Engineers of Victory

by Paul Kennedy

  Nevertheless, though the heroism and fortitude displayed by the Allied crews had been of the highest order throughout, the fact remained that their navies had taken a tremendous beating. The Battle of the Atlantic was being lost.

  The Allies’ Many Weaknesses

  Why did the battle of the convoys and escorts versus the U-boats go so badly for the Allies in these critical weeks? It will be clear to the reader that this field of war was so complex that no single factor was decisive. Indeed, some historians of this great campaign, including such a central participant in the convoy battles as Sir Peter Gretton, feel uneasy at suggesting that any single factor could be described as the major reason for its eventual outcome.16 To this author, however, some causes clearly justify a greater emphasis than others.

  To begin with, there was the sheer imbalance in numbers in early 1943. No doubt the British, Canadian, and American navies had many other very pressing calls upon their stretched naval resources in these weeks, but if this really was the critical theater for the Western Allies in 1943, then to allocate to a North Atlantic convoy of fifty slow merchantmen an escort of only four or five warships, and at a time when it was known that more and more U-boats were becoming operational, was taking a great risk. As one examines the charts of the ships’ dispositions during the hours and days of attacks upon the convoys, it is clear that an escort commander such as Lieutenant Commander Luther faced a totally impossible task—how on earth could his ship possibly try to drive away submarines bunching together on one flank of the convoy, pick up survivors from sunk or sinking merchantmen in the receding waters, and protect the rest of his flock? In the rock-paper-scissors game, all three sides are equally strong and weak. This was simply not true of the battle between U-boat, convoy, and escort that was played with HX 229 and SC 122. Numbers were decisive. There were far too few escorts, and a late rush of aerial and surface support was indeed too late.

  Then there was the imbalance in intelligence, for this was probably the time when B-Dienst was at the height of its competence, reading Admiralty messages at an impressive rate. Of course, individual submarines missed messages intended for them and steamed in the wrong direction, though sometimes with a favorable, unexpected sinking to follow. But, taken all together, those boats were being directed by Doenitz in a rigorous, demanding fashion—it was not just that he was in charge of one orchestra, but rather that he was directing, hustling, energizing four orchestras, four separate U-boat groups, sometimes more than a thousand miles from each other. By contrast, at this particular time Allied intelligence about German intentions seems to have been horribly behindhand. Even when Enigma decrypts came through to Western Approaches Command, they were likely to be late—and such delays came at a high price in the North Atlantic.

  Another important element, as noted above, was that of air cover, or, rather, the lack of it. If anything confirmed the geopolitical significance of the mid-Atlantic air gap, it was the location of the losses of Allied merchantmen in the awful months when 1942 unfolded into 1943. Of course, U-boats were sinking Allied cargoes off Trinidad, Buenos Aires, and the Cape; the Gibraltar route, being so close to German naval and aerial bases in western France, was always under attack. But when one turns to the vital sea line of communications between North America and Britain, the evidence is overwhelming. Captain Roskill’s official history gives us a map of the sinkings of the merchantmen: every one was in the air gap.17

  This was the place where the three-way game with the convoy, the surface escorts, and the U-boats was played out. Even a weakly escorted convoy with some daylight aerial coverage, such as SC 122, had a better chance during those hours than another weakly escorted convoy with no aircraft at all, such as HX 229. But what if the Battle of the Atlantic became a four-way game—merchant ships, naval escorts, U-boats, and aircraft—for the entirety of the voyage? What if, tactically, Allied aircraft operating both day and night made surface attacks by submarines simply too dangerous?

  THE NORTH ATLANTIC AIR GAP AND CONVOYS

  During the first half of the war, almost all Allied merchantmen, whether in convoy or not, were sunk by U-boats in this mid-Atlantic gap; it was not to be covered until the coming of very long-range Allied aircraft in mid-1943.

  Click here to download a PDF of this map.

  A fourth factor was, simply, the quality of the respective weapons systems. The key U-boat that was deployed at this time, the Type VIIC, was a narrow, cramped, and very basic piece of equipment, a mere 800 tons in weight and 220 feet long, carrying forty-four crew members stuck in unbelievable conditions. It was also extremely workmanlike and reliable, and it could dive fast if trouble loomed on the horizon. Another series, Type IX, was larger, with much greater endurance and heavier surface firepower, but it was a rather clumsy vessel if a British or Canadian escort was bearing down on it. Fortunately for Doenitz, he had plenty of the older class in the North Atlantic, allowing him to divert the bigger boats to distant stations, where they could play to their strengths.

  By contrast, the Allied hardware at this stage was unsatisfactory. A number of twenty-five-year-old ships had had new equipment stitched onto their bow or aft decks, a desperate half measure until newer vessels arrived. Almost everything was experimental, and liable to have teething troubles. As noted, escorts lost their sonar readings for around fifteen minutes after firing a depth charge or a torpedo. Important, too, was the fact that most newer weapons, which did have great potential, were encountering operational problems, and Royal Navy crews thus discounted both their value and their use. Older corvettes did not usually have the larger front bow of later types, so they almost drowned in the 150-foot-high Atlantic waves. Not possessing the newer high-frequency direction-finding (HF-DF) radio detectors was obviously a massive disadvantage to most warships. Yet the minuscule radio crew of HMS Volunteer, which did possess such equipment as it shepherded convoy HX 229 across the Atlantic, had to choose whether to spend their time detecting enemy movements or receiving anxious queries from the Admiralty—they could not do both. And almost all the small escorts were manned by scratch crews and naval reserve officers (Royal Navy Reserve and Royal Navy Volunteer Reserve), many of them a year or less away from civilian life or university—naturally, the professional naval officers were kept for the fleet destroyers, the cruiser squadrons, the Home Fleet. Britain was not playing its strongest cards, and Canada was struggling, with extreme effort, to create a very large escort navy out of virtually nothing. Its time would come, but not in the spring of 1943.

  Other aspects appear less central to our story. The weather, for example, seems to have been sublimely neutral. A distinguished Canadian scholar of the convoy battles, Marc Milner, believes that the defenders had a stronger hand when winter gave way to spring and summer, with improved weather, calmer seas, and many more hours of light each day.18 But the record itself suggests that if a convoy battle was fought in terribly rough or icy seas, then all the combatants—submarines, escorts, aircraft—were so battered by the storms that they naturally had less time and energy to devote to finding and sinking the enemy. Calmer waters and longer daylight hours did indeed offer warships and, especially, Allied aircraft more opportunity to detect U-boats on the surface, but it also provided submarines with better conditions for spotting a merchant ship’s giveaway smoke. Thus, improved weather conditions gave all of the operational elements—visibility, detection, maneuverability, firepower—a better chance of being exploited by either side.

  The disadvantages under which the Allied convoys were laboring in early March 1943 were, then, pretty overwhelming: inadequate naval protection, poor intelligence, nonexistent or minimal air cover (and no cover at night), together with a lot of poor, faulty, and outdated equipment. In these circumstances, it seems surprising that the shipping losses were not even greater. After all, HX 229 and SC 122 were scarcely in a better position as they steamed through the Atlantic air gap than the tragic and infamous Arctic convoy PQ 17, which had lost no fewer than twenty-three
ships to German submarines and aircraft when it was forced to scatter while only halfway to north Russia in July 1942.19

  The exceptionally pounding oceanic turbulence of this early spring of 1943—one elderly ore carrier simply snapped in half during the storms—had brought the pugilists’ match to a brief close. Doenitz used this opportunity to give his boats a breather and bring them back to their French bases, while sending others to the South Atlantic. The British used it to rethink and regroup, even as Admiralty planners struggled to ensure protection for ever more distant convoys, such as those carrying troops and supplies across the Indian Ocean.

  But by the beginning of April, as the ice floes diminished, the win-or-lose battle for control of the Atlantic was approaching its climax. By this time, Doenitz possessed a staggering number of operational submarines (around 240, with another 185 in training or refit), so he was capable of sending as many as forty or more U-boats against any particular convoy. The Allied navies and air forces had also rebuilt their fighting power, but perhaps the most noticeable change was in operational policy, in adopting a more aggressive stance, as proposed by the leader of Western Approaches Command, Max Horton, who swiftly gained the backing of his superiors at the Admiralty. The conclusion drawn from the grim experience of the SC 122 and HX 229 convoys being attacked on all sides, as the First Sea Lord, Sir Dudley Pound, put it to the cabinet committee on antisubmarine warfare in late March, was that “we can no longer rely on evading the U-boat packs and, hence, we shall have to fight the convoys through them.”20 In retrospect, this may be the single most important statement made regarding the Battle of the Atlantic. Without perhaps even the First Sea Lord himself fully appreciating it, the decision that the convoys were going to be defended much more vigorously, in a recognizable life-and-death struggle for control of the sea-lanes, gave the Allied side a much clearer focus than before.d

  Newer Elements Enter the Fight

  Admiral Pound’s statement was not just one of those romantic and archaic calls to fortitude in dire times. There certainly would be fortitude in the months to come, but now the convoys would be protected by defending forces that steadily enjoyed vastly improved tools of war, and many more of them, than they had possessed previously. These new resources worked to the Allies’ advantage and blunted Doenitz’s strategy, despite the great increase in the number of U-boats at his disposal. This chapter’s narrative is therefore quite different from that concerning the struggle for command of the air over Germany (chapter 2), where the introduction of a single weapons system was manifestly seen to turn the tide, or that concerning the war in the Pacific (chapter 5), where a remarkable succession of breakthroughs—U.S. Marine Corps amphibious warfare, U.S. Navy fast carrier groups, Seabees construction teams, and B-29 bombers—gave America the upper hand. And although the change in fortunes between the German and Allied navies in the Battle of the Atlantic occurred much earlier in the war than did the shift in those campaigns, the Atlantic story is far messier. The improvements did not arrive according to a grand incremental plan from Max Horton’s office; rather, they entered the Royal Navy’s tool kit episodically, and some of these newer systems took months before they properly fitted in the whole. Yet the commander of a U-boat that had been sent south in late March 1943 to wreak havoc off Freetown would have been completely disconcerted by what he saw when he arrived back at his base in Brest in July.

  That U-boat commander would have learned of a large and growing list of Allied improvements. Aerial support from RAF and Royal Canadian Air Force (RCAF) Coastal Command squadrons was rising fast, if unevenly, as additional squadrons arrived. New escort carriers were beginning to appear, some of them joined by recently formed groups of the newer corvettes and frigates. In addition, a couple of support groups consisting of very fast destroyers had been released from the Grand Fleet, due to the total inactivity of the German heavy warships based in Norway. Newer or vastly improved detection and killing equipment was at last making its way to the escort vessels. More and more HF-DF masts appeared even on small Allied escorts, and a fabulous 10-centimeter radar had also arrived. Wellington and Catalina bombers with the powerful spotlights known as Leigh Lights made the Bay of Biscay dangerous for U-boats on the surface. The Hedgehog grenades, together with much more powerful and better-set depth charges and aerial homing torpedoes, were all arriving on the scene. And the intelligence battle between B-Dienst and Bletchley Park’s code breakers was turning in favor of the Allies. In sum, while the U-boat commander and his crew were enjoying easy kills to the south, things had gone badly wrong in the North Atlantic.

  Interestingly, the poor weather in the north had obscured for a while the many improvements on the Allied side. The record storms of early 1943 continued unabated, so the convoys that did sail in late March suffered enormous physical damage; yet the two main ones, HX 230 and SC 123, got through with only one vessel sunk to the U-boats. The same was true of the early April convoys, HX 233 and ONS 3 and 4. The only convoy that met with serious attack at this time was HX 231, whose escorts fought off a whole group of U-boats on April 5 and 6, inflicted much damage, and brought the vast bulk of its cargoes to port. No fewer than twenty-two of the sixty-one vessels were oil tankers, while many of the others were carrying what might be termed “pre-D-Day” supplies—trucks, tanks, aircraft, landing craft, and vast amounts of ammunition.21 This, of course, was key. Just getting that one huge group of oil tankers alone to Britain staved off the island’s resource crisis of early 1943.

  And there was to be even better news ahead. The powerful winds and great long rollers of the North Atlantic never abated, but the sound of conflict now did. Amazingly, only fifteen merchantmen were lost in those waters between June and mid-September 1943, and only one of them was in a convoy. As the Allies girded themselves for further advances in the Mediterranean and for a really massive buildup of men and munitions in Britain in preparation for the future invasion of France, and as the demands of the Pacific and Southeast Asia campaigns grew ever greater, their shipping crisis actually intensified. But that crisis was essentially one of supply and demand, eventually solved by further stupendous outputs of American industry, and no longer about the hemorrhaging of ships from the convoy routes between New York/Halifax and the Clyde/Merseyside. To the tens of thousands of crew members of the merchant vessels who for the first time steamed those rough seas without a single attack, this must have seemed incredible, inexplicable, even a bit eerie. More U-boats were being sunk than merchantmen.

  One graph captures this dramatic change of fortunes in the Battle of the Atlantic during the months of 1943.

  U-BOAT VS. MERCHANT SHIP LOSSES IN THE NORTH ATLANTIC, 1943

  The dramatic rise in U-boat losses and decline in Allied merchant ship losses during the critical months of 1943 are well captured here.

  Before we understand the swiftest of the changes of fortune in the five major campaigns of the Second World War, it is appropriate to study the critical convoy battles of May 1943, especially the first, since it was the most important of all. It focused, unusually, around the voyage of one slow, westward-bound convoy, ONS 5, which sailed from its gathering place near the Clyde for North American ports between April 23 and May 11. Here was the reprise to the saga of SC 122 and HX 229. This time the Allies won—not easily, but quite decisively. The map on this page captures the overall situation.

  Forty-three merchant ships, emerging from five separate British ports, had gathered off the Mull of Kintyre with their escorts in late April and then set off on a great-circle route to the New World. They were not holding much cargo, but the point was that if they did not get back to America’s eastern ports, there would be fewer and fewer Allied vessels left to carry oil, ores, trucks, grain, and aircraft parts to the British Isles in the future. And it was emphatically an Allied convoy: in addition to the twenty-eight vessels flying the Red Duster, there were five ships from the United States, three from Norway, two from the Netherlands, two from Greece, and one each from Yugosl
avia, Panama, and Denmark. Escort Group B7 consisted of two destroyers and a frigate, two rescue trawlers, and four invaluable corvettes named, deliciously, Sunflower, Snowflake, Pink, and Loosestrife. In charge of the escort was Commander Peter Gretton, whose consuming passion was the sinking of U-boats; he had recently returned from the Mediterranean, where he had received the Distinguished Service Order (DSO) for ramming and demolishing an Italian submarine. Gretton had sailed and fought in the early North Sea convoys, in the Second Battle of Narvik, in the Atlantic and Mediterranean, and with the North African landings. He had just caught his breath after fighting through HX 231, and after this present convoy would soon be out to fight with others. Nelson’s frigate captains would have recognized him immediately.

  THE FIERCEST CONVOY BATTLE?

  A graphic illustration of how Allied convoys were sent on the great-circle route, yet still encountered the lines of Doenitz’s wolf packs. This convoy is ONS 5, which sailed from the United Kingdom on April 21, 1943, and had to fight its way through successive lines of German U-boats before arriving in Canadian and American ports almost three weeks later.

  Click here to download a PDF of this map.

  The convoy headed northwestward toward its welcome “cocoon” of Allied air cover from Iceland, which was available to it between April 25 and 27. After that it skirted north of the U-boat group Star toward Cape Farewell and a period of Greenland-based air cover, and then it encountered horrendous weather. So far there had been no sight of enemy submarines, and the only casualties—apart from weather damage to many ships—were the withdrawal of some of the destroyers, including Gretton’s, due to a shortage of fuel (refueling at sea during these storms was impossible). It is an indication of how bad the weather had become that the whole convoy was forced to heave to for a while and that the five destroyers of the reinforcing 3rd Support Group took more than a day to find out where it was. Ahead, Doenitz’s wolf packs had been instructed to play a waiting game. By the night of May 4–5, two long lines of them (twenty-one U-boats, in the groups Fink and Specht) were waiting as the achingly slow convoy steamed toward Newfoundland. One of the submarines had been attacked and sunk by RCAF flying boats during daylight hours, but at nightfall the U-boats’ time came.