The news he received when he got to his headquarters was bad. The attack out of Faïd Pass was much bigger and more aggressive than CCA had thought at first. The Germans had destroyed an American tank battalion, overrun a battalion of artillery, isolated two large segments of American troops, and driven CCA out of its position. Nevertheless, General Anderson continued to insist that Mockler-Ferryman’s intelligence was correct and that the main attack would come at Fondouk. He refused to release Robinett’s CCB to join CCA in the defense. Ike tried to speed a flow of reinforcements to CCA, but his main strategic reserve, the U. S. 9th Infantry Division, was unable to move with any dispatch because it had no organic truck transportation. As a result, outnumbered and inexperienced American troops had to take on German veterans led by Erwin Rommel himself. The result was one of the worst American defeats of the war. CCA lost ninety-eight tanks, fifty-seven half-tracks, and twenty-nine artillery pieces. It had practically been destroyed—half an armored division gone!5
Fortunately for Ike, the German command setup was almost as muddled as the Allied one. Rommel and von Arnim operated independently. Von Arnim wanted to confine himself to limited attacks against Fondouk. Rommel was after much bigger results—he wanted to break through the mountains at Kasserine Pass, capture the great Allied supply base at Le Kef, then possibly drive on to Algiers itself. He wanted to turn a tactical advantage into a strategic triumph, destroying the II Corps, isolating the First Army, and thus reversing the entire position in North Africa. If all went well, he could accomplish his objectives before Monty was ready to attack the Mareth Line.6
Von Arnim was a vain, ambitious man who refused to cooperate in Rommel’s bold (but wildly impractical) plan. Higher headquarters (Kesselring) had ordered him to give his best panzer division, the 10th, to Rommel for the original attack, but von Arnim had stalled and it was not committed on February 14. Ironically, this turned out to benefit Rommel, because the location of the 10th Panzer was, according to Mockler-Ferryman, the key piece of information. As long as those tankers were facing CCB at Fondouk, that was where Mockler-Ferryman insisted that the attack would come.
Over the next two days Rommel pressed his initial advantage. On February 20 the 10th Panzer, finally released to his command, moved into Kasserine Pass. It was too late. American reinforcements had arrived. The German offensive stalled.
That same day, February 20, Ike asked the British Chief of the Imperial General Staff, Alan Brooke, to replace Mockler-Ferryman “with an officer who has a broader insight into German mentality and method.”7 It was the only time in his three-year career as Allied Commander in Chief that Eisenhower asked the British to relieve one of their officers on his staff. In a cable to Marshall the next day, Ike explained that “due to faulty G-2 estimates” Anderson had not become convinced “until too late that the attack through Faïd was really the main effort.”
Then, in guarded language, he added, “I am provoked that there was such reliance placed upon particular types of intelligence that general instructions were considered inapplicable. In this connection and for your eyes only, I have asked for the relief of my G-2. He is British and the head of that section must be a British officer because of the network of special signal establishments he operates, but Brooke has agreed to make available a man in Great Britain who is tops in this regard.”8 The man was General Kenneth Strong. He stayed with Ike through the remainder of the war and the two officers established a close and effective relationship. Mockler-Ferryman returned to London to head the Special Operations Executive (SOE), which controlled sabotage and underground efforts in occupied France.
But what, meanwhile, was the origin of Mockler-Ferryman’s terrible mistake at Kasserine Pass? It was ULTRA. An entry of February 20 in Butcher’s previously unpublished diary provides some of the details: “An explanation of the defeat, as seen by Ike, lies in a misinterpretation of radio messages we regularly intercept from the enemy. This source is known as ‘Ultra.’ It happens that our G.2 Brigadier Mockler-Ferryman, relies heavily upon this source. It has frequently disclosed excellent information as to the intentions of the Axis. However, the interpretation placed by G.2 on the messages dealing with the place of attack—an attack that has been expected for several days—led Mockler-Ferryman to believe that a feint would be made where the attack actually occurred … and that the real and heavy attack would come in the north.”9
What Butcher did not know was that Rommel’s initial attack was as much a surprise to von Arnim and his superiors as it was to Mockler-Ferryman. Rommel, not for the first time, had disobeyed orders.10
On March 14, after Rommel had been driven back both at Kasserine and then at the Mareth Line (and had consequently left Africa), Ike wrote Brooke again about Mockler-Ferryman. He said that his G-2’s performance, up to Kasserine, had been outstanding, pointing out specifically that “his forecast of the extent of French opposition proved in the event to be more accurate than that of any other authority.” Ike wanted Mockler-Ferryman’s relief to be “without prejudice.” Then he added, “In his successor, I now look for a little more inquisitiveness and greater attention to checking and cross-checking reports from various sources.”11
The Battle of Kasserine Pass has often been pointed to as the contest where the American Army of World War II came of age. Green troops became veterans; new commanders gained badly needed combat experience; over-cocky Americans learned what a tough opponent they were up against. The man who learned the most was the commander himself, Dwight Eisenhower, and one of the most important lessons he learned was that no one source of information, no matter how sensational, is ever by itself sufficient. Mockler-Ferryman had been so confident of ULTRA’S insight and trustworthiness that he had neglected other, more traditional sources. As Butcher noted in his diary, “Ike insists we need a G.2 who is never satisfied with his information, who procures it with spies, reconnaissance, and any means available.”12
In the aftermath of Kasserine, Ike also learned from interrogation of German prisoners that the enemy was “easily and constantly” breaking the low-level codes used by the 1st Armored Division. He decided that the Germans were probably as enamored with this information as Mockler-Ferryman, and that he could take advantage of them. He told Patton, “We should obviously but clumsily change the code at frequent intervals, so that the Hun will not suspect a plant, but never enough so that it will be impossible for him to break them quickly. As long as nothing is hurt the orders given in this way should be faithfully executed (unimportant patrols, etc.). But when the time comes for real surprise, use an erroneous order in order to support your other measures of deception. This effort should not be difficult to make—and it might work!”13 The innocent American was learning quickly.
DESPITE ITS RELATIVE FAILURE at Kasserine Pass, ULTRA was Ike’s single most effective spy throughout the war. It proved itself in every campaign from 1943 onward, beginning with Operation HUSKY, the invasion of Sicily, Ike’s second amphibious assault. Well before HUSKY was launched in July 1943, thanks to ULTRA, Eisenhower had a complete picture of the enemy’s order of battle on Sicily and in Italy. Equally valuable, ULTRA allowed him to penetrate the German mind and judge how successful Allied deception measures had been.
The major attempt at deception for HUSKY showed the British Secret Service at the top of its form. In an imaginative subterfuge, the British managed to convince the Germans that Eisenhower’s troops would land either on Sardinia or in Greece, rather than Sicily. This sophisticated deception scheme was potentially decisive, because the Germans had more than enough troops scattered throughout Italy and the Mediterranean to reinforce Sicily sufficiently to produce another Gallipoli.
The story is well known—it was superbly told by Ewen Montagu in his 1954 book, The Man Who Never Was—and needs only a brief summary here. A British Secret Service team searched the London morgues to find a suitable body—they needed a once fairly healthy, fairly young, and completely unknown man. Once found, they used odds and ends to
give him an identity, a biography, a history. He became “Captain (acting Major) William Martin, 09560, Royal Marines.” His pockets and his briefcase were stuffed with documents, matches, loose change, love letters, a bill or two, a bank statement, a photo of “mom,” all prepared with exquisite care to prove that Major Martin was authentic.
Major Martin was a courier. His briefcase was attached to his wrist by handcuffs. In it were various travel orders and other documents, some labeled “Most Secret.” The planted material consisted of two private letters, one from the vice chief of staff to General Harold Alexander, the overall ground commander in the Mediterranean, under Ike, and the other from Lord Louis Mountbatten to Admiral Cunningham. Each letter hinted that the next operations would strike at Sardinia and Greece.
At dawn, April 30, 1943, Major Martin was dumped overboard from a British submarine off Huelva on the Spanish coast. (At the last minute in London, there had been an anxious discussion about what would happen if the tide failed to sweep him to shore. Churchill gave his verdict: “You will have to get him back and give him another swim.”) The Spanish picked him up, opened the briefcase, gave the documents to a German intelligence agent (who photographed them and sent the film on to Berlin), replaced the documents in the briefcase, then gave it to the British vice-consul in Huelva. Major Martin was interred and his documents returned to London in the freshly sealed briefcase.
Had the Germans taken the bait? ULTRA showed that they had. From the War Cabinet Office to Churchill, then in Washington, the signal flashed, “Martin swallowed rod, line and sinker by the right people and from best information they look like acting on it.” The phrase “best information” meant ULTRA.14 Between early May and July 10, the date of the invasion, ULTRA provided mounting evidence of the successful deception, primarily through order of battle information, the area in which ULTRA was always at its strongest and most reliable. ULTRA reported that the Germans had moved the 1st Panzer Division from France to Greece, that they had moved units from Russia into Greece, that reinforcements from Germany were sent into Sardinia, and so on. In May, the Luftwaffe had had 415 aircraft in Sicily with 125 in Greece; by July there were 305 in Greece and only 290 in Sicily.15
ULTRA was precise about the opposition Ike’s forces would face on Sicily. Field Marshal Kesselring gave Berlin a complete rundown on his dispositions. He had the Hermann Göring Panzer Division on Sicily, along with the German 15th Panzer Division and some Italian troops (who were without transportation and badly equipped). Part of the 15th Panzer was in Palermo, on the north coast; the remainder, along with the Hermann Göring Panzer Division, was in the center of the island, ready to move in any direction. This was priceless information, as was Ike’s knowledge that via ULTRA he would be able to listen in on the German reaction to the landings.16
The initial assault went according to plan. On the morning of D-Day, from his advance headquarters on Malta, Eisenhower sent a cable to the Combined Chiefs: “Fragmentary information obtained mostly from intercept of messages indicates that leading waves of British 5th, 51st and Canadian Divisions are ashore and advancing.”17 ULTRA, in other words, was giving him not only the German reaction—which was slow and confused—but was also his best source on the immediate tactical dispositions of his own troops. The following day, July 11, was the critical one in the campaign, as German armor from the Hermann Göring Division counterattacked against American forces at Gela. ULTRA had provided an alert, and the Americans were ready with a combination of superb naval gunfire, artillery, infantry action, and tanks. The Germans were repulsed with heavy loss.18
The operation in Sicily did reveal ULTRA’S inescapable limitations. The Allies dared not act on ULTRA information that stood alone—i.e., there had to be some explanation other than a code break as to how they found out this or that, or the Germans would realize what had happened and change their code. Churchill and Menzies insisted that those “in the know” had to promise never to use ULTRA information until it was possible to point to some other source.
For example, parachutists, under the command of General James Gavin, dropped onto Sicily on the eve of the invasion, could not be told that the Hermann Göring Division was in their drop zone for fear of revealing the ULTRA secret. The men were not told they would encounter German tanks. They were also not given antitank weapons. They were told that there were some German “technicians” in the area and “nothing more.” In 1979, General Gavin commented, “From the viewpoint of protecting Ultra, I think that this was the proper course for the high command to take, provided they equipped us with adequate antitank weapons.”19
IF THE SECURITY OF ULTRA was a first objective, the question arises, how was ULTRA information relayed to the field commanders safely and swiftly? The British had worked out a system of Special Liaison Units (SLUS) to speed the intercepted messages from Bletchley Park (BP), where the decoding and translating took place, to Churchill and the generals. In 1943 the United States began to create its own SLUS. The result was a huge success and an extraordinary achievement, showing Americans at their best.
The Army’s selection process was superb. It managed to locate precisely the two dozen or so officers who were perfect for the job. They had to be young and healthy, because the SLUS worked long, taxing hours on intricate problems and because the SLUS had to be junior officers, usually captains or majors, so that they would not attract attention by their rank. They had to be diplomatic enough not to offend the senior generals to whom they reported, but firm enough to make sure the generals heard what they had to say (not always as easy as it might seem, especially when Patton or Clark were the recipients). Men who are absolutely trustworthy, mentally quick, tireless, and self-effacing (they knew there would be no battlefield promotions for them in this war, nor any opportunity to lead men into combat) are few in number—but America had enough of them, and the Army found them. To a man, they did an outstanding job during the war; to a man, they kept their trust, not one of them ever revealing the ULTRA secret or his part in the war.* It may not be too much of an exaggeration to say that the ULTRA system, from BP to the SLUS, was a triumph of the Western democracies nearly on a par with the creation of the atomic bomb.
TELFORD TAYLOR headed the American SLU effort. His later career, as was true of all the SLUS, was marked by success after success. Taylor was the prosecutor at the Nuremberg War Crimes Trials and later a distinguished author and professor of law at Columbia University. His young men, selected for brains and ability rather than rank or background, included William Bundy, who became Assistant Secretary of State; Alfred Friendly, who became managing editor of the Washington Post; John Oakes, who became an editor of the New York Times; Langdon van Norden, a businessman who became chairman of the Metropolitan Opera Association; Curt Zimansky, a noted philologist; Yorke Allen, of the Rockefeller Brothers Fund; Stuyvesant Wainwright II, four-term congressman; Lewis Powell, Associate Justice of the U. S. Supreme Court; Josiah Macy, vice president of Pan American Airways; and Adolph Rosengarten, who was a little older than the others but still had a successful postwar career, first as a director of the Fidelity Philadelphia Trust Company, then—in 1975, at age seventy—earning his Ph.D. in history from the University of Pennsylvania. Clearly the SLUS were outstanding junior officers.20
The SLUS served in a new Army organization, Special Branch of the Military Intelligence Service. In defining their responsibilities, General Marshall insisted, without any question of misinterpretation, that these officers were in a special category and that the generals in command must allow them (no matter how young or unmilitary) the necessary scope and authority.
Marshall gave his SLUS more latitude, and demanded of them a great deal more, than their British counterparts. In the British system, the SLUS were only glorified messengers who handed on the complete ULTRA intercepts to their superiors. Under the system that Marshall and Taylor created, the American SLUS synthesized, summarized, and interpreted the intercepts. As Marshall put it, “Their primary responsib
ility will be to evaluate Ultra intelligence, present it in usable form to the Commanding officer, assist in fusing Ultra with intelligence derived from other sources, and give advice in connection with making operational use of Ultra intelligence in such fashion that the security of the source is not endangered.”21 As Lewin notes, “This directive was so comprehensive and permissive that it allowed and indeed encouraged the representative to think of himself as a kind of private intelligence center.”22 As the SLUS were, in fact, for in the field each had his own tent, van, or trailer as an office—under continuous guard—in which his safe contained ULTRA papers plus a great deal more information.
To train these men, Taylor first of all sent them to BP, where they saw 10,000 of the most valuable people in the British Empire at work. They were deeply impressed. The exposure of the SLUS to the inner workings of BP meant that they understood the magnitude and significance of what ULTRA offered in a way that few field commanders could.
In addition, Taylor carefully indoctrinated the SLUS in all aspects of intelligence gathering, which gave them a perspective that was crucial to their effectiveness. The temptation to rely completely on ULTRA was always there, but usually spurned. In 1978, Rosengarten wrote, “I am bold to say that Ultra was primus inter pares, some of the time but not all of the time, among the sources of information which were available to our section. These were principally prisoners, civilians who crossed the line, air photography, and low level deciphering.”23
Rosengarten’s point was made over and over again by the American SLUS. After the war, Taylor had each of his men answer a long questionnaire about their experiences. In his summary of these reports, Taylor noted that “the need for careful study of all sources of intelligence was stressed by most of the representatives.” Everything that the commanding general’s G-2 section knew, the SLU knew, because he made it his business to read all papers passing through the G-2 situation room. This enabled him to fuse ULTRA with other intelligence.