
Petrov knew that LUGGAGE meant mail, DEPARTMENTAL, consular, and PLANNER, cadre worker in the espionage jargon. He consulted a list of formal codenames and found that SPARTA meant Russia, SUDANIA, Australia, and NOVATORS, secret agents. Decipherment of the accompanying sheet of paper told Petrov that, in this letter, “No. 42—Kazanova; No. 43—last will and testament; No. 44—relatives; No. 45—sending; No. 46—lines….” Thus, interpreted, decoded, and with the deciphered equivalents of the numbers inserted, the paragraphs read:
We request you to report to us by the next mail all the information known to you concerning Kazanova [an old Russian woman living in Sydney], who figures in the consular files in connection with her last will and testament and about her relatives in Russia [whom she wanted to see]….
Depending on the availability of full particulars concerning Kazanova and her relatives in Russia, we shall weigh the question of sending to Australia one of our cadre workers along the lines of a secret agent, under the guise of a relative of Kazanova.
This system appears to be used instead of total encipherment because security has bowed to convenience. To encipher everything fully would take too long. Perhaps one reason is that the cipher clerks have to work by hand.
For its agents in the field, however, the Soviet Union uses the best. It takes no chances, cryptologically speaking, with them or their networks. It gives its agents the confidence that they need fear nothing from cryptanalysis. It will not jeopardize their radio links with Moscow by trusting to anything less than the one perfectly secure system of encipherment. The main Soviet spy cipher today employs the one-time pad.
Its form varies. It has been found as a thick, squarish booklet the size of a postage stamp and as a scroll about the size of a cigarette butt. It seems to be growing smaller. A pad captured in 1954 had 40 rows of eight five-digit groups. One captured in 1958 had 30 rows of ten. Pads captured in 1957 and 1961 had 20 rows of four and five groups, respectively. Columns, rows, and pages are numbered. One booklet had 250 pages of a material like very thin gold and silver foil (several scrolls are needed to provide an equivalent supply of key digits). Usually, one part of the pad is printed in red and the other in black, presumably to distinguish the enciphering keys from the deciphering. The “printing” seems to be simple photography—probably the best way to make the one accurate copy of the original key that the agent will need; extra evidence for this is that the Russian word gamma (“scale”), which appears to be the Soviet term for one-time pad, is used in photo
graphy. Furthermore, the “paper” of the pad is cellulose nitrate, which was used for film in the early days of the motion-picture industry. It is highly inflammable, and spies seem to have kept potassium permanganate at hand to turn an ordinary combustion into an almost explosive reaction to destroy the pads rapidly and completely. No latent image would remain.
Interestingly, some pads seem to be produced by typists and not by machines. They show strike-overs and erasures—neither likely to be made by machines. More significant are statistical analyses of the digits. One such pad, for example, has seven times as many groups in which digits in the 1-to-5 group alternate with digits in the 6-to-0 group, like 18293, as a purely random arrangement would have. This suggests that the typist is striking alternately with her left hand (which would type the 1-to-5 group on a Continental machine) and her right (which would type the 6-to-0 group). Again, instead of just half the groups beginning with a low number, which would be expected in a random selection, three quarters of them do, possibly because the typist is spacing with her right hand, then starting a new group with her left. Fewer doubles and triples appear than chance expects. Possibly the girls, ordered to type at random, sensed that some doublets and triplets would occur in a random text but, misled by their conspicuousness, minimized them. Despite these anomalies, however, the digits still show far too little pattern to make cryptanalysis possible.
One-time pads have turned up with a number of top Soviet spies. Rudolf Abel, the highest-ranking Russian agent ever captured in the United States, had the one in the form of a booklet and the size of a postage stamp—1⅞×⅞×⅜ inches. F.B.I. agents found it when they arrested him in his room in New York’s Hotel Latham on June 21, 1957. Abel had wrapped it in paper and concealed it inside a hollowed-out block of wood covered with sandpaper like a sanding block (Abel posed as an artist) that he had tossed casually into the wastebasket. A Greek Communist, Gregory Liolios, had a one-time pad when he was arrested in 1954, as did another, Eleftherious Voutsas, picked up in 1958. In suburban London, early in 1961, half a dozen one-time pads in the scroll form were found hidden in the base of a Ronson cigarette lighter in the cottage of Helen and Peter Kroger, two Soviet spies who were actually two Americans named Lona and Morris Cohen. More pads were found in another lighter in the London flat of their chief, the Soviet Resident (agent in charge) for England, known only by his alias, Gordon Arnold Lonsdale. Later that year, Japanese police rounded up members of a North Korean Communist spy ring, and found among their effects some one-time pads. Atomic scientist Giuseppe Martelli, accused of espionage against Britain for the Soviet Union, was carrying two tiny packs of pads in a pack of cigarettes when he was apprehended at Southend Airport in 1963. Seven cigarettes were intact, but six others were glued together and partly cut away to form a recess for the pads. And a former spy for East Germany, who received his messages in an open broadcast of numerical codegroups and sent them by leaving them in a tin box hidden under a tree root, also enciphered with the one-time pad.
