Grasping Gallipoli

by Peter Chasseaud

  The French helped the Turkish General Staff to set up their Military Mapping Department in 1895. Prior to this, Turkish officers had travelled to Paris for training at the École de Guerre and the Service Géographique de l’Armée, returning to Constantinople in 1892. The role of this new department was to initiate regular surveys and create maps covering the extensive Ottoman Empire.6 The first regular survey was constructed on a base measured at Eski-Shehr in Anatolia, leading to a 1:50,000 map of the town and environs being completed by 1894.7 Increasing German influence led to the French officers, General Defforges and Captain Barisien, who had been assisting the Turks to inaugurate their first regular survey, leaving the country around 1895,8 although French influence in the Turkish survey appears to have continued for some time. The French therefore had access to a certain amount of Turkish survey data and cartographic products, some (subsequent to their 1854 survey) possibly relating to the Gallipoli Peninsula. Presumably, on the basis of their new knowledge and with German assistance, the Turks continued to make local defence surveys on a limited basis, and it is probable that part of the Dardanelles area was surveyed at this time. One German source gives 1902 as the year of the start of 1:25,000 mapping in Turkey, and describes it as ‘Turkish survey (with French assistance)’.9

  The training of a large Turkish cartographic staff started surprisingly late, in 1904, with the institution of a two-year course in map drawing, followed by training in topography, the latter to culminate in practical fieldwork exercises. Clearly no regular surveys could begin, and maps be drawn, by the Turks themselves as opposed to foreign officers and men, until the Turks had built up their own cadre of technical operatives. In 1906, after a period of little or no progress, and even regression, a fresh start was made under Colonel Mehemmed Shevki Pasha, one of those who had trained in Paris, using a Bonne projection whose central meridian ran through the crescent on the dome of the mosque of Ayia Sophia (the old Byzantine basilica of St Sofia) in Constantinople. This meridian was adopted for all Turkish mapping in 1909. In 1907 the drawing of the 1:25,000 map of the Constantinople area was begun. The intention was that European Turkey should be surveyed at 1:25,000, and Anatolia at 1:50,000; though both agreed on the central meridian, they adopted different latitudes, so the surveys could not initially be joined.10

  A 1:25,000 contoured plan was made of Adrianople and its defences, which the British (who had a ‘military consul’ there – Captain Townshend in 1905) managed to obtain and reproduce in 1909 for their Military Report on Eastern Turkey in Europe and the Ismid Peninsula.11 This date is rather earlier than those of 1910–11 or 1913–14 given by some authorities for the printing of the 1:25,000 sheets of the Adrianople area.12 In six weeks in 1912 the Balkan League practically wiped out Turkey-in-Europe. The Greeks captured Salonika and the Serbs were victorious at Kumanova, while the Bulgarians defeated the Ottoman armies in Thrace (at Kirk-Kilisse and Lule Burgas) and pushed them back to the Chatalja Lines. Adrianople was captured by the Serbs and Bulgarians in renewed fighting during the London Peace Conference, and the Treaty of London (30 May 1913) restricted Turkey-in-Europe to little more than Constantinople and Gallipoli. In the Second Balkan War, Bulgaria attacked her recent allies, but was defeated by Serbia and Greece, assisted by Rumania, and Turkey took advantage of this situation to regain Adrianople (Treaty of Bucharest, August 1913). Clearly, in this chaotic military situation, Turkish surveys in the European part of the empire could not continue.

  The ‘Young Turk’ revolution of 1908 may have given an impetus to Turkish surveys, but the direct threats of the Italian (1911) and Balkan (1912–13) Wars certainly seem to have accelerated their execution. German survey practice, rather than French, gradually became the model. In 1911 a new base, over three kilometres long, was measured ‘to the east of the military hospital’ at Chanak for a second 1:25,000 survey of the Straits. There was an earlier, independent, 1:25,000 survey, covering the west end of the Peninsula, apparently preceded by a 1:20,000 survey, so presumably at least one earlier base had been measured.

  This new (second 1:25,000) survey of the Dardanelles, apparently executed in 1912 and 1913, covered the whole of the Gallipoli Peninsula and the Asiatic Shore although the disturbed conditions of the Balkan Wars, and later the First World War, rendered impossible the careful execution of a systematic geodetic survey; what was produced was more in the nature of a high-quality reconnaissance survey (based on a triangulation). The fear of invasion during the Balkan Wars caused an acceleration of the Straits survey, and eight sheets were completed of the south-west end of the Gallipoli Peninsula, the Narrows and Chanak and the Asiatic coast of the Aegean and Dardanelles. These were soon followed by others. The regular 1:25,000 survey of the Gallipoli Peninsula also formed the base for smaller-scale maps, being reduced by the Turks to 1:50,000 and 1:100,000 by 1915.13

  According to Lt-Col. Wood, OC 8th Field Survey Company RE, who checked the sheets in 1918–19, the 1:25,000 survey was of the highest order. However, a British report dated 22 September 1915 noted that though ‘the main outline … has proved to be very good … the contours are not up to much and the coastline is badly out in places’.14 The map was only tested, however, within a small area. A British 1924 report stated that the sheets represented. ’a good reconnaissance survey, similar to the Ismid 1/50,000 but not of so high an order as the Constantinople and Chatalja 1/25,000…. The 1st Corps Survey Platoon in 1923, however, could find little in it to quarrel with.’15

  In 1914 a map reproduction and printing office was set up in Constantinople, equipped with material acquired from Vienna and Berlin, and in 1914–15 two Turkish topographical officers were sent to these cities to learn the techniques of terrestrial stereophotogrammetry16 (a survey technique which the British, notably Lt F V Thompson RE, had experimented with before the war, but which they did not use during the war) which the Germans and Austrians had pushed forward rapidly, being particularly suited to mountainous areas where plane-tabling was difficult. The German survey of the route of the Berlin–Baghdad Railway through the Taurus Mountains was made in this way. German photogrammetric technicians, experienced in plotting alpine terrain from co-planar photos, also travelled to Turkey; on 30 January 1916 two Feldphotogrammeters of Bavarian Vermessungsabteilung (Survey Company) 11, a unit with their 6th Army in France, were working in Constantinople.17

  Turkish maps in Allied hands

  A crucial – and unresolved – question is how long before the war large-scale maps of the Gallipoli Peninsula, the Dardanelles and the Asiatic Shore were surveyed, drawn and printed, and why copies were not acquired by the Allies? Given the rearming of some of the Straits’ forts and batteries with modern Krupp guns in the early 20th century, it is probable that at least a local coast defence survey was executed to tie together guns, range-finders, searchlights, etc., for fire control purposes. A further reason, and one closely related to the development of fortress plans directeurs, artillery maps, and trench maps before and during the First World War, was that the mobile howitzer batteries defending the Straits’ minefields were intended to fire ‘indirect’, i.e. to fire from defiladed positions, depending on survey for their line and range and on observation from a crest or flank for correction of fire. These batteries would have to shift position frequently to avoid being located and neutralised by air-directed naval gunfire (or by the fire of enemy artillery landed for the purpose), and depended on an accurate, large-scale map for determination of their position, the position of observation posts, and for target location.18

  It is possible that a preliminary 1:25,000 ‘Southern Gallipoli’ sheet (part of the first 1:25,000 survey of western Gallipoli) was specially surveyed and printed at this stage as such an artillery map. Such a map of the Helles–Krithia area, unnamed and undated, was captured by the British after the first landings in April 1915.19 Another of the Helles–Kilid Bahr area, on sheetlines parallel to the graticule, was dated 1915 and allegedly used by Liman von Sanders.20 Such surveys and maps of strategically
sensitive areas were, of course, subject to security restrictions, and would have been particularly difficult for Allied agents to acquire. There seems to be a prima facie case for arguing that the Allies should, before the war, have made greater efforts to obtain copies of these recent Turkish surveys, but the Turks had only just made, or were still making, their own new large-scale regular survey, and issued some of the new sheets only just before the Allied attack.

  The Greeks may well have obtained copies of the Turkish large-scale sheets. Greek villagers and farmers would certainly have been aware of the work of the Turkish surveyors on the Peninsula and the Asiatic Shore. Lt-Colonel Cunninghame, military attaché in Athens from 1 March 1915, was shown the Greek operations maps by Colonel Metaxas on 3 March (he gave no details of scale, etc.), but told the Dardanelles Commission that the earlier absence of a British military attaché, and poor diplomacy by the British minister in Athens, Sir Francis Elliott, placed difficulties in the way of obtaining copies of these. When asked whether it would have been possible for the British military authorities to have had access to these maps, he replied that a good deal of suspicion had been generated by the very strong line which the British representative in Athens had taken for Venizelos and against the Greek General Staff, and that there had not been an English military attaché there before he arrived. The French military attaché was not popular, while the Greeks were (naturally) extremely suspicious of the Russian military attaché. All this meant that ‘it would have been very difficult for one to have got the maps’. The crux of the matter was the lack of British military representation. Asked whether he thought it would have been desirable to have had a military attaché at Athens from the beginning, he replied: ‘Absolutely, and it probably would have been decisive and made the whole difference.’21 An irony of this situation is that there was a British military attaché (including Frederick Cunliffe Owen) at Athens at various stages before war broke out. It was clearly unforgivable, given the situation in the Balkans and vis-à-vis Turkey, not to have had an officer in place from August 1914 to March 1915.

  The British War Office did, however, manage to get hold of some Turkish 1:25,000 sheets of the Constantinople–Chatalja (Bosphorus) area in 1912–13 during the Balkan Wars22 (possibly through Ashmead-Bartlett), but none of the Gallipoli Peninsula, despite reconnoitring visits in 1910 by Major L Samson, and about 1912 by the British military consul at Adrianople,23 and subsequent visits by Cunliffe Owen in 1914. This omission could perhaps be counted an intelligence failure, particularly as several different surveys, made over several years, were involved.

  After the winter of 1912–13 further efforts were made to complete the 1:25,000 survey of the fortified zone of the Straits, with detailed 1:5,000 plans ‘of some areas of extreme military sensitivity near Anafarta and Sid-ül-bahr’ (these were based on enlargements of the 1:25,000 surveys and do not appear to have been surveyed until after the British evacuation in January 1916). In February 1914 the trigonometrical and topographical sections restarted work on the north coast of the Gulf of Saros, north of the Bulair area.24 As a result of the interrupted nature of these surveys, there was some overlapping and detail and contours from different surveys do not always agree. Lieutenant T C Nicholas, Hamilton’s ‘Maps’ Officer, suggested in May 1915 to Hedley at GSGS that maps derived from at least three different Turkish surveys had been captured in April and May,25 and this is confirmed by the examination of archive material.26

  As suggested above, an area or block of 1:20,000 or 1:25,000 sheets, or both, appears to have been surveyed and printed before the regular Turkish 1:25,000 series; it differed in significant details from the later Turkish survey, and was on different sheetlines. It dated apparently from 1912, but may have been earlier, and coincided with the three sheets of the 1909 Turkish 1:50,000 map, except that sheet III (the southernmost) of the latter included a further row of 1:25,000 sheets surveyed in 1913. If this is the case, it is most unfortunate, given the time available, that the Allies did not obtain a copy before the campaign. The evidence for the existence of this earlier 1:25,000 survey and map is a Turkish index map dated 1917,27 and captured maps (see Chapter 9). However, another index map of Turkish 1:25,000 sheets28 gives 1912 dates for all sheets of the whole Gallipoli Peninsula except for sheet 44, immediately to the east of the Helles sheet (43), which was dated 1913. None of the Turkish index maps seen by the authors show the special large Helles–Krithia area sheet, and there remains the possibility that this was derived from an even earlier large-scale survey, particularly as some captured sheets provided evidence of different surveys of the same areas, and a few sheets were at 1:20,000 rather than 1:25,000.

  Although Turkish sources only refer to the 1:25,000 survey of the Peninsula, the evidence of captured 1:20,000 and 1:25,000 maps makes it clear that there were several surveys, certainly of the south-western end of the Peninsula and the Straits, and that the 1:20,000 scale probably pre-dated the 1:25,000. This fits in with the transition from French to German survey influence in Constantinople. The use of the 1:25,000 scale was evidence of German influence, the French preferring the 1:20,000 scale for their plans directeurs. The use of the Bonne projection and grade graticule, however, was evidence of the strong initial French input.

  Earlier 1:20,000 surveys were certainly made. On 8 May 1915 Lieutenant Nicholas described one of the first captured maps, to this scale, as of:

  … the area around Maidos and Kilid Bahr, contoured every 10 metres and printed rather roughly, and hastily, in black. [Like the captured 1:25,000 map of the ‘Gallipoli Peninsula from Cape Helles to a little south of Maidos’] it … looked very nice at first sight, but on comparing it with the other I found a difference of 30-40 metres in several of the heights; the topography differed considerably, and a test measurement of a length of about 4 km. showed a difference of 20%. One map is lying, and perhaps both.29

  One British map of the Anzac area seemed to referred to another Turkish 1:20,000 survey; this was: ‘Trench Map 1/1800 from Chatham’s Post to Lone Pine (Provn.), Compiled from Photos and Traverses. Contours compared with enlargement of 1/20,000 Turkish Plan. Note any additions or corrections to be sent to Major Newcombe, 4th Field Coy., 2nd Austr. Div.’30 However, this may refer to one of the British 1:20,000 maps based on captured 1:25,000 Turkish mapping.

  In summary, the Turkish maps existing in the decade before the war were the following (there is considerable uncertainty about the dates of much of this Turkish mapping):31

  Table 2: Summary of Turkish (Ottoman) maps of Gallipoli and the Dardanelles

  Scale

  Remarks

  1:5,000

  Helles, Anzac, Suvla, etc. Based on enlargements of 1:25,000 sheets, but considerably more accurate. Limited coverage. Not printed until 1916.

  1:20,000

  One or more military surveys of part(s) of the Gallipoli Peninsula (1895–1910?).

  1:25,000

  Regular sheets of the Dardanelles area, and a large special sheet of the south-west end of the Gallipoli Peninsula, surveyed in 1912–14. Most sheets of Peninsula not apparently printed until 1915, though there is a suggestion that the printing date range is 1913–14. There was also the 1:25,000 plan of Adrianople and its defences, which the British reproduced in 1909.

  1:50,000

  1909 edition of Gallipoli Peninsula (in three sheets) existed, if date can be trusted. This possibly incorporated 1:20,000 or 1:25,000 survey. There was also a 1915 Turkish Staff Map, reduced from the new 1:25,000 survey; the three sheets were named: Anafarta Sagir, Kilid-ul-Bahr and Sed-ul-Bahr. Pre-war (1909) contoured 1:50,000 sheets of the Constantinople area were produced, but these were enlargements of von der Goltz’s 1:100,000 map; the British pre-war one-inch sheets of the Constantinople and Bosphorus area were apparently derived from these.

  1:100,000

  Gallipoli sheet, of uncertain date, but as script is in French and Turkish this appears to be pre-war. 1:100,000 contoured sheet reduced from new 1:25,000 survey dated 191
5.32

  1:100,000

  Bosphorus sheet 1908. Von der Goltz produced a 1:100,000 map of the Bosphorus and Environs in 1895.

  1:120,000

  Les Dardanelles. Apparently pre-war.

  1:126,000

  Pre-war Russian-Bulgarian series covering Balkans and European Turkey. ‘Very old’.33 The Russians described this as a ‘contoured map of the Balkan Peninsula’.

  1:200,000

  Pre-war Austrian General Staff series. Turkish series produced from 1910.

  1:210,000

  Series produced in 1898 covering Imperial Turkey in Europe (southern Balkans and Greece).

  Allied maps: the French Carte de la Presqu’Ile de Gallipoli, 1854

  This map, in two 90 x 65 cm sheets,34 was contoured (form-lined) at ten-metre vertical interval, and was a fine-looking map which gave a very good picture of the terrain. Particularly noteworthy was the way in which the confused, even chaotic, topography in the Sari Bair range (Anzac–Suvla Bay) was indicated. It also showed the spur between the coast and Gully Ravine covered with brush or scrub (Broussailles); this information was not shown on the British one-inch or 1:40,000 map, but was again picked up from reconnaissance before the landings.

  The survey methods used by the officers who made this map were not stated on the original sheets, but presumably it was a typical military reconnaissance survey of the period, with a few points fixed by theodolite, sextant, plane-table or compass intersections, the details and form-lines sketched in (perhaps using a plane-table), and spot-heights obtained by taking vertical angles with a clinometer. Points marked on the map indicate that churches, mosques or shrines, windmills and other significant points were fixed, and used to control the map. To provide trig points in wild hilly country, temporary cairns or beacons were probably constructed. Either a cairn was erected on Achi Baba (Atchi-baba on the map), or an existing beacon was used; the map shows a sign for a trig point. Such constructions were a normal part of field triangulation. Azimuth would have been found from astronomical observations.