by Kahn, David
Like Radio Cipher H, the Dockyard system also replaced pairs of plaintext letters with pairs of ciphertext letters. But it used one of twenty tables that were changed every other month; later thirty tables were used and were changed monthly. Its cryptograms consisted of five-letter groups, easily distinguishable from the four-letter groups of Enigma cryptograms. Soon after Denniston made his pessimistic remark, a captured document yielded a crib to Dockyard. It enabled the hand-cipher unit of Naval Section to achieve its first solution in that system; subsequent solutions were derived from frequency analysis and probable words. The Dockyard Cipher proved to be used more widely than its name implied: it served not only for messages to and from shipyards and their associated vessels and posts but also for messages to and from many smaller vessels, such as patrol boats and auxiliaries. At G.C.&C.S., seven cryptanalysts, the best of them a woman named Ruth Briggs, worked on it, rotating around the clock, aided by seven women who filled in the cipher tables as the cryptanalysts pulled out the bigram equivalences during their solutions. A certain Teutonic rigidity helped the cryptanalysts. Cipher regulations called for Dockyard plaintexts to end with a word of padding and gave as examples Wassereimer, Fernsprecher, Kleiderschrank (“water pail,” “telephone,” “clothes closet”). The young signalmen often used these very words as padding, varying them from time to time with Rosengarten. For the Bletchley cryptanalysts, Dockyard began to serve mainly as a source of cribs to Enigma cryptograms in cases where the Germans radioed virtually the same content both to vessels that had no Enigma and to those that had it. In Bletchley’s colorful jargon, such cribs were called “kisses.”
Naval Section exploited another form of radio intelligence called traffic analysis—in this case a study of the flow and volume of German naval communications. It worked from direction-finding and the external indicators of messages, such as addressees and radio frequencies, to build up a picture of the communications organization and of its normal activity. The analysts looked for any deviation from these norms, which might reflect some unusual activity. Gordon Welchman had made a similar study for German army messages.
Unlike cryptanalysis, which can explicitly reveal the intentions of an enemy in his own words, traffic analysis rests almost entirely upon inferences. A radio circuit normally quiescent springs to life. Does this mean that the squadron it serves is preparing for a sortie—or only that some disciplinary problem has arisen? These possibilities may be distinguished by the priority level and the length of the messages, but such indicators seem insubstantial and thin, and thus second-rate to someone who is not intimate with them.
Perhaps that was one reason Naval Section had assigned its traffic analysis to a very young man, a new recruit with no standing, given to wearing longish hair and corduroy trousers. In fact, having been plucked from university before he had completed his studies, he was technically still an undergraduate.
Francis Harry Hinsley, called Harry, had been recruited when Denniston sent a letter to the heads of about ten Oxford and Cambridge colleges asking them to recommend half a dozen of their best students for war work. Denniston realized that G.C.&C.S. needed able young generalists in addition to mathematicians and linguists. Hinsley, who had grown up in Walsall, a town eight miles northwest of Birmingham, had gotten into St. John’s College, Cambridge, on a history scholarship, having chosen both the field and the school because the schoolmaster at Queen Mary’s Grammar School in Walsall was a St. John’s historian. Hinsley had been at Cambridge for two years and had just taken first-class honors in Part I of the history tripos when, on a morning in October 1939, he was summoned to an interview in the third court of St. John’s with Denniston and two other G.C.&C.S. officials.
“We understand you’ve been abroad a lot,” one of them said. Hinsley had spent his summer vacations from 1937 to 1939 with pen pals in Germany to learn that language, staying, for example, with a German family in Koblenz. Denniston and the others asked no penetrating questions, no questions about security or whether Hinsley was a Marxist, only such matters as whether he preferred government service to being conscripted. Their decision rested in part simply on liking his face and on the fact that the master of the college had recommended him; the old-boys method of recruiting—which failed so miserably with Kim Philby, Guy Burgess, Donald Maclean, and Anthony Blunt, all of whom were Communists—succeeded brilliantly with Hinsley and about twenty other Oxford and Cambridge undergraduates. He was hired into the Foreign Office at £150 a year; he was not told what work he would be doing but only to present himself on a certain day at London’s Euston Station, when he would receive further instructions.
At breakfast on the day of his departure, he repeated these directives to the St. John’s medieval history don. The don, who had himself been recruited by G.C.&C.S., realized that Hinsley was bound for the same destination and drove him there that day. At Bletchley Park, Hinsley learned what the work was about. Told that G.C.&C.S. needed men in both the air and naval sections, he chose the naval and was sent to the section chief, Frank Birch.
Naval Section had only started to accumulate information. Birch gave Hinsley lists of intercepts and told him to see what he could make of them. For each message, the lists gave the call signs of the sender and the addressee, its time of origin, the radio frequency of the transmission, the time of interception, and the number of cipher groups in the message. By examining and comparing these lists, together with intercepts of radio operators’ chatter and service messages requesting retransmissions because of some error or other, plus direction-finding fixes on transmitters, Hinsley learned a great deal about German naval signals. He figured out the locations of the fixed transmitters ashore and tracked the movements of those aboard ship, saw which stations “talked” to which, and gained a feel for the normal message lengths and the traffic volume on each link. He did well in this rather mechanical work, from whose results he and others in Naval Section gradually sketched out the structure of the Kriegsmarine signal system and, by implication, of the Kriegsmarine organization. Then they identified the call signs of individual ships. Naval Section hoped eventually to warn when some potentially threatening enemy activity was afoot. Hinsley’s familiarity with the German system made him the person to report any such activity to the Admiralty.
But the Admiralty’s Operational Intelligence Centre paid little attention to this inferential evidence. On April 7, 1940, for example, Hinsley telephoned it to say that German naval radio activity had been heard in the waters west of Denmark and in the exit from the Baltic—activity that was unprecedented since the start of the war. This was G.C.&C.S.’s first substantial contribution to naval intelligence, but O.I.C. essentially ignored it. Clayton, the conservative head of O.I.C, and Denning, a deputy, had never met Hinsley; he was but a youthful voice on the telephone to them. He was, moreover, a civilian, with no experience of naval matters. They had built up no trust in him; they had had no experience with traffic analysis and had gained no confidence in it. Two days later, it turned out that the communications volume represented German warships and troopships sailing toward Denmark and Norway, which Germany was invading. This confirmation of Hinsley’s traffic analysis did not, however, convince O.I.C. It took a catastrophe two months later to change its attitude.
The Allied front in France was collapsing when the British decided to bring their forces home from Norway. Their attempt to repel the German invasion had failed. But they were still fighting fiercely near the end of May, when the Germans ordered three cruisers to aid a German garrison isolated at Narvik in northern Norway. To alert German naval commands to this sortie, a flurry of communications swept up the coast of occupied Norway. The British overheard them, and Hinsley reported to the O.I.C. about it. That unit duly recorded that “from a study of German W/T [wireless telegraphy] traffic … there would appear to be a movement of certain enemy ships, class and type unknown, from the Baltic to the Skagerrak,” the strait between Denmark and Norway. But the O.I.C. took no action. On June 4 th
e German flotilla sailed. It consisted of the Gneisenau, the Scharnhorst, the Admiral Hipper, and six support vessels. On Friday, June 7, Hinsley warned that the German warships might take “offensive action.” The O.I.C. noted this but otherwise ignored it. As a consequence, the Admiralty issued no warning to its forces at sea.
The next day, the British aircraft carrier Glorious, with two destroyers as escorts, was steaming toward Scapa Flow with men and planes rescued from Norway. Built as a cruiser in 1916 and converted to a carrier in 1930, the Glorious was commanded by Captain Guy D’Oyly-Hughes, a man undoubtedly brave, undoubtedly energetic, but whose inept leadership had greatly reduced morale aboard his ship. The ship was carrying seven Hurricane fighters and ten Gladiator biplane fighters that had been flown aboard in a dramatic rescue operation: it was the first time that Hurricanes had landed on a carrier, a risk taken because the planes and their crews were badly needed in Britain to defend against the expected German air onslaught.
The day was clear and warm; visibility was excellent. The sea was calm except for a long swell from the northwest. There was little wind. Lassitude seemed to claim the ship. When one of the pilots commented about the warheads being removed from the torpedoes, the chief torpedo instructor said with a happy smile that they were getting “a lap ahead” to go on leave. The laxness led to carelessness. Though the Glorious carried no radar, D’Oyly-Hughes had not sent a lookout aloft to the crow’s nest and had not flown patrols. In the afternoon, the carrier was zigzagging at 17 knots around a mean course of 250 degrees.
Suddenly, four fountains of water sprang up on the ship’s starboard side. A sharp-eyed midshipman in the foretop of the Scharnhorst had, a while before, spotted a puff of smoke from the carrier at a range of twenty-eight miles. The Scharnhorst and the Gneisenau tracked it until they discerned the Glorious. At 4:32 the Scharnhorst opened fire with the 11-inch guns of her two forward turrets.
“The first salvo arrived 20 yards on the starboard side,” said one of the British aviators. “On the horizon were two little bits of smoke. I thought, this is extraordinary! I thought I’d go up to the bridge to see how they fought a naval action. The second salvo arrived just as I stepped on the flight deck. It wrote off the ladder I had just come up! There was a damned great hole in the flight deck.”
The German ships’ excellent optical range finders and outstanding gunnery soon had their shells hitting the Glorious and the destroyers again and again. Fires broke out in the hangars. Airplanes blazed. Men died. One airplane fitter was hit in the chest with a heavy fragment of black smoking shell. He fell flat on his back, exclaimed, “Oh my wife and children!” and died. A shell struck the bridge, destroying it and killing D’Oyly-Hughes and others but leaving the action helmsman immediately below unhurt, though he was now in the open and the body of the navigating officer lay in halves on either side of him. Shells tore into the carrier almost continuously. Corpses lay about, looking more like rag bundles than men.
The Glorious was listing 15 to 20 degrees to starboard and was burning out of control from stem to stern when the order was given to abandon ship. She sank at 6:10 P.M. in an enormous flurry of spray and foam. Both destroyers were also sent to the bottom. Many of the men who had survived their ships’ destruction were dying in the water, some of them crying out for their mothers and for God to help them. The killed and missing for the three ships totaled 1,519, including most of the badly needed air crews. Only thirty-nine men survived.
The disaster taught a terrible lesson. The O.I.C. realized that had it listened to Hinsley and urged the Admiralty to warn its forces about the German ships, Britain might not have suffered its awful loss. Clayton and Denning brought him down to the Admiralty, where he stayed for three weeks during the Blitz (the German air raids on London) getting to know the personnel at O.I.C. and letting them get to know him. Hinsley had won their confidence.
Thus when, at the end of January 1941, G.C.&C.S.’s traffic analysis revealed that heavy German warships were about to sortie from the Baltic, and when an attaché report confirmed the presence of battle cruisers in Danish waters, the Admiralty took action: major elements of the Home Fleet stood out from Scapa Flow to catch and destroy them. It was not the fault of G.C.&C.S. or of O.I.C that the Germans brilliantly evaded the Royal Navy and later sank more than twenty ships in a two-month sortie.
The growing closeness between the codebreakers and the intelligence agency was further exemplified by the establishment of an assistant directorship of the O.I.C. to supervise the action taken by the Naval Intelligence Division on radio intelligence. The post was filled by Captain Jasper Haines, a tall, gray-haired officer so handsome that many of the young women in Naval Section regarded his regular visits to Bletchley Park from O.I.C, where he was based, as occasions—and woe to the boss who didn’t find some excuse to keep Haines around a little longer. Haines supplemented the day-to-day contacts between London and Bletchley.
But the naval Enigma still could not be read. Seeing no chance of analyzing the machine, the British began to consider ways of capturing keys. Some vague discussions took place during the summer, but the first clear proposal came from the assistant to the director of naval intelligence, an imaginative civilian who later became world famous as the creator of superspy James Bond. In a note of September 12, 1940, to his chief, Ian Fleming wrote:
D.N.I.
I suggest we obtain the loot by the following means:
1. Obtain from Air Ministry an air-worthy German bomber.
2. Pick a tough crew of five, including a pilot, W/T [wireless telegraph] operator and word-perfect German speaker. Dress them in German Air Force uniform, add blood and bandages to suit.
3. Crash plane in the Channel after making S.O.S. to rescue service in P/L [plain language].
4. Once aboard rescue boat, shoot German crew, dump overboard, bring rescue boat back to English port.
In order to increase the chances of capturing an R. or M. [Räumboot, a small minesweeper; Minensuchboot, a large minesweeper] with its richer booty, the crash might be staged in mid-Channel. The Germans would presumably employ one of this type for the longer and more hazardous journey.
Clayton supported the idea, and Fleming produced a more detailed plan. The bomber would take off shortly before dawn as one of the big London raids was ending. It would try to spot an isolated small minesweeper, cut an engine, emit smoke, drop fast, and pancake into the water. Fleming drew up a list of the personnel and material required, including a pilot who was “tough, bachelor, able to swim.” In case word got out that a German boat had been brought into an English port, Fleming proposed to put out a cover story that the capture was done “for a lark by a group of young hotheads who thought the war was too tame and wanted to have a go at the Germans. They had stolen plane and equipment and had expected to get into trouble when they got back.”
Frank Birch, the head of Naval Section, liked the plan, which he called “very ingenious,” in part because it would not give away British cryptanalytic efforts if it failed. He provided a three-page memorandum of “Activities of German Naval Units in the Channel.”
The plan was approved and given the codename Operation RUTHLESS. Fleming assembled the needed men. The director of naval intelligence, Rear Admiral John H. Godfrey, scrounged a bomber from the Air Ministry. A twin-engine Heinkel III, shot down during a raid on the Firth of Forth, had crash-landed on the moors not far from Edinburgh without great damage. The British restored it to flyability. Group Captain H. J. Wilson, in charge of evaluating captured German aircraft, flew it to the Royal Aircraft Establishment in Farnborough, southwest of London.
A few months later, Wilson was visited by some naval intelligence officers. They explained the plot, referring to the capture not of cipher documents but of secret radio equipment. The plan—which, they assured him, had the blessing of Churchill himself and would guarantee a decoration for the pilot—had elaborated: Spitfires would mock-attack the bomber. Wilson pointed out that the crash landing in
the rough autumn seas would unquestionably collapse the Heinkel’s all-acrylic Perspex nose, would probably sink the plane almost at once, and would almost certainly seriously injure the crew. He asked how these shaken-up, soaked, cold crewmen could leap from their dinghy to intimidate the crew of the rescue vessel and capture it. He himself declined the navy’s offer to participate, saying he preferred to work on successful operations than to win a posthumous Victoria Cross.
The intelligence officers were disappointed at his reaction but did not give up. When they advertised for men for a suicide mission, a love-sick pilot volunteered. Wilson reinforced the Heinkel’s nose and found a way to inject oil into the exhaust to simulate an engine fire. All that the airmen could do had been done. In October, Fleming went to Dover to await his chance. None came. Air reconnaissance found no suitable German ships operating at night, and radio reconnaissance likewise found nothing. On the sixteenth, the Dover command postponed RUTHLESS but suggested trying it at Portsmouth. Four days later Birch wrote: