Grasping Gallipoli
Page 37
A lack of strategic clarity had obvious implications for the survey organisation responsible for making or ordering the required maps, and for the distribution of such maps. There was also a disastrous imbalance between the secrecy requirement and operational efficiency. Both of these factors – strategic muddle and obsessive secrecy – restricted or curtailed opportunities for gathering geographical information and for studying the terrain. As usual, the consequence of all this was failure, with heavy casualties among the attacking infantry.
In his Despatch of 11 December, Hamilton retrospectively defined his aims and objectives as:
1. To break out with a rush from Anzac and cut off the bulk of the Turkish Army from land communication with Constantinople [i.e. the primary attack, to capture the key height of Koja Chemen Tepe (the highest point of the Sari Bair ridge) as the key to the capture of the line across the Peninsula from Maidos to Gaba Tepe].
2. To gain such a command for my artillery as to cut off the bulk of the Turkish Army from sea traffic whether with Constantinople or with Asia [i.e. capture the Kilid Bahr Plateau].
3. Incidentally, to secure Suvla Bay as a winter base for Anzac and all the troops operating in the northern theatre [i.e. a secondary attack; diversionary operations were also carried out at Helles].
Point No. 2 was connected with the desire, expressed by Churchill, to emplace heavy and super-heavy artillery (9.2-inch guns and 15-inch howitzer) in positions where it could destroy the Turkish defences and forts in the area of the Kilid Bahr Plateau. Churchill had minuted as early as 14 May that:
The operations have now reached a point where they may easily develop into a great siege similar to that of Port Arthur, though not so formidable. Our preparations should therefore consider and cover the following points: . . The provision of heavy artillery, which could be used against the semi-permanent works, and the mounting on shore of heavy long-range naval guns which can, from the existing positions held by our troops, bring accurate fire to bear on the permanent defences.
On the same day he minuted to Fisher and his Chief of Staff: ‘The fifth 15-inch howitzer with fifty rounds of ammunition, should go to the Dardanelles with the least possible delay… The two 9.2-inch guns will go to the Dardanelles, either in the two monitors prepared for them or separately, for mounting on shore…’7
In the event, the 15-inch howitzer was sent to Mudros but, even if landed, did not reach the Peninsula and never fired a shot. The naval guns were never mounted on land but, together with a 14-inch gun, mounted in monitors performed very useful service against Turkish batteries on the Asiatic side. Unfortunately the enthusiasm of Churchill and Keyes (de Robeck’s Chief of Staff) was not matched by that of others in key positions, notably Stopford and Reed (Stopford’s senior staff officer). And by the time of the Suvla operations, Churchill was no longer at the Admiralty. These memos to Fisher, allocating more naval resources to the Dardanelles, had finally triggered the resignation of the First Sea Lord, bringing about Churchill’s downfall a few days later, when he was replaced at the Admiralty by Balfour. As Chancellor of the Duchy of Lancaster, however, he remained influential, retaining a seat on the War Council (now renamed the Dardanelles Committee).8
Various questions arise from Hamilton’s statement. The strategic aim was much the same as it was in April. But was the Navy to cooperate, and what was the role of the other Allies? And what of terrain difficulties? The capture of the Sari Bair range committed Hamilton’s force to operations in almost impossible country, as he well knew from the Anzac experience in the Sari Bair foothills. Did the corps commander understand his instructions? General Stopford was persuaded by his orders, and by conversations with Hamilton and Braithwaite, that his main task was to secure a harbour and base area, so his operations did nothing to support the attack from Anzac and did not form part of a synergetic whole. As a result he failed to drive forward from Suvla to crown the surrounding heights (how he thought he could secure the area without doing this remains obscure – he had made it clear to Hamilton that he fully recognised the need to capture the hills) and push southwards past Anzac to the Narrows before Turkish reinforcements could arrive. Stopford’s over-cautious approach was also coloured by his view – flying in the face of all intelligence reports – that a heavily entrenched Turkish position, supported by heavy artillery, had to be tackled in a Western Front manner.9
A major contributor to early failure was the excessive secrecy, in defiance of field service regulations which stressed that the need for secrecy had to be balanced against possible loss of operational efficiency. The Suvla operation, at first a mere afterthought to a big attack from Anzac and mostly planned, like the April landings, by Aspinall, Dawnay and Deedes, was cloaked in such secrecy that the formation commanders themselves were kept in the dark for far too long. Several senior officers remained unbriefed until 30 July, and many didn’t see a map before the landing. The same concern for secrecy had restricted opportunities for land, sea and air reconnaissance (low-level reconnaissance flights were forbidden10) to a minimum far below the absolute minimum required for conduct of operations. Thus most commanders, officers and men were totally lacking in an understanding of the ground over which they had to operate.
Knowledge of the Suvla terrain and Turkish defences
The terrain of the Suvla Bay area was fairly well known and appreciated before the landings, but it could be argued that, as at Anzac, no amount of terrain intelligence would compensate for lack of experience in fighting in these conditions. This is particularly true of the ground encountered once troops had left the flat Suvla Plain.
In his letters to Kitchener, his Despatch, and subsequently, Hamilton makes much of the difficult terrain, particularly the dense scrub, with the implication that he had not been fully aware of this before the operation. In fact he was in possession of a great deal of terrain intelligence from various sources. The 1909 War Office Report11 stated (emphasis added):
South of Cape Suvla is the wide bay of the same name, well sheltered from the northern winds which frequently prevail, and having a sandy beach suitable for landing. A good beach runs south of Nibrunesi Point also, and from both these places access is fairly easy to the villages of Biyuk and Kuchuk Anafarta, and thence into the plain about Turshtenkeni and Selvili. A rough coast track leads south also.
Suvla Bay. The northern shore of the bay is lined by low, rocky cliffs, some 30 feet high, and rocky hills covered with low scrub extend along the coast to the east. Between the cliffs and the mouth of the Salt Lake is a stretch of firm, sandy beach, 900 yards in length, on which a landing could easily be effected in calm weather or when the wind is north. In a southerly wind landing would not be practicable, but northerly winds are prevalent, especially in winter, and from these the bay is quite sheltered. Deep water extends close up to the beach, but there are occasional shoals. A landing on the north side of the bay gives access to a stretch of sandy hillocks, covered with tufts of grass, which offer no obstacle to the passage of guns or wheeled vehicles; 800 yards inland cultivation (principally maize) is reached, with firm ground passable for all arms. The hill [Lala Baba] between Salt Lake and Nibrunesi Point is of sand and stones, covered with low scrub, and has an old tomb on the summit. The plain between the Topalin Mezar Dere and the Kizlar Dere is covered with cultivation and dotted with large trees, oak and walnut. Low scrub-covered hills of sandy soil descend to the plain from the south and east. The Salt Lake or Lagoon is ordinarily dry in summer, but with a south wind it may fill up again. From November to April it always contains water. When dry its bed is easily passable for infantry. The opening to the sea is 80 yards wide, and also becomes dry in summer.
Further extracts from the 1909 report are given in Appendix III, including information about the Sari Bair range immediately to the south, where the Anzacs had encountered impossible terrain and a tough Turkish defence.
Of the area north of Suvla Bay, the British 1:50,000 map (GSGS 4001) produced in the late-summer (possibly A
ugust) of 1915 by the Survey of Egypt and based on captured Turkish 1:25,000 originals, carried this description (possibly written by Nicholas or Ogilvy, clearly with a geologist’s eye):
The sharp ridge running from the neighbourhood of Ejelmer Bay to Suvla Point is formed of beds of hard sandstone alternating with less resistant rocks and dipping to the S.E. The seaward slopes are steep, rocky and in many places precipitous, while the S.E. slopes are more gradual but corrugated by numerous minor parallel ridges, where the harder beds crop out. The whole ridge is covered with low scrub…. Hills around Anafarta Sagir are covered with thick scrub and traversed by ridges of hard rock with a general E–W trend.
Although the map was printed after the Suvla landings, the information was available before August; it was also printed on the British 1:20,000 sheets of the GSGS 4000 Dardanelles series, some of which appeared in August.
Reconnaissance by landing parties and patrols from Anzac
As early as 30 April, the Navy landed a fighting patrol from a destroyer on Suvla Bay’s southern beach to deal with Turkish artillery observers directing fire onto the Anzac beach, and this patrol attacked a Turkish piquet entrenched on the summit of Lala Baba and cut a telephone wire. This feat was repeated inside the bay two days later by Captain C Cribb and fifty men of the Canterbury (New Zealand) Battalion, landed by the destroyer Colne, who surprised sleeping Turks in the Lala Baba trench. The New Zealanders searched the area, discovering two trenches covering the beach to the south of Nibrunesi Point, and also a spring on the western side of the hill, before successfully re-embarking.12
Two brave and adventurous ground reconnaissances made from the northern flank of the Anzac beachhead during May – the first by Major Overton and Corporal Denton on 15–16 May, and the second by Major Overton and Captain Hastings on 27 May – proved that the terrain was practicable for a left-flanking movement towards Chunuk Bair, and that the Turks considered the terrain to be so difficult that the area was undefended. Only old and unoccupied stretches of trench were seen on Chunuk Bair.13 More incredibly, a New Zealand reconnaissance patrol under Lieutenant G R Blackett was landed from a trawler on the night of 20 June at Nibrunesi Point, the tip of the southern arm of Suvla Bay under the northern slopes of Kiretch Tepe, to locate Turkish artillery positions on the western hills. After successfully crossing the Suvla Plain, they returned to their trawler, after locating an abundant water supply on Kiretch Tepe, without encountering any Turkish soldiers or defences. Turkish patrol activity was reported, but there were few enemy forces north of the Anafarta Spur.14
Invaluable terrain intelligence, which was to form the foundation of the Anzac and Suvla operations in August, was gathered through these reconnaissances. More information could have been gathered in this way, but as it was feared that frequent patrolling might alarm the Turks, reconnaissance patrols were discouraged in the weeks before the landings. Thus new confirmation regarding the paucity of the Turkish defences, which might have encouraged Stopford and Reed to adopt a more forward operational approach, was not acquired. It has also been argued that the many wells and other water supplies should have been located and verified by patrols before the landings,15 but not only is it unlikely that the Turks would have countenanced this degree of activity, but it is also probable that they would have taken more effective steps to render these supplies unusable.
Hydrographic intelligence and landing places
While there is disagreement regarding the accuracy of existing Admiralty charts and other hydrographic information for the Suvla area, there is no doubt that the Navy could have collected more data before the landings had it been warned to do so.16 Again, excessive concerns about secrecy may have played a part in this, but given that a force was landed at Anzac, just to the south of Suvla Bay, in April, and maintained in that position thereafter, there were ample opportunities for further hydrographic reconnaissances off a land area where Turkish forces were extremely thin on the ground.
According to Aspinall, the hydrography of Suvla Bay itself was largely unknown, and the old Admiralty chart was unreliable. We have seen that although the Suvla Bay area had been charted in the 19th century, little specific hydrographic work in that area was done before or in April 1915, or for the Suvla operation in August, as it was feared that any obvious survey work would alert the Turks and lead to stronger opposition. Hamilton was extremely angry when a destroyer fired a shell into the Salt Lake to test the hardness of its surface before the August landings, believing that this would give the game away to the Turks.
The Navy originally rejected the idea of landings at Suvla Bay because it believed that shoal waters and reefs made landings too hazardous, and that the charts were poor and insufficient accurate hydrographic data were held. On the other hand there were now more general hydrographic data compared with April, as Turkish maps had been captured, and there had been months of close observation by the Navy’s fire-support ships as well as aerial reconnaissance (seaplanes and aeroplanes), air photography and landing parties, which all contributed their bit to the overall picture.
Naval reconnaissance suggested that the northern shore was rocky and not suitable as a landing place, while in the southern area around Lala Baba it was decided that the low cliffs bordering the sandy beach ruled it out also as a landing place. A further adverse factor was that the sea near the shore at Lala Baba was very shallow. A suitable landing place was located just north of ‘The Cut’ (the apparently dry watercourse connecting Suvla Bay with the Salt Lake) on the eastern side of the Bay. Here a long, wide beach looked very promising, while there was an apparently easy exit from the beach to the Suvla Plain across low sand-dunes.17 The Salt Lake itself was known to be dry in summer, and between The Cut and Lala Baba was separated from Suvla Bay by a low and narrow ridge of sand-dunes.
However, the Navy’s considered view was that, owing to shallow water and the possibility of rocks, no landings should be attempted within Suvla Bay. Instead, they advocated landings on the excellent beach, over a mile long with easy exits to the Suvla Plain, south of Nibrunesi Point, the southern arm of the bay. This firm sandy beach, with its deep-water approaches very close to the shore, was well-known to the Navy, through close observation by the destroyers guarding the Anzac northern flank.18 Visual reconnaissance and sketches made from the sea gave a useful general impression of the terrain, but ‘the distance prevented any useful information being gained’.18 While the beach seemed to have an easy, almost flat gradient, it was difficult to estimate distances and heights, and the sketches did not disclose any Turkish defences.
In the event, despite its misgivings and although certain beaches were shelving or difficult in other ways and some of the approaches were bad, the Navy found that it was able to land Stopford’s Corps in the face of weak opposition. New purpose-built, shallow-draught landing craft, originally designed for Fisher’s pet Baltic operation, were used. Because of their pairs of narrow, extending ramps, giving the appearance of antennae, they were known as ‘Beetles’. Most of Stopford’s force was landed south of Nibrunesi Point, the Navy’s preferred beach, but Hamilton’s plan was changed at Stopford’s insistence to land Sitwell’s 34th Brigade inside the bay north of The Cut, at the beach earlier identified. This Brigade, however, ran into trouble. The covering force destroyers anchored in the wrong positions, and as the Navy had feared, an uncharted reef was encountered, two of the special landing craft ‘Beetles’ struck the reef, and their troops had to wade shore in neck-deep water. The net result of all this was that the landing was made south of The Cut. Despite the usual chaos associated with troops (and these were relatively untrained and inexperienced) landing in the dark, good footings ashore were made against light opposition, but bad leadership and a failure to push on meant that the initial successful landing was not exploited. The ‘Beetles’, as well as tank landing craft (the tanks specially modified to climb the sea wall), were later prepared to make coastal landings in Belgium in mid-1917, but as Haig’s Flanders offensive failed
to make sufficient progress this operation was cancelled.20
Air photography and reconnaissance, and the Turkish defences
Aerial reconnaissance and photography provided the most detailed and accurate information about the Turkish defences and troop movements, and revealed the small size of the garrison defending the Suvla area.21 They did not, however, give much detailed information about the terrain, and their results were not interpreted with this in mind. Apart from the beaches and their exits, observers and air photo interpreters were simply not told to consider the terrain.22
Details of air reconnaissance and photography at Suvla were given in Chapter 7, but it is useful to summarise them here. On 23 July Samson flew low over the area and saw little activity and few trenches.23 On 4 August all existing trenches and gun emplacements were identified (most were unoccupied), and the position sketched from the air. The next day the information about the gun emplacements was confirmed and it was reported that the trenches north of the Salt Lake were unoccupied. Photos were taken of the Chocolate Hill defences, one of the first objectives. The air photos generally showed little or nothing in the way of enemy defences except at Lala Baba, and Knatchbull Hugessen recorded: ‘Note complete absence of all trenches, etc.’24 On 6 August the only enemy troops seen were moving away from the area.25
The results of earlier reconnaissance by landing parties had not necessarily been well-collated or appreciated; the Final Instructions issued on 29 July from GHQ to Stopford’s IX Corps made a specific reference to ‘an aeroplane photograph [which] has also disclosed the presence of a few trenches on Lala Baba. A sketch of these trenches [not reproduced in the Official History], which have apparently been constructed for some months is attached.’26 The air photos confirmed the Intelligence staff in their belief that there were only some 4,000 men and a few guns to oppose a landing by a whole British corps of over 20,000 men, and also revealed which localities and hills were undefended, and the relative strength and dispositions of those which were defended.27