'These Russian convoys are becoming a regular millstone round our necks, and cause a steady attrition in both cruisers and destroyers'. Admiral Pound (First Sea Lord) to Admiral King, U.S.N. (Chief of Naval Operations), 18th May, 1942.
At the beginning of the year Admiral Tovey commanded, in the Home Fleet, the battleships King George V and Rodney, the battle cruiser Renown, the aircraft-carrier Victorious, four 8-inch and six 6-inch cruisers and about eighteen destroyers. The new battleship Duke of York, which was not yet fully 'worked up', had taken the Prime Minister to America; the Nelson, three cruisers and some destroyers were refitting. Very heavy demands on destroyers were arising in connection with the Russian convoys, and the Commander-in-Chief was, as ever, concerned over the perpetual shortage of ships of that class. As to the enemy, it was believed that the new battleship Tirpitz, the pocket-battleship Admiral Scheer, the heavy cruiser Admiral Hipper, four light cruisers and about a score of destroyers were ready for sea. Most of them were known to be in the Baltic. The Scharnhorst, Gneisenau and Prinz Eugen were still in Brest, and constituted a permanent threat to our Atlantic convoys. They were all believed to have repaired damage received in the previous year1, and to be ready for sea, though in need of further training. The increasing signs of the Brest squadron's readiness caused the Admiralty to press for renewed bombing on a heavy scale, and in January 612 aircraft of Bomber Command dropped 908 tons of bombs on the base; but no further damage of a serious nature was done to any of the ships. The story of the squadron's escape back to its home waters up the Channel between the 11th and 13th of February will be told in the next chapter.
The readiness for sea of the enemy's Baltic and Brest forces compelled Admiral Tovey to keep as close a watch as possible on the northern passages.. One cruiser, therefore, stayed on patrol in the
Denmark Strait, and she was supported by a battleship and two American heavy cruisers, based on Hvalfiord. The Iceland-Faeroes passage was now closed by the minefield which had been laid by stages during the previous eighteen months. It was by no means an impenetrable barrier and needed reinforcement with more mines; but it was now better watched by the patrolling aircraft of Coastal Command's No. 18 Group, and it was far less likely that enemy warships would attempt to break out that way than it had been during the first two years of the war. Offensive measures were constantly considered, and Admiral Tovey wished by frequent coastal raids 'to turn the mind of the enemy to defence'. In this our strategy was more successful than could possibly have been realised at the time, because of Hitler's 'intuition' that we intended to invade Norway.2 It was actually the German dictator's insistence on the defence of Norway which caused the next movement by the enemy's main forces, for at his conference on the 12th of January he gave orders for the Tirpitz to be moved to Trondheim. 'Every ship', he declared at this time, 'which is not stationed in Norway is in the wrong place'. The German Naval Staff knew from recent experience that the movement of the great battleship from the Baltic to the North Sea by the usual passages was almost certain to be reported by British intelligence.3 They therefore brought the Tirpitz through the Kiel Canal to Wilhelmshaven whence she sailed to Trondheim, escorted by four destroyers, on the night of the 14th-15th of January. The German plan was successful, and on the 16th she reached her destination undetected. But the destroyers were at once needed in the south to help bring the two enemy battle cruisers up-Channel. Thus the lack of adequate escorts, added to the serious shortage of oil fuel in Norway, reduced the offensive possibilities of the new German dispositions from the start.
On the 17th of January the Admiralty gave warning that the Tirpitz might be at sea. Though it was not expected that she would this time attempt to break out into the Atlantic, Admiral Tovey was taking no chances. He shifted his main concentration to Iceland to cover the northern passages, postponed the sailing of the next Russian convoy (PQ 9) and cancelled an operation off the Norwegian coast. On the 23rd of January, after much strenuous searching, our reconnaissance aircraft at last found the battleship at anchor, camouflaged and heavily protected by nets in Aasfiord, fifteen miles east of Trondheim. Whether her purpose was to assist in the expected return home of the Brest squadron, to threaten our Russian convoys, to deter us from raiding the Norwegian coast or to break out into the Atlantic was still obscure; but from the Admiralty's point of view it
was plainly desirable to drive her away from a position in which she could exert any or all of these threats, or to immobilise her where she lay. As long as she was present in Norway her influence was bound to make itself felt in all the waters, from Murmansk to the American seaboard, for which the Home Fleet was mainly responsible. For a start it was decided that the Russian convoys should continue, though not more than one should ever be at sea between 10° West and 15° East; secondly, that the Northern passages must meanwhile be left uncovered, and lastly that raids on the Norwegian coast, 'which had so annoyed the German High Command' should continue.
On the 25th the Prime Minister drew the Chiefs of Staffs' attention to the cramping influence exerted by the battleship, and asked for plans to be prepared to attack her with shore-based and carrier-borne aircraft. He considered that 'the entire naval situation throughout the world would be altered' by her successful destruction.4 Unfortunately she was a difficult target to attack. Carrier-borne aircraft could not use their torpedoes in the narrow, steep-cliffed inlet, and Bomber Command's longest-range aircraft could only reach her from bases in Scotland at the limit of their endurance. The bombing plan was approved on the 28th of January, and was carried out by nine Halifaxes and seven Stirlings on the night of the 29th-30th. No damage was done.
There now followed a trying period for Coastal Command. They had to watch the Tirpitz continuously at a range and in conditions which made it almost essential to employ Mosquitos, of which the command possessed very few; secondly, 'break-out' patrols had to be flown in case she escaped from Trondheim unseen and made either for the Atlantic or for the Arctic convoy routes. At the same time it was becoming daily more urgent to watch the Brest squadron's movements. There is little doubt that the need to attempt so many duties at the same time was the main cause of the failure to have the Command's torpedo-bombers ready and at southern bases when the Brest squadron's expected move up-Channel took place. We shall return to that matter later.5
Between the 14th and 19th of February, by which time the Brest squadron had reached its home bases, great enemy air activity was noticed in Norway. On the latter date the Tirpitz was seen to be under way and exercising in the fiord. Air patrols were at once strengthened, submarines were stationed off-shore and Admiral Tovey left Hvalfiord with his main strength and steamed towards Tromsö. If nothing of greater moment took place, he intended to attack enemy shipping in that port. His anxieties had been made
heavier because a big WS troop convoy was due to leave the Clyde for the Middle East on the 15th and a strong escort, partly drawn from ships of Force H which had specially returned home, and partly from the Home Fleet and Western Approaches commands, had to be provided. After the Brest squadron had successfully reached its home waters, Admiral Tovey reviewed the future outlook and found it little to his liking. 'In a few months' time', he wrote in his despatch, 'the enemy would be able to confront us with a considerable battle fleet'. He did not expect the enemy to risk the Tirpitz by herself in an Atlantic foray, but that they would 'gradually assemble the rest of the fleet round her'. In this forecast he was soon to be proved correct, though the actual strength which the enemy could 'assemble round the Tirpitz' was less than the Commander-in-Chief feared. Though Admiral Tovey considered the outlook to be 'profoundly changed', he none the less realised that a period of quiescence was likely, while the battle cruisers were repairing the damage sustained during their dash up-Channel. He therefore seized the opportunity to refit some of his own ships. Moreover, the removal of the threat from Brest at once reduced his escort commitments. No longer need a battleship be sent with each important south-bound convoy.
On the 20th of February the Admiralty received indications of another intended warship movement from Germany to Norway. Air patrols were maintained, and the torpedo-bomber squadrons of Coastal Command came to immediate readiness. At 11:10 a.m. the next day two large warships (actually the Admiral Scheer and Prinz Eugen) and three destroyers were sighted off the Dutch coast, steering north at high speed. Relays of reconnaissance aircraft were sent out to keep in touch, and all the available Beauforts were despatched to catch the enemy force off Utsire that afternoon.6 Unhappily, the weather worsened, contact was lost and none of the torpedo aircraft found the enemy who had, we now know, turned back on his tracks for a time and so threw our searchers off the scent. Early next morning, the 22nd, two of our aircraft employed on other missions, did, somewhat luckily, sight the enemy squadron as it was entering the Inner Leads. By 3 p.m. the ships were located at anchor in Grimstad Fiord, just south of Bergen.
Admiral Tovey had meanwhile cancelled his intended attack on Tromsö and sent the Victorious, Berwick and four destroyers to a position 100 miles off Stadlandet7, whence the carrier's aircraft were to attack the north-bound warships at 1 a.m. on the 23rd. He himself followed in the flagship King George V to cover the lighter forces, and four submarines were sent to patrol off Trondheim. Though snowstorms and bad weather again defeated the air searches and strikes,
the submarine Trident (Commander G. M. Sladen) torpedoed the Prinz Eugen at 6 a.m. on the 23rd as she approached the entrance to the Leads off Trondheim. She was badly damaged and, for a time, stopped. The Scheer went on to Aasfiord and took up a berth not far from the Tirpitz. The Eugen managed to limp into the same sheltered anchorage at about 11 p.m. that evening. Though less damage had been inflicted than we would have wished, the enemy's plan to form a squadron composed of the Tirpitz, Admiral Scheer and Prinz Eugen 'to conduct offensive and defensive operations from Trondheim in northern waters' had not gone exactly as he intended. Admiral Tovey was, however, anxious, because 'no disposition of the Home Fleet could adequately protect both the Russian convoys and the northern passages from this threat'-that is to say from a powerful squadron based in Norway.
We must now turn to the Russian convoys, which occupied so large a place in the plans and operations of the Home Fleet at this time. Admiral Tovey was reluctant to use his full strength to cover them throughout their long, outflanked passage, because he could only do so if the northern passages to the Atlantic were left unguarded. The Admiralty, however, considered the risk acceptable and pressed the Commander-in-Chief to afford the convoys the most powerful cover possible.
The early convoys of 1942 fared well. PQ 7, which had been delayed by defects and sailed finally from Hvalfiord in two parts on the 31st of December 1941 and the 8th of January 1942, consisting of two and nine ships respectively, lost only one of its number. PQ 8 had one ship damaged by torpedo off Kola Inlet, but all the eight merchantmen managed to reach their destination. The Matabele, one of the two escorting destroyers was, however, sunk on the 17th of January; and from her company there were only two survivors - an unpleasant reminder of the very poor prospects for the crews of ships sunk in those ice-bound waters. The next three convoys, PQ 9 and 10 (which sailed from Hvalfiord together on the 1st of February and totalled only ten ships), and PQ 11 of thirteen ships from Loch Ewe, were never located by the enemy, and got through unscathed. But it was realised that this run of good luck could not last much longer. The lengthening days were steadily depriving the convoys of the friendly shield of darkness; yet for two or three more months they would be forced by ice to pass close off the enemy coast. To Admiral Tovey it was as unacceptable as it was unnecessary that U-boats should be able to lie in wait off the entrance to Kola Inlet, The Russians should, in his opinion, easily be able to make such confined waters untenable to them; and he considered that proper fighter protection should be given to the approaching convoys by our Ally from his shore airfields. The cruiser Nigeria (flagship of Rear-Admiral
H. M. Burrough, commanding the 10th Cruiser Squadron) was at Murmansk all February with the object of representing these needs to the Russians, and of covering convoys between Bear Island and Kola Inlet or vice versa.
Then occurred the concentration of enemy surface forces at Trondheim already described. In Admiral Tovey's opinion attack by any or all of them was now to be expected between Jan Mayen and Bear Island8, while the eastern part of the route would, he considered, probably be left to U-boats and aircraft. This meant that cover by heavy ships had to be given throughout the first half of the journey; and to accomplish this at the minimum of added strain the Commander-in-Chief asked, on the 26th of February, for the outward and homeward convoys to be sailed simultaneously. PQ 12 and QP 8, which sailed from Iceland and Kola Inlet on the 1st of March, and were both comparatively large convoys of sixteen and fifteen ships respectively, were the first to be covered by the main Home Fleet. At the same time close escorts were strengthened, Coastal Command took special steps to watch the Trondheim fiords closely, and Liberators flew long-range patrols to the north-east from Iceland. Nor were these measures introduced a moment too soon. On the evening of the 6th the Seawolf, one of our submarines on patrol off Trondheim, reported that a large enemy warship, either a battleship or a heavy cruiser, had sailed. It was actually the Tirpitz, flying the flag of Vice-Admiral Ciliax, with three destroyers in company. Her departure was missed by our air patrols. We now know that a long-range Focke-Wulf had given the enemy the position of the outward convoy at noon on the 5th of March, and that it was her report which caused the squadron to put to sea.9 That, however, was the end of the usefulness of the German air reconnaissance.
On the forenoon of the 6th the King George V (fleet flagship), Duke of York, Renown (flagship of Vice-Admiral A. T. B. Curteis, second in-command, Home Fleet), Victorious, Berwick and twelve destroyers had concentrated under Admiral Tovey. Until the following afternoon the main fleet patrolled on a line some 50-100 miles south of the convoys' routes, during the dangerous period when they were passing each other. Soon after midnight on the 6th-7th Admiral Tovey received the Seawolf's report, relayed to him by the Admiralty. At 8 a.m. the main fleet raised steam for full speed and turned more to the east. The Victorious had been warned to be ready to make an air search to the south of the convoy routes, but unhappily - for such a search 'would almost certainly have located the Tirpitz' - the weather prevented flying all day. Meanwhile, the Tirpitz, which was
Map 12. The Sortie of the Tirpitz in the Arctic - Phase 1. Map 12. The Sortie of the Tirpitz in the Arctic - Phase 2.
also unable to use her reconnaissance aircraft, was closing our main fleet.
At noon on the 7th, when the two convoys were passing each other some 200 miles south-west of Bear Island, they, the enemy and the Home Fleet were all within about eighty or ninety miles of each other.10 Ciliax had no idea that powerful British forces were at sea, let alone so close to him. He, on the other hand, passed only a few miles astern of PQ 12 and ahead of QP 8. In the prevailing low visibility neither side's forces saw anything of the other's. The German destroyers, which had been detached to search separately to the north, also passed very close to the homeward convoy. At 4:30 that afternoon, the 7th, one of them sank a Russian merchantman, which had dropped astern of the homeward convoy. Not long afterwards Ciliax sent his destroyers back to fuel. He continued to search with the Tirpitz alone.
The Russian merchantman's distress message was intercepted by Admiral Tovey, but the sender's position was not clear to him. However, wireless bearings of an enemy vessel, which might be the Tirpitz, caused the Commander-in-Chief to alter to the east at 5:50 and to the north-east at 8 p.m. At the latter time he detached six destroyers to spread and sweep along the enemy's most probable return route. They searched to the north from 2 a.m. to 6 a.m. on the 8th, but sighted nothing. It will be seen from Map 12 (Phase 1) that while this game of blind man's buff was in progress the Tirpitz was actually still some 150 miles to the north of the Home Fleet. She was now a long way astern of QP 8, and the outward convoy's north-easterly course was taking it away from the enemy. As the gap between the convoys widened, so did the greatest danger recede.
Having heard nothing more of his quarry Admiral Tovey turned south at midnight on 7th-8th, to get his carrier aircraft within striking range off the Lofoten Islands at dawn. Four hours later he came to the conclusion that he had missed the enemy and, having no destroyers to screen his ships in these dangerous waters, he turned towards Iceland 'to collect some destroyers'. This westward movement temporarily took the fleet directly away from the enemy, and the Commander-in-Chief's guess that the Tirpitz had already slipped home past him was wrong. Nor was the outward convoy by any means yet out of danger, to the south of Bear Island, since Admiral Ciliax searched to the north during the forenoon of the 8th, and then turned west.11 At noon he passed only about eighty miles south of the convoy, which thus had a second narrow escape. The Admiralty had meanwhile ordered that convoy to pass north of Bear Island; but ice conditions determined the senior officer of the escort and the Commodore
to disregard an order which could not even be attempted without grave risk of damage to the ships; for heavy pack-ice had been encountered while still to the south of the island. In fact, from noon on the 8th till early next morning the convoy steered east or south-east, more or less along the edge of the pack-ice. By the afternoon of the 8th the Home Fleet was some 500 miles to the south-west of the outward convoy. The Admiralty, however, considered that the Tirpitz might still be seeking it in the waters south of Bear Island-as in fact she was. At 5:30 p.m. a signal from London to that effect caused the Commander-in-Chief to turn back to the north-east. It was, we now know, at 8 o'clock that evening that Ciliax abandoned the search and turned south towards the Lofoten Islands. The Admiralty's intelligence again proved accurate later that night, when they told Admiral Tovey that his quarry was now moving south. Accordingly at 2:40 a.m. on the 9th the Commander-in-Chief altered towards the Lofotens and increased speed. He was then about 200 miles to the west of the enemy.12 Reconnaissance aircraft were flown off from the Victorious at 6:40 a.m., followed by twelve torpedo-carrying Albacores. 'A wonderful chance' Admiral Tovey signalled to the latter: 'God be with you'. One of the searching aircraft sighted the Tirpitz at 8 a.m., and her report was picked up by the leader of the torpedo-bombers; forty minutes later the striking force itself made contact. The enemy, we now know, sighted the Albacores a few minutes after they had sighted him. Surprise was thus not achieved, and although the German account says that the attackers showed great 'determination and dash', it is a fact that their tactics lost a great deal of the advantage of conditions which were unusually favourable. The aircraft came in from astern and to leeward, which meant that, with the Tirpitz steaming into the wind, their relative speed of approach was far less than would have been the case had they attacked from ahead and to windward. The enemy thus gained time to take skilful avoiding action, and all the torpedoes missed. Two Albacores were lost, and the chance of bringing the battleship to action off the Lofoten Islands, in the waters where the Renown had fought the Scharnhorst and Gneisenau in April 1940, was lost to the Home Fleet.13 The Tirpitz anchored off Narvik that evening, and Admiral Tovey reached Scapa next day.
The failure of the striking force to slow down the Tirpitz, so that the heavy ships could bring her to book in the same manner as happened with the Bismarck14, was most disappointing. In fact, such an opportunity was never to recur. The tactics employed were undoubtedly open to criticism, but it is only right to mention that the
leader of the Albacores had only just taken over command, and had not even flown with his squadron previously. Moreover, both he and the other aircrews lacked the intensive training so necessary to success in air torpedo attacks. To be called on to carry out so critical an operation in such circumstances was a very severe, even unfair test. The only lesson that could be drawn from the failure was the well-known one that success in such attacks could only be achieved after prolonged individual and squadron training, and by experienced as well as gallant leadership.15 The rapid expansion of the Fleet Air Arm then in progress made it inevitable that its squadrons should be constantly diluted by semi-trained crews, but steps were now taken to ensure that a higher standard of training in torpedo attacks was achieved before individual pilots, or complete units, became operational.
There remained the possibility of catching the Tirpitz if and when she left Vestfiord to return to Trondheim. Eight destroyers were sent to sweep along the Norwegian coast south of Vestfiord early on the 13th; and submarines were disposed to catch her at the points where she had to leave the shelter of the Leads. But the Tirpitz actually sailed at 11 p.m. on the 12th and steamed south in very bad visibility at high speed, close to the coast.16 Although she certainly passed very near to some of our submarines none managed to get in a shot. She reached Trondheim at 9 p.m. the next evening; but the weather continued so persistently bad that it was not until the 18th that a Coastal Command reconnaissance aircraft located her in her old berth once more.
Thus ended the first foray by enemy heavy forces in the far north. The results gave the naval authorities of both sides much to ponder on. Although this time our convoys had come through almost unscathed, the realities of the danger in which they lay during the whole of the eastern section of their 2,000-mile journeys received new emphasis. It was plain that by the mere presence of his heavy forces in Norwegian waters the enemy would force us to hold great strength in readiness to deal with them. Reinforcements which were urgently needed elsewhere, and especially in the Indian Ocean, could not be sent out; and every convoy to North Russia would now involve us in a major fleet operation.17 The enemy, on the other hand, realised that it was only 'sheer good fortune' which had this time saved the battleship from damage, and that such luck would probably not be repeated. Admirals Ciliax and Raeder both stressed the risks they
ran in such sorties, chiefly because they lacked both an aircraft carrier and effective co-operation from shore-based aircraft. In consequence the German Naval Staff decided on a more cautious policy in the future. Their last battleship would be held in reserve against the anticipated invasion of Norway, and only their lesser ships would be committed in the north. Hitler, however, who always insisted that he could and would have the best of both worlds, ordered that the offensive against the supply route to Russia was to be intensified by air and U-boat attacks, as well as by raids by surface forces. With typical lack of realism he also ordered the lack of carrier aircraft to be rectified by the completion of the Graf Zeppelin 'forthwith', and other ships to be converted to auxiliary aircraft carriers.18
But the German command was not alone in finding difficulties in operations of this type. Admiral Tovey was critical of the Admiralty instructions that the protection of the convoys must be his main object, and that he was to provide fighter protection for all capital ships within range of enemy shore-based aircraft. Such orders appeared to him a radical departure from the tradition that the destruction of the enemy's principal forces should be the object of our fleet. To carry out such instructions he was forced to work his carrier and capital ships as one unit, often without a destroyer screen, in highly dangerous waters. On the present occasion Admiral Tovey, who considered the sinking of the Tirpitz 'of incomparably greater importance to the conduct of the war than the safety of any convoy', had found himself 'seriously embarrassed' by these instructions. Secondly, the Commander-in- Chief criticised in forthright terms 'the detailed instructions for the handling of his forces' which had been signalled from London. This was an old issue, from which Admiral Tovey's predecessor had also suffered, and which was fully discussed in our first volume.19 The circumstances of the Arctic convoys do, however, appear to have been somewhat different from those earlier cases in which the Admiralty's interventions had aroused the critical comment of Commanders-in-Chief. In the first place the intelligence derived by the Admiralty and sent to the fleet flagship was, we now know, more accurate than the appreciations made afloat. Neither the signalling of the intelligence, nor the issue of orders when the intelligence available in London indicated that the assumptions on which our forces' movements had been based were wrong (as happened on the evening of the 8th of March), is open to criticism. Secondly, communications in those waters were proved to be so difficult that Admiral Tovey himself had once broken wireless silence to ask the Admiralty to operate the cruisers and destroyers of his
fleet. True, some of the Admiralty's messages (such as that ordering the convoy to pass north of Bear Island) now seem to have been unnecessary, and even dangerous; but the conduct of the whole operation produced many novel problems for the Admiralty as well as for the Commander-in-Chief and his staff. If any conclusion is to be drawn, it is perhaps that, as was suggested in an earlier context, it is the extent to which interventions are made from London, rather than the principles involved in making them, which requires constant watchfulness ashore.20
We will now take leave temporarily of the distant and dangerous waters across which the war supplies, on which our Russian allies so greatly depended, had to be carried, in order to glance briefly at an operation which took place at this time further south. Readers of our first volume will remember how in January 1941 five Norwegian merchantmen, loaded with valuable cargoes, broke out from Sweden and safely reached British ports.21 In spite of the greater difficulties inherent in repeating a successful coup, plans to do so had long been brewing. On the 11th of March the Admiralty issued definite orders to those concerned; but several postponements were caused by the weather and by the enemy's obvious alertness to what was in the wind. Finally, in the small hours of the 1st of April ten Norwegian ships sailed from Gothenburg. It was not possible to give them surface ship protection for the first day's passage, but for the second day six destroyers were sent out, as well as Coastal and Fighter Command aircraft. The Germans reacted strongly, and many fierce air combats resulted over the North Sea. The final result was that five of the ships were sunk in the Skagerrak by surface forces, grounded, or were scuttled to avoid capture; one was so badly damaged by air attack that our destroyers had to sink her, two returned to Gothenburg and only two reached Britain safely. The results were disappointing after such a determined effort.
The next pair of Russian convoys, PQ 13 and QP 9, sailed from Reykjavik and Murmansk on the 20th and 21st of March respectively. The enemy had brought his revised plans into force and was organising heavier attacks by aircraft and U-boats. Moreover, his policy of concentrating his heavy ships in Norway had just been carried one step further by the transfer of the Hipper. She left Brunsbüttel on the 19th and, although our intelligence had once again detected what was afoot, neither our reconnaissance aircraft nor the torpedo striking force of Coastal Command Beauforts managed to find her. On the 21st she anchored in a small fiord near to the Tirpitz, and next morning the Home Fleet sailed from Scapa to restart the long and arduous process of covering two Russian convoys.
During their passage Coastal Command did its utmost to watch the four big ships lying in the Trondheim fiords. The weather greatly hindered our aircraft, and the enemy could certainly have evaded their patrols had he wished to do so. In fact, this time none of the four ships put to sea.
The arrangements to protect PQ 13 and QP 9 (both consisting of nineteen ships) were similar to those made for the preceding pair of convoys. The homeward convoy had a safe and comparatively uneventful passage. The only serious encounter with the enemy ended in the ramming and destruction of U-655 by the minesweeper Sharpshooter of the close escort. The passage of the outward convoy told a very different story; but all went well until the 24th. The cruiser Trinidad was providing close cover, and two destroyers from Iceland had by that time reinforced the close escort. Then a violent gale scattered the convoy far and wide. By the 27th not a single merchantman was in sight of the escort. Next morning the convoy was strung out over about 150 miles of the Arctic Ocean south of Bear Island; and the enemy's searching aircraft had reported its presence. Air attacks soon started, and the wide dispersal of the ships made their protection very difficult. In spite of this only two ships, both stragglers, were sunk by bombs that day, the 28th. But three German destroyers had sailed from Kirkenes on the strength of the first air reports and by the evening of the 28th were sweeping westwards along the presumed track of the convoy. Very early on the 29th they picked up a Panamanian straggler, whose survivors disclosed to the enemy a good deal about the composition and progress of the convoy and its escort. The enemy adjusted his sweep accordingly and, just before 9 a.m. encountered the Trinidad and the destroyers Fury and Eclipse of the escort. A series of sharp actions followed, in conditions of atrocious difficulty - low visibility, snow, and the spray freezing solid as fast as it came aboard. One enemy, the Z-26 was finally sunk. Unhappily, the Trinidad was hit by a torpedo and suffered serious damage. After a good deal of difficulty she made Kola Inlet on the 30th under her own steam. The subsequent removal of debris from her boiler rooms produced, quite unexpectedly, a relic from which it was established beyond doubt that she had, in fact, been hit by one of the torpedoes which she herself had fired at the German destroyers during the surface actions. It is almost certain that the extreme cold caused its steering mechanism to behave erratically. But, as Admiral Tovey remarked, 'it was cruel hard luck' for a ship which had just successfully fought off the enemy 'to torpedo herself'.
Meanwhile the convoy, still in several groups, plodded slowly eastward. Two ships got caught in the ice, and one gallant British merchantman, the Induna, towed a disabled comrade for more than a day, only to be sunk herself by a U-boat on the 30th. Another fell
victim to the same cause a short time later. On the 30th of March and 1st of April the fifteen surviving ships reached their destination. Five had been sunk-two by aircraft, one by destroyers and two by U-boats. Though the escort had sunk one enemy destroyer and also a U-boat (U-585, by the Fury on the 29th), the Trinidad and Eclipse had themselves been badly hurt. And there remained the comfortless fact that one quarter of the convoy had gone down.
Again both sides reviewed the Arctic struggle. The Germans were satisfied, and even claimed a 'notable success' in the operations just described. Admiral Tovey urged the need for stronger close escorts, and more destroyers and corvettes were promised for the next convoys. He also wanted to reduce the number of convoys during the spring and summer, when conditions would increasingly favour the enemy; but he realised that political considerations would probably make this unacceptable. He realised that the enemy was 'determined to do everything in his power to stop this traffic'. Yet efforts to persuade the Russians to make the air and anti-submarine protection more effective at their end of the route continued to produce 'little response'.
On the last day of March Bomber Command made another attempt to reduce the threat from the Tirpitz to the Arctic convoys. Thirty-three Halifaxes of No. 4 Group set out to attack her; but the weather was exceedingly bad, and the majority of the bombers failed to identify the target. Those that managed to drop their bombs did no damage, and five aircraft were lost.
The First Sea Lord himself had no illusions regarding the difficulties and dangers of the northern route. Early in April he represented the matter forcefully to the Defence Committee, and ended with a warning that geographical conditions were so greatly in favour of the Germans that losses might become so great as to render the running of these convoys uneconomical. But pressure in the opposite direction was being exerted in the highest quarters, in particular from President Roosevelt to the Prime Minister, and in consequence the next convoy (PQ 14) was made a larger one, of twenty-four ships. It sailed on the 8th of April and soon ran into heavy ice. Sixteen ships returned, and of the eight which went on one was sunk by U-boat. The corresponding homeward convoy, QP 10 of sixteen ships, lost four of its number. The German destroyers again tried to intervene, but the very bad weather frustrated their two attempts.
On the 28th and 29th of April our heavy bombers again attacked the Tirpitz, but on both occasions they were met by a very heavy volume of anti-aircraft fire, and had to attack through the dense smoke screen in which the Germans shrouded the battleship. No hits were obtained, and in these two operations we lost seven bombers. Early in May, after the Admiralty had repeatedly represented the
need for the Russians to accept a larger share of the responsibility for the safety of the convoys during the latter part of their journeys, the Prime Minister telegraphed to Stalin pointing out how essential this was. Stalin replied that his 'naval and air forces would do their utmost' on the section of the route east of the meridian of 28° East, but pointed out how small were the forces which he could make available. Though there are no grounds for suggesting that, within the limits imposed by their somewhat primitive conceptions of maritime war, the Russians did not do what they could with what they had, it is none the less. the case that they never relieved the Home Fleet of any appreciable share of the responsibility for defending any Arctic convoy.
PQ 15 and QP 11, of twenty-five and thirteen ships respectively, sailed at the end of April. The dispositions followed in general those made for their predecessors, but an anti-aircraft ship joined the close escort, and the covering force included American as well as British warships. We will briefly follow the homeward convoy first. The covering cruiser was the Edinburgh (flagship of Rear-Admiral S. S. Bonham-Carter, commanding the 18th Cruiser Squadron), and the close escort comprised six destroyers, four corvettes and a trawler - far greater strength than had so far been employed. In addition British minesweepers and two Russian destroyers kept company for the first stretch of the convoy's passage. On the 29th its presence was reported by enemy aircraft and U-boats, and next afternoon the Edinburgh was hit while zig-zagging ahead of the convoy by two torpedoes fired by U-456. Her stern was blown off, and she started back towards Murmansk, 250 miles away, at slow speed escorted by two destroyers. The U-boat meanwhile shadowed the lame cruiser, and the weakening of the convoy escort encouraged the enemy to send three destroyers to sea that night. On the afternoon of the1st of May, after an air torpedo attack had been unsuccessful, the German destroyers appeared on the scene. They made no less than five separate attempts to reach the convoy, but were each time foiled by the aggressive tactics of the far weaker British escort, which was most ably led by Commander M. Richmond in the aptly-named Bulldog. One of our small force, the Amazon, was damaged, and one Russian merchant ship, which had straggled, was sunk by an enemy destroyer's torpedo.22 But that was all. Throughout the afternoon, from 2:00 p.m. till nearly six o'clock, the enemy's repeated lunges at the convoy were successfully driven off. Finally, the Germans abandoned the attempt and went off to find the damaged Edinburgh. To Commander
Richmond's congratulatory signal to his consorts one of them instantly replied, 'I should hate to play poker with you'; and there is indeed no doubt that he thoroughly outfought the enemy's 'three of a kind'. QP 11 reached home without further trouble.
Meanwhile the Edinburgh, unable to steer except with her engines, and also unable to be towed, was making very slow progress eastwards. On the evening of the 1st she was joined by four minesweepers, but early the following morning the German destroyers found her. A series of confused fights followed, and the cruiser herself, for all her disablement and grievous trouble, managed to hit and stop the large destroyer Hermann Schoemann. But the Forester was also heavily hit, just at the moment when the enemy had fired torpedoes. By ill luck one of these, almost at the end of its run, hit the Edinburgh, which was unable to take any avoiding action, amidships on the opposite side to her earlier damage. The ship was thus almost cut in two. She continued to fight her armament - and one enemy described her gunfire even then as being 'extraordinarily good'-but she was plainly doomed. Another misfortune followed quickly, when the Foresight, the last effective destroyer, was badly hit and brought to a standstill. There were thus three British ships all lying stopped at the same time, and all with much of their armament out of action. The two surviving enemies could have finished them off at leisure, but chose instead to take off the crew of the damaged Schoemann. This they successfully accomplished, the Schoemann sank and the other German destroyers then withdrew. The Forester and Foresight next managed to get under way at slow speed, the minesweepers took off the Edinburgh's crew, among whom casualties were remarkably light, and the cruiser was then sunk by one of our own torpedoes. The enemy had undoubtedly scored a success; but he might have annihilated our whole force had he not mistaken the minesweepers for destroyers and, we now know, greatly overestimated the opposition by which he was faced.
Meanwhile the east-bound convoy PQ 15 had entered the critical part of its passage. It was powerfully escorted, and covered by Admiral Tovey's full strength. Up to the 2nd of May no losses had been suffered. On that day the west-bound convoy (QP 11) was passed, and a gloomy prognostication of what probably lay ahead was received from its escort. Enemy shadowing aircraft were soon in touch, and in the half light of the small hours of the 3rd six torpedo bombers came in low. They sank three ships. The U-boats never succeeded in getting in an attack, and subsequent bombing caused no more losses. On the 4th, visibility closed right down, and shielded the convoy for the rest of its passage. Of the thirty-eight merchantmen involved in this double movement only four were lost; but casualties among the escorting warships had been heavy. In addition
to the loss of the Edinburgh and the damage to the Foresight and Forester, the destroyer Punjabi had been rammed and sunk in low visibility by the King George V on the 1st of May, and the Polish submarine P-551 was destroyed by our own forces when she was suddenly encountered nearly 100 miles from her patrol area, near to PQ 15, on the 2nd. Nor was the story of our losses yet ended, for on the 13th the damaged Trinidad left Murmansk escorted by four destroyers and covered by powerful forces. She was quickly sighted by enemy reconnaissance planes, and on the evening of the 14th air attacks started. Some twenty-five bombers and one torpedo attack failed to damage her, but at 10:45 p.m. a lone Ju-88 dived out of low clouds, and scored a hit with a bomb not far from where she had previously been damaged. This started a serious fire, which spread rapidly; her condition was made more precarious by a near miss blowing a temporary patch off her side and causing flooding. She was still able to steam, but by midnight the fire was out of control and it was decided that, situated as she was far from any friendly port, in the presence of U-boats, and in certain danger of renewed attack by aircraft, salvage was impossible. She was sunk by our own torpedoes at 1:20 a.m. on the 15th. Thus did Admiral Bonham-Carter suffer the loss of two valuable cruisers, both of them his flagship, within a matter of two weeks. Admiral Tovey fully endorsed his recommendation that unless the airfields in north Norway could be neutralised, or some cover obtained from darkness, the convoys should be stopped. 'If', he went on, 'they must continue for political reasons, very serious and heavy losses must be expected'. Nor did the First Sea Lord disagree. On the 18th of May he wrote to his American colleague, Admiral King, about these convoys, saying that 'the whole thing is a most unsound operation with the dice loaded against us in every direction'; and Admiral King replied in sympathetic agreement. Mr Churchill has revealed the extent to which political pressure overruled such strong professional opinion.23
It thus came to pass that, far from the convoys being suspended, the next one (PQ 16) of thirty-five ships was the largest yet sailed - and that despite the time of year being now even less favourable. It was plain to all involved in the work of planning the convoys and the associated fleet movements and in the long-drawn anxieties of their execution, that we were gambling with fate to an extent which was bound, sooner or later, to provoke nemesis. All realised that a disaster was likely; but when and on which convoy would it fall? Though, as we realised, the threat from German destroyers was now more or less eliminated, the Scheer had moved north to Narvik, and had been joined there by the Lützow on the 26th of May. To
escort the convoys with cruisers throughout the passage was therefore deemed less essential, but four cruisers and three destroyers were detailed for close cover against the pocket battleships west of Bear Island. The battle fleet again provided more distant cover against the Tirpitz. PQ 16 sailed on the 21st, as did QP 12 of fifteen ships. Early on the 25th the covering cruisers Nigeria, Norfolk, Kent and Liverpool, under Rear-Admiral H. M. Burrough, with three destroyers, had joined the east-bound convoy and greatly reinforced its escort. The first shadower promptly arrived, and thereafter for five continuous days the convoy was hardly ever unaccompanied by a watchful enemy reconnaissance plane. On the afternoon of the 25th it passed the homeward convoy, and soon afterwards there began an air battle which lasted throughout virtually the whole of the rest of the journey. Torpedo-bombers (He 111s) alternated with dive attacks by Ju 88s; but the first victim fell to a U-boat early on the 26th. A Catapult Aircraft Merchantman (C.A.M. Ship), the Empire Lawrence, had been included in the convoy, and her single Hurricane destroyed one enemy and damaged another.24 The gunfire of the powerful escort proved effective in holding off the attackers and in destroying some of them, while the anti-submarine escort constantly harassed and chased away the U-boats. But on the 27th, after the cruiser force had left the convoy, yet heavier air attacks took place. The A.A. ship Alynbank recorded, with, as we now know, complete accuracy, attacks by 108 aircraft that day. They were generally pressed well home; four merchantmen, including the C.A.M. ship, were sunk, and two others and the Polish-manned destroyer Garland badly damaged. The small escort vessels rescued survivors, even while they themselves were being bombed.
Late in the evening heavy attacks were renewed. Two more ships went down and the Commodore's ship, the Ocean Voice, was set on fire and badly holed. 'I had little hope of her survival', wrote Commander R. Onslow, senior officer of the escort, 'but this gallant ship maintained her station, fought her fire, and with God's help arrived at her destination'. In the escorts ammunition was beginning to run low; yet there were three more days, and twenty-four hour days, too, to be endured. 'We were all inspired', continued Commander Onslow, 'by the parade-ground rigidity of the convoy's station-keeping, including the [damaged] Ocean Voice and the Stari Bolshevik [a Russian merchantman], who were both billowing smoke from their foreholds'.
However, the worst was actually over. One more damaged ship went down on the 28th, but a welcome reinforcement of three Russian destroyers arrived. More attacks followed, but no more
losses were suffered. Next evening six British minesweepers from Murmansk arrived, and the six ships destined for Archangel were detached. The expected U-boat attacks did not occur, and on the afternoon of the 30th convoy PQ 16 'reduced in numbers, battered and tired, but still keeping perfect station' entered Kola Inlet.25
While the east-bound convoy was being subjected to this prolonged ordeal, the fifteen ships of the west-bound QP 12 had a comparatively uneventful passage. Apart from one Russian ship which had to return, it arrived intact at Reykjavik on the 29th of May. Of the fifty ships which started out on the double journey only seven were lost. 'This success was beyond expectation', wrote Admiral Tovey, and gave high praise to the officers and men of both escorts and merchantmen. Dönitz himself paid tribute to the work of the Allied escorts, and admitted that his favourite weapon (the U-boat) had failed him. The Luftwaffe had, with great exaggeration, claimed that the convoy had been totally destroyed. This misled Dönitz into recommending that aircraft rather than U-boats should be used against the summer convoys. On our side, Commander Onslow urged that many more C.A.M. ships or an escort carrier, and more A.A. ships as well should be included in the escort of future convoys. It was indeed realised that, in face of the air strength now deployed by the enemy in north Norway, anti-aircraft defence must take equal precedence with anti-submarine measures and protection against surface attack.26 Thus was catastrophe in the Arctic deferred-but not for long.
When, in the spring of 1942, the Russian convoys loomed so large in the responsibilities of the Home Fleet and the problems of their defence were of constant concern to the Admiralty, it was natural that the eyes of the British authorities should once more be turned towards the island of Spitzbergen. It was important to prevent the enemy establishing any form of base there, and especially from stationing aircraft in its bays. In August 1941, after we had evacuated all the Allied inhabitants27, it was known that a German meteorological
party had been set up ashore; but we had never been able to spare the forces to turn them out, or to re-occupy the island. Now, in May 1942, after a preliminary and very difficult reconnaissance by a Catalina of Coastal Command, a small Norwegian expedition sailed from Iceland. Unfortunately, enemy bombers sank both its ships before all the stores had been unloaded. There now followed a curious period of hide and seek among the fogs which so often shroud this remote Arctic island; for there were Allied and enemy expeditions ashore in different places, and each side tried to support and supply its own party and attack the other's. On our side the denial of Spitzbergen to the enemy owed much to a series of remarkable flights by a Catalina of No. 210 Squadron of Coastal Command, commanded by Flight-Lieutenant D. E. Healey. In the most difficult conditions conceivable, during flights which generally lasted about twenty-four hours, he carried supplies to the stranded Norwegians, attacked the enemy base on Spitzbergen, picked up some of the Allied party, sighted survivors of ships sunk in Russian convoys and performed a dozen other various duties. Unhappily, he was killed in September, in a chance encounter with a German bomber off Kola Inlet. Warships accompanying Russian convoys were several times diverted to relieve, reinforce and supply the Norwegian expedition. Thanks to all these various measures, by the autumn it was plain that we had prevented the enemy establishing himself in Spitzbergen, and had a reasonable hold on it for our own use.
It must not be thought that Coastal Command's long-range reconnaissance and escort work was the only way in which the Royal Air Force tried to mitigate the dangers of the Arctic route. Bomber Command did its best to put the Tirpitz out of action while she lay near Trondheim. It has already been told how, on the last day of March and twice during April, strong forces of from thirty to forty-five heavy bombers were sent to attack her28; but no hits were obtained on any of these missions.
Our experiences in defending PQ 16 and earlier convoys to Russia had emphasised how difficult it was to carry out the air patrols off north Norway, which were essential to obtain early warning of enemy warship movements, as long as they had to be sent out from British bases. Early in June the Commander-in-Chief, Coastal Command, suggested to the Air Ministry that the establishment of a flying boat base in Kola Inlet would greatly ease such difficulties. The feasibility of providing a similar base at Advent Bay in Spitzbergen was also discussed, but in that case the difficulties were finally found to be insuperable. Another way of increasing the threat to the German warships in the North was to station torpedo-bombers near Murmansk. The Admiralty told the Senior British Naval Officer, North
Russia (Rear-Admiral R. H. L. Bevan) to investigate the latter, but they insisted that the Russians must not treat such a proposal as relieving them of responsibility for the defence of our convoys at the end of their journeys. On the 7th of June the First Sea Lord also told Admiral Miles, the head of our naval mission in Moscow, that if the convoys were to continue the Russians must make a proper contribution to their defence by such steps as bombing enemy air stations, and keeping submarines on patrol east of Bear Island. One-fifth of our losses were incurred at the Russian end of the Arctic route, and Admiral Pound considered that a large proportion of them had been avoidable. Admiral Miles replied that the Russians would welcome our torpedo-bombers, and intended to devote all their resources in the north to improving the defence of the convoys. Though the acute shortage of torpedo-carrying aircraft in Coastal Command prevented the proposal being carried out29, by the end of June arrangements had been made for Catalinas of Nos. 210 and 240 Squadrons to patrol off north Norway, land at Russian bases and work from them for a time before returning to Britain.
Owing to the desperate need to relieve Malta, considerable strength had to be detached from the Home Fleet early in June, to help fight through a convoy.30 It was nearly the end of the month before the survivors returned to Scapa, and meanwhile no Russian convoy could be run. There was thus a lull for the rest of the Home Fleet, during which H. M. King George VI came to Scapa. He stayed in the fleet flagship Duke of York, and visited several of the ships present, including the U.S.S. Washington, flagship of Rear-Admiral R. C. Giffen, U.S.N., who was in command of the American Navy's Task Force 99. But the interlude was a brief one. PQ 17 and QP 13, of thirty-six and thirty-five ships respectively, sailed on the 27th of June, except for the Archangel section of the latter which started a day earlier.
From what has already been written the reader will have understood the profound misgivings with which the continuation of the Russian convoys throughout the summer, and especially of large convoys, was regarded by the officers responsible for their safety. They accepted the need to carry on with them because the political leaders of the Allied nations desired it, and because they knew that a hard-pressed Russia was clamouring for the munitions which had been promised her, and which were steadily piling up in British and American ports. But they did so without any illusions regarding the dangers of the undertaking. So far our forces had not had to deal
simultaneously with all the enemy's weapons-his heavy ships, his light surface forces, his aircraft and his U-boats. Different convoys had been threatened in the Barents Sea by one, two or even three of these four; but never by all four at once. Now, unknown to the Admiralty, the German Naval Staff had just decided to commit the Tirpitz to the attempt. True, the instructions issued with Hitler's approval to Admiral Schniewind, the new Commander-in-Chief afloat, were hedged with such cramping restrictions as would have eased the Admiralty's anxiety, and probably altered their actions, had they known of them. But they could not know that Raeder had given warning that a naval reverse at this time was particularly undesirable, nor that Group Command North had told Schniewind 'on no account to allow the enemy to score a success against the main body of the fleet'. The Tirpitz and Hipper, with four destroyers, were now at Trondheim; and the Scheer and Lützow, with six destroyers, were at Narvik. 'The strategic situation', wrote Admiral Tovey, 'was wholly favourable to the enemy'; and apart from submarine attacks off the coast he could see no way of mitigating it, except by tempting the enemy heavy forces to attack further to the west. A suggestion that this might be accomplished by turning the convoy back on its tracks temporarily did not meet with Admiralty favour, though they agreed that in certain circumstances they themselves might order it. Their instructions laid down that, to the west of Bear Island, our surface forces would be responsible for the convoy's protection against attack by heavy ships; to the east of that mark our submarines must meet the need. The cruiser covering force was not to go east of Bear Island, unless the threat to the convoy consisted of a surface force which it could fight-that is to say, a force which did not include the Tirpitz; nor in any case were the cruisers to go east of 25° East. These instructions did not altogether appeal to Admiral Tovey; and we now know that the Commander-in-Chief was very near the mark in holding that, particularly after his experiences against PQ 12 and QP 8, the enemy would not again risk committing the Tirpitz to an attack on a convoy in the Barents Sea. He and the First Sea Lord discussed the new convoy operation, regarding which they were not wholly in agreement, on the telephone from Scapa to the Admiralty. Admiral Tovey had always disliked sending heavy cruisers into the Barents Sea with the convoys. They could not be given adequate protection against U-boats or air attack and, if they were damaged so many miles from home bases, to extricate them was bound to be difficult; for there were no proper facilities at Murmansk for repairing damaged ships. The recent loss of the Trinidad and Edinburgh from combinations of these causes had lent support to his view that the risks were too great; but the Admiralty still considered cruiser support for the smaller ships of the close escorts essential.
With particular regard to the forthcoming operation Admiral Tovey represented that, apart from the time of year being unsuitable, the close escort was too weak and the convoy too large. If it must be sent, he considered that it should sail in two sections. The First Sea Lord was, however, insistent that the operation should take place as planned. Though no record was kept of these conversations, Admiral Tovey's recollections are clear on one other important point. He first learnt by this means of the possibility of the Admiralty ordering the convoy to scatter, if it appeared to be in imminent danger. Nor is there any doubt that this suggestion shocked the Commander-in-Chief deeply, because all his experience had been in exactly the opposite sense. It had, in fact, been repeatedly shown in all theatres, and very recently confirmed with respect to the Russian convoys, that so long as close order and disciplined movements were maintained, the merchantmen and escorts could afford each other effective mutual support; but once a convoy lost cohesion, its individual ships fell an easy prey to whatever enemy next found them.
The general dispositions made for PQ 17 corresponded to those which had proved successful on the last occasion. Four cruisers under Rear-Admiral L. H. K. Hamilton, the London (flagship) and Norfolk and the American ships Tuscaloosa and Wichita with three destroyers, were to provide close cover as far as Bear Island. The Commander-in-Chief in the Duke of York with the Washington (American battleship), Victorious, Nigeria, Cumberland and fourteen destroyers formed the distant covering force in the waters north-east of Jan Mayen Island. It should be mentioned that this was the first occasion on which substantial American forces were placed under British orders for an operation of this nature. Mr. Churchill later 'surmised' that this new factor may have influenced Admiral Pound's actions31; but no indication that this was the case has been found in the Admiralty's subsequent investigations, nor does the memory of staff officers who were close to the First Sea Lord lend support to the suggestion.
An attempt to deceive the enemy by sailing a false convoy ahead of the real one was unsuccessful, because the Germans never sighted it. Meanwhile, Hitler's overriding powers had produced still more precautionary orders, and of such stringency that to attack the convoy at all with the heavy ships was made virtually impossible; for Raeder was told that before the ships sailed he must have ascertained the disposition of our aircraft carriers, in order that the Luftwaffe might attack them. This restriction was almost certain to delay sailing the heavy ships until it was too late to attack the convoy; but the Admiralty could not possibly have been aware of this. Raeder tried to overcome the handicap thus imposed by transferring the ships to a temporary base in the extreme north as soon as
Map 13. Convoys PQ 17 & QP13, 3rd - 6th July 1942.
the convoy was known to have sailed. He hoped to get Hitler's final sanction to the operation while they were there, and thus gain enough time to attack. Admiral Schniewind issued his plan on the 14th of June. As soon as the convoy was known to be approaching, the Narvik force (the Lützow and Scheer) would move to Altenfiord32, and the Trondheim force (the Tirpitz and Hipper) to Vestfiord. The two squadrons would sail as soon as the convoy had passed the meridian of 5° East, and would meet each other 100 miles north of North Cape. The attack would be made when the convoy was between 20° and 30° East. Reconnaissance by U-boats and aircraft was arranged in order to give early warning of our movements.
The convoy was routed further north than before, since the summer limit of the ice made it possible to pass north of Bear Island. This route was, of course, longer, but it kept the convoy further away from the enemy air bases in north Norway. All the merchantmen were destined for Archangel, because Murmansk had been put out of action by bombing. The convoy was in the charge of Commodore J. C. K. Dowding, R.N.R., in the River Afton. An oiler was, as usual, included in order to refuel the escorts of the homeward as well as the outward convoy. The escort for the first part of the journey, of three minesweepers and four trawlers, left with the convoy; on the 30th the long-range escort of six destroyers, four corvettes and two submarines under Commander J. E. Broome in the Keppel joined up, as did the two anti-aircraft ships Palomares and Pozarica. There were also three rescue ships. On the 1st of July U-boats and shadowing aircraft made contact, but attacks by the former were all driven off. The outward and homeward convoys passed each other that after noon in 73° North 3° East, and in the evening an unsuccessful air torpedo attack took place. Admiral Hamilton with his four cruisers had now overtaken the convoy. He was keeping out of sight some forty miles to the north of it in order 'to keep the enemy guessing' as to his whereabouts. From the evening of the 2nd till the following forenoon the convoy was protected by fog. At 7 a.m. on the 3rd course was altered due east, to pass Bear Island and enter the Barents Sea.33 Shadowing aircraft were temporarily thrown off the scent. Meanwhile the Admiralty had reported that the ice edge was further north than had been anticipated, and the convoy altered somewhat in that direction at Admiral Hamilton's suggestion. Commander Broome, however, was anxious to make eastward progress as quickly as he could, and did not therefore fully accept the cruiser Admiral's proposed northward diversion. By 10-15 that evening, the 3rd, the convoy was thirty miles north of Bear Island. Our reconnaissance aircraft had meanwhile discovered that the
German Trondheim force had sailed, but no news had yet been obtained regarding the Narvik squadron. Admiral Tovey and the battle fleet were approaching their covering position, and Admiral Hamilton decided to exercise the discretion allowed to him and remain for a time with the convoy after it had passed Bear Island. So far all had gone very well indeed.
But the enemy forces were both meanwhile on the move. The Tirpitz and Hipper arrived in the Lofoten Islands on the 3rd, and the Scheer reached Altenfiord as well. The Lützow and three destroyers of the Tirpitz's group had all run aground near Narvik, and they took no further part in the operation.34
Early on the 4th PQ 17 suffered its first loss when a single aircraft torpedoed an American merchantman 'through a hole in the fog'. At about noon the Admiralty gave Admiral Hamilton discretion to carry on east of the limit of 25° East laid down in his orders 'should [the] situation demand it'. Admiral Tovey, who had no intelligence to justify the change of plan, qualified the Admiralty's message by telling Hamilton 'once the convoy is east of 25° East or earlier at your discretion you are to leave the Barents Sea unless assured by [the] Admiralty that [the] Tirpitz cannot be met'. At 6 p.m. Hamilton reported his intention of withdrawing at 10 o'clock, but at 7:30 the Admiralty signalled to him 'Further information may be available shortly. Remain with convoy pending further instructions'. We will return shortly to the 'further information' referred to in the Admiralty's message. The next development was a more serious attack by some two dozen torpedo aircraft at about 8:30 p.m. Three ships were hit, and two of them had to be sunk by the escort; the third, a Russian tanker with at least one woman in the crew, was found, in Commander Broome's words, to be 'holed but happy and capable of nine knots'. She eventually reached port. The convoy and escort defended themselves and each other with splendid discipline, and with good results. All felt that 'provided the ammunition lasted PQ 17 could get anywhere'.
In Germany Hitler's approval for the departure of the heavy ships was still lacking; but Raeder had ordered the Tirpitz to join the pocket battleships in Altenfiord, so that no time should be lost if the Führer's approval was forthcoming. The main body of the Home Fleet had not been sighted since early on the 3rd, so that the restriction imposed by Hitler, forbidding the employment of the battleship while the whereabouts of our aircraft carriers was unknown, still held good. Raeder decided he could do no more, and the German ships remained in Altenfiord till the afternoon of the
5th. So much for the enemy's actual dispositions and intentions. Let us now see how they appeared to the Admiralty at the time.
Between the 1st and 4th of July a number of Catalinas of No. 210 Squadron had flown to North Russia, making very thorough reconnaissances off the Norwegian coast on the way. Continuous patrols by these and by home-based aircraft were arranged during the critical period of the convoy's progress. Because of an accident to an aircraft there was, however, a gap in the air patrols from 11 a.m. to 5 p.m. on the 4th of July, and it is likely that the uncertainty produced by this failure influenced the Admiralty's subsequent actions. On the afternoon of the 4th of July our intelligence suggested that although there had been no verification of the photographic reconnaissance which had revealed that the German warships had left Vestfiord, it was 'tolerably certain' that the Scheer and Lützow were at Altenfiord. There had been no news of the Tirpitz and Hipper since 2 p.m. on the 3rd, when they were known to have left Trondheim. Thus by the afternoon of the 4th all four heavy warships might have been at sea making for the convoy, and at a time when our long-range air reconnaissance was known to have temporarily failed. The anxiety which this state of affairs must have produced in London is easily to be understood.
At about the time when PQ 17 was repelling the torpedo-bomber attack (8:30 p.m. on the 4th) the First Sea Lord called a staff meeting at which the various possibilities were fully discussed. By that time it was known that the Tirpitz had joined the Scheer in Altenfiord, and it was therefore considered that surface ship attack might take place at any time after 2 a.m. next morning. It seemed to the Naval Staff that it could only result in Admiral Hamilton's cruisers, the convoy and its escort all being overwhelmed. On the other hand, the convoy still had 800 miles to go, and the enemy aircraft and U-boats would find things much easier for them if the convoy dispersed. The surface attack was held to be the greater of the two dangers and, shortly after 9 p.m. a signal was sent to Admiral Hamilton 'Most Immediate. Cruiser force withdraw to westward at high speed'. This was followed at 9:23 by 'Immediate. Owing to threat of surface ships convoy is to disperse and proceed to Russian ports' and, at 9:36, by 'Most Immediate. My 9:23 of the 4th. Convoy is to scatter'.
Responsibility for the main decision lay, of course, on the First Sea Lord's shoulders. But it must be made clear that so critical a decision was not quickly taken by him.35 He and the Naval Staff had previously discussed the problem in all its aspects, but memories are not unanimous regarding whether any opposition was then expressed. It is known that the Vice Chief of Naval Staff (Vice-Admiral H. R.
Moore) pointed out that if the convoy was to scatter it must do so soon, because the further east it steamed the less sea-room would it have in which to scatter.
Three important points must be discussed before we turn to the tragic and disastrous sequel. Firstly, the order to disperse was based on anticipation of the enemy's intentions. It was not known whether the enemy surface ships were already at sea and threatening the convoy; but it was reasonable to suppose that such was their intention. To scatter the convoy would certainly incur grave dangers, and the decision to do so gambled on these being less than the risks from surface attack, in spite of the fact that the latter had not yet arisen, and might never arise. Secondly, the Admiralty issued a categorical order without telling those who would have to carry it out whether it was based on positive or negative intelligence. The 'threat of surface ships' mentioned in the second signal was practically meaningless; for such a threat was known to have existed for the past several days. Whilst making every allowance for the strain and anxiety felt in London, it is hard to justify such an intervention, made in such a way. If it was felt that there was a possibility that dispersing the convoy would turn out to be the less perilous action, such a proposal, and the grounds on which it was made, could justifiably have been sent to the responsible officers, for them to carry out or not as they saw fit. It is beyond doubt that had this been done the convoy and escort would have been kept together. Thirdly, the manner in which the decision was signalled by the Admiralty was almost bound to convey a false impression to the recipients of the three messages. In fact, emphasis was placed on the use of high speed in the first signal, only because U-boats were reported to be concentrating on the cruisers' withdrawal route; the distinction between 'disperse' in the second and 'scatter' in the third was merely a technical amendment36; and that the final message had a more urgent priority (Most Immediate) than its predecessor appears to have been an error or mischance in the drafting of it. To Admiral Hamilton and Commander Broome, however, the three signals, read together, were bound to signify firstly that they constituted the 'further information' promised to them in the earlier message; secondly, that a moment of extreme urgency, demanding drastic action, had arrived, and thirdly, that the enemy surface forces were really close at hand. None of these deductions was in fact correct. A decision, the wisdom of which was doubtful from the start, was thus made disastrous when translated into action.
Map 14. Convoy PQ 17, Approximate Movement of Ships, 4th - 28th July 1942.
The two senior officers both expected the enemy's masts to appear above the horizon at any moment, and for a desperate action to be joined. Commander Broome therefore took his six destroyers to reinforce Admiral Hamilton's cruisers; he left the two submarines of the escort with the convoy to attack the enemy warships, and ordered the rest of the escort (the A.A. ships, minesweepers, corvettes and trawlers) to proceed independently to Archangel.37 At 10:15 p.m. the order to scatter was passed to the Commodore, and Commander Broome took what he described as the hardest decision of his life - to leave the convoy.
To Commodore Dowding the order to scatter his convoy came as such a surprise that he asked for it to be repeated. 'It must', wrote Broome later, 'have come as a shock to him : he was sharing the wave of confidence which swept through the convoy and escort after the air attack . . . The tails of PQ 17 were well up'. Commander Broome's actions were subsequently fully supported by Admiral Tovey. Fifteen minutes later Admiral Hamilton turned westwards with his four cruisers and the destroyers, passing close to the astonished convoy.
As the hours passed without any drastic developments occurring, Admiral Hamilton and Commander Broome both became increasingly puzzled; but the former had received peremptory orders, and felt bound to continue to carry them out. Moreover, he had no knowledge of the hint given to Admiral Tovey by the First Sea Lord before the convoy sailed, to the effect that the Admiralty might order the convoy to scatter. Broome, on the other hand, 'felt certain that [his destroyers] would be ordered to turn back' to help defend the scattered merchantmen, once the anticipated threat had subsided. The situation was further complicated during the night by thick fog, which persisted until about 6:30 a.m. on the 5th. That afternoon Broome signalled to Admiral Hamilton 'I am always ready to go back', which message he 'intended as a hint as to where I knew my duty lay'; but the cruiser Admiral hoped that he was leading the enemy towards the main British fleet and considered that, since a large-scale surface action might be imminent, 'the most useful service the destroyers could perform would be with the battle fleet'. Not till 6:30 p.m. on the 5th, twenty-one hours after the withdrawal, did a message from Hamilton open the Commander-in-Chief's eyes to the fact that the destroyers of the escort were with the cruisers. He later supported the decision to concentrate the destroyers initially with the cruisers; but he condemned the failure to send Broome's flotilla
back as soon as it became apparent that the Tirpitz was not, in fact, in the offing.
Meanwhile Admiral Tovey had been cruising in the waters northwest of Bear Island. Early on the 5th, he turned towards Scapa. Later that day he received the first firm intelligence of the enemy's movements, from Russian and British submarines and from one of our reconnaissance aircraft. These reports all placed the enemy off North Cape, steering an easterly course, but still some 300 miles from where the convoy had scattered. Actually the enemy's sortie was very brief, for Hitler only gave his permission for the Tirpitz to sail during the forenoon of the 5th. She, the Hipper, Scheer, seven destroyers and two torpedo-boats left Altenfiord between 11 and 11:30 a.m. and steered to the east.38 When the Allied sighting reports already mentioned were intercepted, and it had become clear that the scattered convoy was suffering heavily at the hands of the U-boats and aircraft, Admiral Raeder cancelled the operation. At 9:30 that evening Admiral Schniewind reversed his course. Though preparations were made to attack him with carrier aircraft, and our submarines were redisposed to try to catch the enemy, the Tirpitz and her consorts reached Narvik safely. On the 8th the main British forces had also reached harbour.
We must now return to the convoy, whose long-drawn agony began very soon after the cruisers and destroyers had withdrawn. It scattered 'in perfect order', and ships proceeded singly or in small groups, escorted by the A.A. ships Palomares and Pozarica and screened by the smaller units - so that, in effect the order to scatter was partially, but insufficiently, undone. We cannot here follow the fate of each small group. Their adventures and, all too often, their tragic endings have been told in various eye-witness accounts.39 Less than half the merchantmen got even as far as Novaya Zemlya40 During the next three days seventeen of them, the oiler Aldersdale and the rescue ship Zaafaran were sunk by U-boats and aircraft. The Commodore's ship, River Afton, was among those lost, but happily the gallant and imperturbable Dowding and the ship's Master were both saved, after more than three hours spent on rafts in those icy waters. By the 7th five merchant ships and most of the escorts had reached the Matochkin Strait. They formed themselves into a small convoy and started off on the evening of the 7th to make a hazardous and difficult passage south towards the White Sea. For four hours
during the night of the 9th-10th they were heavily bombed, and two more merchantmen went down. Three ships reached Archangel on the 11th. 'Not a successful convoy' was the concluding sentence of Commodore Dowding's report - surely one of the classics of understatement. Actually a rescue ship and two merchantmen had already arrived on the 9th, so Dowding's little group was not the first to make port. On the 16th he left Archangel once more, in one of the three corvettes sent to bring in other ships known to be sheltering off Novaya Zemlya. The inhospitable, ice-bound coast was searched; one ship was found aground, another at anchor, and survivors from a third were collected. Then, entering again the Matochkin Strait, he found five more of PQ 17's number at anchor. Three of these - the Silver Sword and Ironclad (American) and Troubadour (Panamanian) - had been collected by the trawler Ayrshire (Lieutenant L. J. A. Gradwell, R.N.V.R.) when the convoy scattered, and taken twenty miles into the ice. There they remained for nearly two days, during which they camouflaged themselves by painting their upper works white. They then continued the southward journey and reached the Matochkin Strait safely. The little Ayrshire's conduct was, in Admiral Tovey's words 'a splendid example of imagination and initiative'. Her Captain had been a barrister and his First Lieutenant a solicitor before the war; yet they acted as though they had spent a life time acquiring naval outlook and traditions.
As soon as Commodore Dowding arrived off Novaya Zemlya he organised another convoy, and all the ships sailed on the evening of the 20th, with the Commodore leading in a Russian ice-breaker. One more merchantman was collected next day, and they all arrived safely on the 24th. Four days later the American ship Winston Salem, which had been aground in Novaya Zemlya, was refloated, and she too made harbour. But that was all. Of the thirty-six merchantmen and three rescue ships which had set out from Iceland, two of the former had returned to the starting point early in the passage; thirteen of the convoy and a rescue ship were sunk by air attack, and ten by U-boats. Only thirteen ships (eleven of the convoy and two rescue ships) survived the ordeal. The figures below give details of the cargo which reached Russia, and the quantity lost:
Delivered Lost Vehicles 896 3,350 Tanks 164 430 Aircraft 87 210 Other Cargo 57,176 tons 99,316 tons
The enemy accomplished his success at trifling cost to himself. Of the 202 attacking aircraft employed41, only five were lost. The poignancy of the tragedy is only accentuated by our present knowledge of how easily it could have been avoided. Yet the courage, endurance and resource displayed by the merchantmen and by the escorts which went on after the convoy had scattered have never been excelled; and it is they who provide the one redeeming feature in so dark a story. When Admiral Tovey wrote his despatch on the operation he gave it as his opinion that 'the order to scatter the convoy had been premature; its results were disastrous'.
These events had far-reaching repercussions, not least because so many Allied (and especially American) ships were involved. Lurid reports circulated on the other side of the Atlantic, and suggestions were put abroad that the Royal Navy had abandoned its charges at a moment of crisis. It is easy to see how the action of that unhappy afternoon of the 4th of July 1942 could give such an impression to members of the crews of the ill-fated merchantmen. Happily the pertinacity, and resource of the A.A. ships and of the little escorts (all British or Free French), who saw things through to the bitter end, give the lie to any such statements. Nor can Admiral Hamilton's withdrawal of the destroyers, now that the full circumstances are known, be regarded as more than an 'error of judgment', as the First Sea Lord described it.
The whole matter was, of course, fully investigated in the Admiralty, and on the 1st of August the First Sea Lord gave to the Cabinet an account of the events which led up to the crucial order being sent. The only new knowledge to be derived from that report is Admiral Pound's statement that on the night of the 3rd-4th July the Admiralty became possessed of intelligence indicating that the Tirpitz had eluded our patrolling submarines, and could be in a position to attack the convoy on the morning of the 5th. The existence of such precise intelligence has not been confirmed by postwar research. According to the record of the meeting, Admiral Pound told the Cabinet that the Admiralty had given the orders to disperse, then to scatter the convoy. Mr. Churchill's statement that he 'never discussed the matter with him [Admiral Pound]', and that 'so strictly was the secret of these orders being sent on the First Sea Lord's authority guarded by the Admiralty that it was not until after the war that I learned the facts'42, seems therefore to show a lapse in the Prime Minister's memory.43
As we look back on this unhappy episode today, it is plain that the enemy was never likely to risk the Tirpitz in close attack on a convoy protected by an escort which was heavily armed with torpedoes. That, as was pointed out earlier, had always been Admiral Tovey's opinion; but the Admiralty had never accepted it. The latter could not, of course, know of the restrictions imposed by Hitler and Raeder on the employment of the battleship. Yet all experience of German warship raiders so far gained had shown how reluctant they were to engage a convoy closely, except when it had scattered or was completely unescorted. The Scheer's attack on HX 8444, the Hipper's on WS 5A and SLS 6445, the wariness of the Scharnhorst and Gneisenau when they approached, but did not attack, the escorted convoys HX 106 and SL 6746, all indicated the same unwillingness to accept action except where an escort was very weak, or totally lacking. Furthermore the earlier experiences of German heavy ships had been such as might well make them chary of approaching our destroyers' smoke screens.47 It may therefore be felt that evidence derived from recent experience was available in London to suggest that, if faced with a similar problem, the Tirpitz's actions would probably follow on the same lines as the other raiders. If that be accepted, then the real nature of her threat could have been reassessed, and it might well have been realised that to scatter was to court far greater perils than to stand on and show fight.
In conclusion the tendency of the Admiralty during Admiral Pound's time as First Sea Lord to intervene excessively in the conduct of fleet operations has been commented on in several other contexts; and it will be remembered that, in spite of the First Sea Lord having expressed quite different intentions early in the war, the practice continued.48 It was suggested that the First Lord himself bore a share of the responsibility for interventions made in the Norwegian campaign49; but the habit persisted, though in varying degrees, long after Mr. Churchill had left the Admiralty. There can be no doubt that Admiral Pound himself became markedly prone to make such interventions, often on quite trivial matters, such as telling individual ships to steer a particular course or to steam at a particular speed. Nor did attempts to discourage such practices, made by senior members of the Naval Staff who fully realised the dangers,
meet with any success. When the Russian convoys became such difficult and dangerous operations, signalled interventions from London became very common indeed; and it has been mentioned that Admiral Tovey protested strongly on that score. That, sooner or later a serious misunderstanding would arise seemed all too likely to the Commander-in-Chief and the Flag Officers concerned; and the inevitable nemesis came with the attempt to exercise direct operational control over widely-spread forces, some of which were 1,500 miles or more from London, and working in conditions of which those ashore could not possibly be constantly aware.
The homeward convoy QP 13, of thirty-five ships, had an uneventful passage until one section of it ran into trouble off the northwest corner of Iceland. A large iceberg which suddenly loomed up through thick fog was mistaken for the North Cape of Iceland, and the error caused the Senior Officer of the escort to lead the merchantmen into our own minefield. The Senior Officer's ship - the minesweeper Niger - and four merchantmen were sunk, and two more of the convoy were seriously damaged. Coming so soon after the disaster to PQ 17 this was a doubly cruel misfortune.
Table of Contents
Previous Chapter (IV) ** Next Chapter (VI)