This is a fascinating expedition report on the British Somaliland-Ethiopia boundary, accompanied by a color map and some photographic plates. The area was surveyed thoroughly; numerous aspects are recorded, including geographical features and tribes. The boundary commission was a very successful one.

The British Somaliland-Ethiopia Boundary:

A paper read at the Evening Meeting of the Society on 20 January 1936, by LIEUT.-COLONEL E. H. M. CLIFFORD, C.B.E., M.C., R.E.

The British Somaliland-Ethiopia BoundaryAuthor(s): E. H. M. Clifford

Source: The Geographical Journal, Vol. 87, No. 4 (Apr. 1936), pp. 289-302

An Arab sultanate with its capital at Zeila was founded by emigrants from Yemen in, it is said, the seventh century A.D., and in the thirteenth century became powerful as the Empire of the Adals. It is interesting to note this name. Zeila is called by the Greek geographers ‘Αδυλη, and Somalis today know it as Awdal. In the sixteenth century, the Arab influence was decreasing and the capital was transferred inland to Harar. This was the epic period of the Empire; Mohamed III, Ahamed ibn Ibrahim el Ghazi, the notorious Mohamed Gran of Ethiopian history, became emir in 1525 and before his death in 1543 had invaded, ravaged, and become a virtual master of the greater part of Ethiopia. Thereafter the Empire was greatly harassed by Galla invaders in the seventeenth century and broke up into a number of petty independent emirates and sultanates, mostly Somali. Zeila became a dependency of Yemen, and thus nominally part of the Turkish Empire. During the first half of the nineteenth-century British influence – exercised by the East India Company, later by the Government of India – grew steadily on the south of the Gulf of Aden; but British authority was not established. In 1874-75 Egypt occupied Tajura, Bulhar, Berbera, and Harar, obtained a firman from Turkey by which Zeila became Egyptian, and secured British recognition of her jurisdiction as far east as Cape Gardafui. But, in consequence of the situation in Sudan, all Egyptian garrisons were withdrawn in 1884-85. Great Britain then occupied Zeila, Bulhar, and Berbera, in order to secure the Red Sea route to the Far East and also the supply of fresh meat for Aden; Turkey being given the option of resuming possession of Zeila, an option which was not taken up. At the same time, France annexed the Danakil and Somali coasts at present held by her. An Arab emirate was established in Harar as a buffer state but fell early in 1887 to the arms of Menelik II, King of Shoa, and shortly to become Emperor of Ethiopia. During 1884-86 treaties of protection were concluded with the various Somali tribes, considerable rivalry resulting in the western part of the country between ourselves and the French. A number of entertaining anecdotes of this period have been related by Mr. Prendergast-Walsh in his book ‘Under the Flag and Somali Coast Stories.’ Early in 1888, an Agreement was signed with France, defining the boundary between French and British spheres of influence, extending inland as far as Harar, but imposing self-denying ordinances on both parties as regards Harar itself. The boundary between British and Italian spheres was next defined, by the Anglo-Italian Protocol of May 1894.

The British Somaliland-Ethiopia BoundaryThen came the Franco-Ethiopian Boundary Convention of 20 March 1897; and, on its way back to the coast, the French party was passed by a British Mission to Emperor Menelik, headed by Mr. Rennell Rodd (now Lord Rennell of Rodd). The labors of this Mission resulted in the Treaty of 14 May 1897, the boundary clauses of which are expressed in an appendix negotiated at Harar with Ras Makonen, father of the present Emperor. When the discussions were opened, Mr. Rodd was confronted with a circular letter which the Emperor had addressed to the heads of all European states on 10 April 1891, in which he made a declaration of the frontiers of Ethiopia. This had never reached London, though an inaccurate translation had appeared in La France Militaire in March 1896. The agreed line was a compromise between the Emperor’s claims and that of the Anglo-Italian Protocol of 1894. The great objection to it is the administrative one that it cuts across tribal areas, though this was met by the inclusion of a trans-frontier grazing clause. But in any case, the difficulty is not entirely avoidable with these nomad tribes whose areas overlap in the most confusing manner. There the matter remained until 1924, when the present Emperor, then Regent and known as Ras Tafari Makonen, during his visit to Europe expressed the desire to see all the frontiers of Ethiopia demarcated. Great Britain, alone of its neighbors, paid any serious attention, and it was agreed that a start should be made with the Somaliland frontier. Negotiations to this end proceeded slowly but on the whole surely, and at the end of 1930 reached the stage of definite preparations; but the Boundary Commission did not actually meet until 8 January 1932, at Berbera.


The British Somaliland-Ethiopia BoundaryThe general configuration of Somaliland has often enough been described. I will therefore confine myself to remarks on the parts we worked through. The Haud is a vast area of thorn bush desert sloping very uniformly south-wards, with little, or only very minor, physical relief. It contains no permanent water except at Doomo (Damot on current maps) and Harodiget; in fact, it is the position of the permanent water points round it that determines its limits. It is nevertheless a region of the utmost importance to the tribes on account of its grazing. From the point of mapping, the principal features are the water points: mostly depressions called bale or hara which hold rainwater; and the tracks. Most of the latter run north and south, and until very recently were caravan routes leading principally to Berbera or Bulhar. The development of Ethiopian control has however diverted trade via Dagahbur and Jijiga, and motor transport has put the seal on this change. So that now these tracks are virtually only used as the guiding lines for the seasonal migrations of the tribes. West of the line Hargeisa-Dagahbur it is very striking how all the tracks swing round westwards to Jijiga; there can be little doubt that originally they led overpasses on the watershed between the Fafan and Jerer basins, and that the growth of Jijiga has been responsible for the change. Though the Haud includes many open stretches, as a wide generalization it is fair to say that in the east and south, it is thick, even very thick, thorn bush, and in the north-west parkland. I feel that this is probably partly a result of the disturbed period of the activities of Mohamed Abdulla Hassan, the “Mad Mullah”; for during this time, which lasted some twenty years, tribal arrangements were completely dislocated; only the Mullah’s adherents were able to frequent the south-eastern Haud, and there was considerable congestion in areas far enough away to appear safe, such for instance as Odweina, where one now finds large extents of dead bush, unquestionably due to over-grazing.

The British Somaliland-Ethiopia Boundary
1. Looking N.W. from the crest of the main ridge of Marmar Gedle

North-west of the Haud is a completely bare plain, at an average elevation of about 1500m, with its northern point about 15 miles in British territory, and bounded on the west by the escarpment of the Harrawa valley and on the south by the hills behind Jijiga. It has no collective name that of Marar assigned to it on existing maps belonging in reality to a water-point near Jijiga. A large proportion of the southwestern half of this plain is under cultivation. North-west of this plain the hills start, at first as isolated peaks, or even masses, rising out of the undulating country, well covered with bush and cultivation. Four of these peaks reach to over 1800m. Then comes the broken country which forms the western end of the maritime mountains, the biggest element being the Libah-hele range. This terminates somewhat abruptly at its western end, where the Harirad gorge separates it from the striking mass of Gerigoan. The main part of this block of lithographic limestone has a remarkably smooth, tilted top, and for the greater part is surrounded by precipices. Except however for the western end of Libah-hele, the maritime mountains seem to consist in the main of two types: towards the west, rounded features with grass and sparse thorn bush and, on some of the higher parts, what is perhaps optimistically described as cedar forest – I should call it juniper scrub myself; towards the east, flat-topped hills of the type so common in Africa, showing erosion from a former even level, perhaps a sea-bed. Both types of course are much cut by ravines, some of which are not lacking in grandeur.

The British Somaliland-Ethiopia Boundary
2. Broken country W. of the Zeila plain: the Mirmir range with the boundary in left foreground

North-west of Gerigoan lie the tumbled masses of Marmar Gedle and Marmar Heir, the former a system of parallel ridges running north-west to south-east, the latter circular in plan, with a central peak. Both have in the past been rather zealously guarded as refuges by the Esa tribe. Marmar Gedle has now been mapped in full detail, but Marmar Heir is more complicated and we have a less thorough knowledge of it. The sheep and goat grazing in these mountains is particularly valuable. To the west of this lies the appalling desert through which the Chemin de Fer d’Ethiopie runs – completely bare and to a great extent covered with lava boulders, though parts of it are sandy. To the south, the broad Harrawa valley runs westwards through a region of dense thorn bush. North-east of Marmar Heir is a large plateau called Sa-wer, with a very level top and an escarpment all round.

Between Sa-wer and the sea is the Zeila Plain, a vast expanse of sand and grass, with several large watercourses running through it towards the coast. Though normally dry, after heavy rain in the mountains these are liable to cause a wide sea of mud that effectively bars the coastal motor-road from Zeila to Berbera. Parts of this plain have a light sprinkling of thorn trees of one sort or another, but in general, trees are restricted to the neighborhood of the watercourses. On one of these watercourses, well out in the middle of the Plain, the Somaliland Government have installed a wind-pump and
concrete troughs; a quaint sight in such surroundings, but needless to say it has more than justified itself. West of the Zeila plain is more broken country, which is in fact the edge of the mountain system of the French Somali Coast. Lava is an outstanding feature, particularly in the form of plateaux covered with boulders. The coastal half of this frontier zone is really one huge table, sloping eastwards from the Bura Mountains, and, though much cut up by watercourses, nearly devoid of water. It is consequently little visited by the Somalis, and guides for it are extremely difficult to obtain. The coastline between Zeila and Djibouti is low and shelving. Unfortunately, my geological knowledge is nil, so that I cannot enter into any discussion on that score except in the very widest terms. The mountains are heavily faulted, and there can be no doubt of considerable volcanic activity in the past, particularly in what is now French territory. Beside that however about 15 miles west of Bulhar, on the coast, the mountain of Elmis has a well-preserved crater; while a few miles south of Beyu Anod there are a number of striking pinnacles that must be the plugs of a volcano the rest of which has been worn right away.

The British Somaliland-Ethiopia Boundary
3. Milmilagd, S. of Beyu Anod: remains of a volcanic plug

The Somali has been described not infrequently. I will therefore endeavor to mention only some of the less well-known points about him. But to put these in their proper setting, a certain amount of the general picture must be painted. The race, in the first place, is divided into the Aji or pure blood, and the Sab, or outcast. The latter consists of three tribes – Midgan, who are hunters and menials and provide the sweepers for the European community; Tomal, who are the blacksmiths; and Yibir, a begging tribe who play an important part in Somali life, living largely by the survival of superstitious beliefs. In general, the Sab are scattered among the Aji, to whom they are subject. The Aji comprise Asha and some others mainly living in Italian Somaliland. There are two big branches of the Asha – Ishaak (which I should prefer to spell Isahaq) and Darod. The former essentially people the British Protectorate. The latter have two very big units in the Herti and the Ogaden and a number of smaller ones. The Herti occupy the coast from some distance east of Berbera fairly continuously all the way round to the Kenya frontier, though one element, the Dolbahanta, live inland, mainly in the eastern part of British territory. En passant I should like to quote the opinion of other Somali tribes about the Dolbahanta – “the children of a blind old man by a mad old woman.”

West of the Dolbahanta are the Ishaak, and west of them are the Gadabursi, a tribe said to be a hybrid of Ishaak and probably Ogaden. They seem to be fairly generally unpopular, perhaps because they are to a considerable extent agricultural, whereas the true Somali is essentially a nomad. They are also not a little feared on account of the belief that they have a practice corresponding to that of werewolf, but with hyena instead of wolf. The Ogaden are south of the Dolbahanta and Ishaak. Until quite recent times they used to send big caravans through to Berbera and Bulhar, but those days are over and only individual Ogaden visit British territory now. To the west of the Gadabursi are the Esa, a wild tribe with a tremendous reputation for fierceness. As their western neighbors are the still wilder Danakil, no doubt they need it, particularly as there is no love lost between the two. The origin of the Esa is not certain though they are recognized as Asha. Their country lies partly in British, partly in French, and partly in Ethiopian territory; a circumstance of which they naturally make the most. Indeed only once have the authorities astride the frontier cooperated in punitive measures, and I believe that the lesson will not be forgotten for a considerable time, for the operations lasted for between three and four months, and of all those “wanted” only one was not captured.

The British Somaliland-Ethiopia Boundary
4. Esa dance

As I have already said, the Somali is essentially a nomad. Tribal areas are well enough recognized, but they overlap and sometimes side-step in the most puzzling way. In bad years of course a tribe may wander off into its neighbors’ country, but it is generally arranged amicably. A Somali never marries a woman of his own section, and preferably not even of his own tribe. This may be due in part to a knowledge of the rudiments of eugenics, but there is no question that a very important reason is to obtain the right to visit the wife’s family – in other words to secure additional grazing ground for his stock. The stock consists essentially of camels, whose milk is the main item of Somali food. The Somali camel is an excellent beast, willing and gentle, and quite a number are nowadays trained as riding animals. Except in the latter case, they require no rations and subsist entirely by grazing. In the western Ishaak and Gadabursi country there are a good many cattle, but elsewhere the rainfall is insufficient. Sheep and goats are everywhere. The sheep are a variety of the Persian fat-tailed species, and I believe that about 1,000,000 “Somali blackhead” skins are exported annually, and are much sought after by glove-makers. On the coast between Zeila and Djibouti, but nowhere else that I know of, the goats graze on top of the acacia trees and bushes. The Ishaak, Dolbahanta, and Ogaden also have ponies; but African horse sickness periodically takes a heavy toll of them. It is said for instance that during the operations against the Mad Mullah, he lost 1,500 from that cause alone. Most water places consist of one or more wells or water-holes. But at a few there are a whole series of wells in the bed or on the banks of a watercourse, and the organization is then interesting. At Dagahbur individual wells have specific ownership but are used for all requirements, human and stock. At Bulale and I believe also at Dabawloq the wells are in groups – one for donkeys, one (the biggest group of course) for camels, one for sheep and goats, one for cattle, and another for humans. Inside each group there is individual ownership of particular wells. In this N.W. Ogaden-W. Ishaak area, the buckets used are wooden, shaped like and about the size of a 4.5-inch howitzer shell without a base, and these are thrown up from man to man from the bottom of the well to troughs at the top. It was fascinating to watch, for the movement is exceedingly graceful.

The British Somaliland-Ethiopia Boundary
5. Bulale: part of the group of wells for camels

In his true native state the Somali leads a very hard life and his character is hard in consequence. He is tough and wiry, but far from robust, and the death-rate from tuberculosis is high. Like, I suppose, most nomads contact with civilization does not to my mind improve him. I at all events cordially disliked the sophisticated specimens near the coast and very much preferred the simpler types of the Esa and Ogaden, greatly averse though the former are to contact with the Government. Under strict discipline and good handling, a lot can be done with them, whether as police, military, or merely as coolies for economic circumstances have compelled them to manual labor. But they are quite useless as porters. Two anecdotes will serve to illustrate their quickness of temper – and various other characteristics. Two coolies who had been with the Ethiopian Section for nearly two years had an argument over the rations, in the course of which one suddenly drove his knife into the other’s stomach – a fatal wound. On one occasion our Demarcation Party sent three runners for a considerable distance through the Haud; after doing 80 miles they had a rest which one of them was unwilling to end, whereas one of the other two urged a move. During a momentary absence of the third, the lazy one drew his knife and pursued his tormentor, finally succeeding in inflicting a severe wound in the back. The third man ran all the way to the Demarcation Camp for help, and Major Godfrey-Faussett, in the doctor’s absence, set off in a six-wheeler with bandages, etc. On arrival he found the victim lying in the sand, breathing through his wound and the assailant looking after him to the best of his ability. Lavish bathing of the wound in distilled water (that proved to be the entire stock of liquid quinine) and bandaging that compelled breathing through more normal channels were followed by a bumpy return journey along the boundary, and on reaching camp the patient was so far recovered that he asked to be sent into Burao, 100 miles away mainly through the bush. He arrived there sitting up alongside the driver. After a very short time in the hospital, he was fit to attend the District Court for the case to be tried, and only a few weeks later was back at duty. The assailant, a healthy sturdy young fellow, retired to Berbera Prison where he died within six weeks without symptoms of any sort!

The British Somaliland-Ethiopia Boundary
6. Old caravan road, boundary between “the hill of Somadu” and Habaswein

Somaliland used to be the great shikar resort for Aden. I think it was about twenty years ago, an epidemic of rinderpest occurred, and it was found necessary to make the Greater Kudu royal game. That protection was entirely effective and has now been relaxed. On the other hand, Swayne’s hartebeeste which used to be extremely plentiful on the plains north and east of Jijiga, seems to be nearly extinct; I have only heard of one being seen for certain in recent years. Lesser Kudu are numerous, as are Soemmering’s gazelle, Pelzeln’s and Speke’s dhero, and genenuk. Klipspringer and baira, the latter peculiar to Somaliland, are by no means rare in the mountains, and that peculiar creature the dibataq, a gazelle with the horns of a reed-buck and a short, stiff, furry tail which is erected vertically when the animal is alarmed, is to be found all over the Haud, except in the west. Dikdik of several sorts abound. There are lions, leopards, and cheetahs, and a good variety of vulpines. Of birds, there are ostrich, francolin, and bush partridge, guinea fowl, both common and vulturine, an occasional duck or Egyptian goose, and we once met snipe in the south of the Haud. But it is markedly a country of bustards, greater and lesser, in a number of different varieties. Besides that, immense numbers of sand-grouse visit certain water centers. There was therefore no justification for the advice given to Major Godding – that it was no use his bringing out firearms, though he might find a fishing-rod useful!

The British Somaliland-Ethiopia Boundary
7. Demarcation across cultivation at Dababur, near Borama

The Anglo-Italian Somaliland boundary had been demarcated in 1929-1930, the work being carried as far west as 47° E., 8° N. Our work, therefore, started from this point, but with the qualification from the Ethiopian Government that this is not necessarily the Anglo-Italo-Ethiopian trijunction point, since the Italo-Ethiopian boundary is still indeterminate. From here the boundary was to run in a straight line to Aranare (spelt Arran Arrhe in the Treaty). This however is an area of about 50 square miles, and the name is not particularly rare, for aranar is a very common variety of small thistle, much appreciated by camels. So we agreed to adopt 44° E., 9° N. instead. The boundary, by Treaty definition, should then run in a straight line to Moga Medir, a point which we had little difficulty in identifying as the rock nowadays called Jifu Meider. It is the most striking feature in this part of the country, consisting of a number of huge rocks wedged together in quite a remarkable manner. We then persuaded the Ethiopian Section to agree to abandon the straight line and to use the bed of the Tug Wajale, a long depression containing a number of important seasonal water-points, where it would be possible to exercise a certain amount of veterinary control on migrating stock. The westernmost corner of the boundary is defined in the Treaty as “the hill of Somadu.” But there is no hill of this name, and it seemed evident to us from old documents that the most prominent feature of the neighborhood, called Beyu Anod, was intended. A compromise was eventually found. The Treaty gave us little help as to how the line should run between Jifu Meider and this corner – which was just as well, as the ground was unknown. At the cost of much clambering, we eventually selected what I feel sure is the best natural feature line through this stretch. Plate I shows for instance what a perfect line is afforded by the crest of the main ridge of Marmar Gedle. From Somadu towards the sea, we were required to follow an old caravan track, passing through the lava country which appears to have been the route generally used forty or fifty years ago between Zeila and Harar in preference to an easier one a little farther east, on account of the water problem. Nowadays however the eastern route is customary.

The British Somaliland-Ethiopia Boundary
8. The boundary scraped through lava boulders and gravel, between Somadu and Dulusbo. The T-shaped enclosure is an aeroplane identification mark

The Anglo-Ethiopian Treaty merely defines the boundary of British Somaliland, without specifying the trijunction point at either end of the Ethiopian portion. This is understandable as regards the eastern end since the Anglo-Ethiopian Treaty preceded the Italo-Ethiopian. The Anglo-Franco-Ethiopian trijunction point had to be deduced by a compromise between the three relative documents – the Anglo-French Agreement of 1888, the Franco-Ethiopian Convention of 1897, and the Anglo-Ethiopian Treaty of 1897. This necessitated determining the position of the Anglo-French boundary point on the coast, described in the Agreement as “opposite the wells of Hadou,” a name that is unknown. Fortunately for us, the local authorities had agreed a year or two after the signature of the Agreement to recognize a prominent palm-tree at La-waada as the point in question, and, though this tree has been dead for many years, we were able to locate its stump. For this part of our work, we were joined by representatives of the French Government, and I would like to record that, though circumstances forced some extremely tiresome discussions on us, we made some very good and enduring friendships. Having settled the trijunction point, we selected a natural feature boundary back to Somadu, in place of the caravan track.

Our instructions required us to start demarcation work at the eastern end. So, after locating and marking the corner at 44° E., 9° N., both Sections sent their main parties to the boundary just south of Bohodle, an important center in the days of the operations against the Mullah at the beginning of this century, where the ruins of our old two-storeyed stone fort are still visible. On the way there one passes at Kirit another relic of that period, a derelict steam engine that was installed for water-supply duties by the Expeditionary Force. Having located a point on the boundary, the two sections started demarcation in opposite directions, working to a detailed specification, the main features of which were: (a) a lane cleared of bush for 2 meters on either side of the line so that our motor vehicles could use it as a road. (b) boundary pillars so sited that from any point on the line one should be visible in each direction, the interval between pillars in no case exceeding 2 kilometers. The Tug Wajale presented rather a problem. Its channel, though definite, is small and in many places could easily be passed by anyone not on the lookout for it; and we were afraid of doing anything that might tend to reduce its water-holding capacity. The channel was cleared of bushes, and small survey marks were erected in pairs on either side. In addition, to serve as a warning of its proximity as well as to facilitate frontier patrolling, a motorable track was made along each bank. I am afraid it remains to be seen whether this will provide a really satisfactory solution; we could not think of a better. In the undulating country between Jifu Meider and the maritime mountains, the boundary crosses a certain amount of cultivation. The first attempt to guarantee the preservation of the line was made by the Ethiopian Section and took the form of laying a bed of stone. Needless to say, the transport involved proved prohibitive, and another scheme was evolved which has stood the rains in a way I did not expect. This consists of pairs of ditches, the bank between them being the boundary; the ditches are in short lengths, to reduce the danger of their becoming watercourses.

The British Somaliland-Ethiopia Boundary
9. Gerigoan, east end, a detached mass opposite the western end of Libah-hele range

We met another problem when we reached the lava country. After a good deal of head-scratching and some trial work, we decided to “scrape” a path through the boulders and gravel, astride the line. I am confident that this gives a permanent result; for paths are rare and bad, and this will undoubtedly be used by men and animals; while its straight line and regular edges declare it to be artificial. The majority of our boundary pillars were dry stone cairns, though some were cement concrete. Altogether there were 1,166, of which 33 were masonry; and to this must be added 54 pairs of cement survey marks along the Tug Wajale. Across the central part of the Haud stone was a difficulty. Special reconnaissances were necessary, and our two six-wheeled Lorries were kept fully occupied on this duty alone, the distance to the outcrops being sometimes as much as 10 or 12 miles.

The tribes displayed a good deal of disapproval of the boundary, for they feared that it implied restrictions on their grazing and would not be persuaded to the contrary. Consequently, a good deal of pillar destruction occurred. The Somaliland Government imposed – and collected – a considerable total in fines and rebuilt the pillars for whose maintenance they are responsible, and the trouble appears to have been checked. The Ethiopian Government had similar measures under consideration, but other and greater preoccupations have so far prevented their application. The rates of progress of demarcation realized by the British Section were from 1300 to 1750 meters a day across the Haud; on average just about 1000 meters a day between Sau and Somadu; and 1600 meters a day between Somadu and the Anglo-Franco-Ethiopian trijunction point. In the case of this last stretch, however, all survey work had already been done, so that all effort was concentrated on demarcation. These figures afford complete testimony to the excellent organization and drive of the officers in charge of our demarcation party Major Godfrey-Faussett, and his successor, Captain Taylor. But one must not fail to recognize that it also shows the energy displayed by our three survey N.C.O.s and by the coolies. We completed the demarcation of the boundary at the end of June 1934, after which we were occupied till the end of February 1935 with completing our records, drafting the Boundary Agreement, and studying the trans-frontier grazing problem. Mr. Curle, Captain Taylor, and myself then went up to Addis Ababa, where the Agreement was signed on March 28; and so back to this country.

In the first part of our work in the Haud water was an unceasing preoccupation for the Demarcation Party. Every available barramil – the Somali name for the 12½-gallon water-tank used for camel transport – was handed over to them; but the camels took five and six days for the return journey, and it was never possible to establish any reserve of water at all. During this period it was usual for the last drop to have been issued by the time the daily convoy arrived. Finally, we filled a number of 4-gallon petrol cans with water, soldered them up, and sent them down. Fortunately, they were never actually required. I say, fortunately, for we found that odd bits of cotton waste and rags had been included with the water. As the party got nearer to Hargeisa the situation eased very greatly. All through the latter part of our work in the mountainous country water was never a real source of anxiety, as it was never more than one day away for the camels, and generally only a matter of hours. In a number of watercourses, there are what the Somali calls durdur, that is to say, stretches of running water that die out and perhaps appear again a little farther on. The strength of flow, of course, depends on the season, and some durdur are liable to cease completely in a dry season; but even so one can generally count on finding water in a durdur stretch with a very little digging. Thanks to the geological faults, there are quite a number of springs, some of which are mineralized and hot.

All across the Haud our work was controlled by astronomical methods, for the flat, bush-covered nature of the country made triangulation impracticable. Once out of the bush, however, triangulation came into its own, right through to the coast. It is not often nowadays that one is required to measure a base for triangulation. A good deal of study was therefore needed for both sitting and procedure; Brigadier Jack’s admirable report on his work on the western frontier of Uganda eventually told us what to do and why. By a curious coincidence, our base measurement gear was also that used by him in 1908. It had been done up for us, but I personally should hardly have described the “frictionless” pulleys in those terms. I understand that the age and behavior of these invar tapes are a matter of some interest, but I have not been able to examine the point yet. Our theodolites were all British, from Messrs. Cooke Troughton and Simms – an 8-inch micrometer, against which we had a number of criticisms, but which nevertheless produced most satisfactory answers; two of the new Tavistock model, which were used for the triangulation and all general work with very remarkable results (though we had trouble in one respect with one of them, I cannot sufficiently congratulate the makers on the production of this very delightful little instrument) and an ordinary 5-inch micrometer, which was kept in reserve and hardly ever used. Topography for the map of the boundary was obtained by air survey methods, and for this purpose, a special Air Survey Flight, under Flight-Lieutenant Cator, was detached from Aden. The particular point of interest, in this case, was that photography was carried out ahead of the groundwork. This fact produced a crop of problems of its own in the navigational field, particularly during the first few months when working over the Haud. Lack of meteorological data was also a difficulty, records being naturally available only for the stations. Photography actually started at the end of January 1932, but suffered from interference from cloud (frequently in the form of that bug-bear of survey photography, high detached cloud) until early September. Delay had also been experienced on account of sickness and one or two misfortunes such as crashes, though I do not wish to suggest that the Flight had anything but a thoroughly good record in that respect. September, October, and November saw rapid progress; but, with the job all but completed, the weather broke up completely about the middle of December. Little was achieved during January and February, and still less in March; at the end of which month the Flight was withdrawn to Aden. The last few runs of photography were made later in 1933 by aircraft specially sent over from Aden. Some idea of the magnitude of the task may be gained from the resulting total of about five thousand exposures for mapping: though it must be admitted that, owing to the photography being done in advance of the groundwork, and indeed as far as the western part is concerned, in advance even of the political decisions, a much greater area had to be covered than the narrow band 8 kilometers wide required for the boundary map. This figure also includes an 8-kilometer belt along the Anglo-French-Somali frontier, which we surveyed but did not demarcate.

The performance of the Air Survey Flight calls for comment. Even allowing for Flight-Lieutenant Cator’s very considerable experience, the standard of the photography was first class and that of the navigation as near perfection as is, I should say, possible. Having the photographs taken ahead of the groundwork was, as you may imagine, of the very greatest help to the surveyors. Not only did it provide a most valuable reconnaissance, but it made matters so easy, working on the photographs themselves, to provide all the data required for making the map. With the practice they got, Captain Taylor in particular, and Corporal Griffiths to a lesser degree were able to extract almost any information needed from the photographs, one very useful item towards locating themselves being the age of a thorn zariba.

I have already mentioned that the Treaty accords trans-frontier grazing rights; and, besides demarcation of the frontier, the Commission was required to submit a joint report on the extent of the individual grazing grounds on either side of the frontier. We, therefore, made a reconnaissance map on a 1/125,000 scale almost wherever we went. In the west, this was controlled by our triangulation, in the east by the boundary demarcation and a number of astronomical points. In connection with this work, I cannot refrain from mentioning the very fine performance of Captain Taylor, who in one particular period of sixty-two days traversed 1435 miles, 153 of them on foot, the remainder by six-wheeler, with an additional 939 miles traveling but not traversing. During this same period Corporal Griffiths, in addition to helping me with astronomical and various other duties, traversed 160 miles on foot and 626 by six-wheeler. The mapping of Somaliland is still very incomplete. The Expeditionary Force at the beginning of this century and the Anglo-Italian Boundary Commission in 1929-30 have more or less completed a large block. The work of Captain H. G. C. Swayne, R.E., and Lieutenant E. J. E. Swayne, Bengal Infantry, between 1886 and 1892 covered a big area in the west; while in 1903-04 Major S. L. Craster, R.E., ran a railway reconnaissance from Berbera to Harar. The rest of the map is compiled from route reports and “travelers’ tales,” and is highly unreliable. Tracks are shown where they do not – or at least do not now – exist, under the wrong names, and leading to the wrong places.

We had a certain amount of difficulty in establishing a datum point on the coast for our heights. The behavior of the tides between Zeila and Bab el Mandeb is somewhat eccentric, and we endeavored first of all to connect on to some recent French hydrographical triangulation round Djibouti. But I regret to say – or should I be glad to say? – that we hit on an unexplained error in that work, which put our point on the shore about 25 meters below sea-level. So we had to fall back on the obvious and level between high and low watermarks on several days. Place-names were a heavy task. We had 1500 of them. For ordinary British purposes, spelling was required under the R.G.S. II System; but the Commission’s official language was French and a French version of the names was therefore required. This has forced on us the preparation of a gazetteer with the equivalent spellings. In some areas it was very difficult to ascertain place-names, different guides having quite different ideas. Another trouble arose from the frequently apparent indifference of the Somali in the matter of choice of vowels; in general we worked by the meaning and, for the sake of consistency, ignored minor variations of pronunciation. The Somali language belongs to the Arabic group, though quite distinct from and with a greater variety of sounds than Arabic. The R.G.S. II System consequently is easily applied.

In matters of health, we had our share of bad luck. Early in 1933 two of our officers developed Malta fever and we were fortunate in being able to evacuate them by air to Aden. Both were extremely ill and indeed nearly died. Major Twigg has fully recovered; but Squadron-Leader Lindop, after more than a year in hospital and convalescence, has been obliged to retire from the Service. Then in June 1933, Mr. Plowman developed acute nephritis and again the Air Force came to our rescue and flew him to Aden. In February 1934 the Senior French representative, M. Roussan, had sinus trouble and had to be evacuated. Early in March 1934 Herr Beitz, the German chef technique of the Ethiopian Section was ambushed and killed by a party of Esa tribesmen. A week later Lij Zaude Balaine, Assistant Ethiopian Commissioner, had to be evacuated to Addis Ababa with acute appendicitis. Ato Lorenzo Taezaz, who replaced him, developed pneumonia in June. We had only three cases of malaria among the Europeans of the British Section; but one of these, Corporal Marshall, who must have been infected only a day or so before he left us, died twenty-four hours after landing in England. Among the native staff we had one epidemic of malaria and a variety of other diseases, but relapsing fever was the favorite and could always be relied upon to appear after a short spell at any of the towns. But when all is said and done our casualties attributable to climate were remarkably few, considering that they were spread over a period of thirty-eight months all spent under canvas, without a break. Local leave offered little in the way of attraction – least of all to the N.C.O.s – it would have been too much of a busman’s holiday. So, besides encouraging all forms of outside interests, we relied on a change of occupation wherever possible, for maintaining moral and preventing staleness.

It is difficult to describe the climate briefly, as there are such contrasts. The coast is delightful from November to February; but appalling between May and September during the kharif season. The wind then seldom eases and carries vast quantities of sand, while the shade temperature ranges round about 108° F., and there is little relief at night. When demarcation reached the Anglo-Franco-Ethiopian trijunction point at the end of June 1934, thinking to afford a welcome change from the monotony of demarcation, I arranged with Captain Taylor for the party to return to headquarters at Borama, traversing for our reconnaissance map on the way. We had neither of us realized in the least the severity of conditions in the Zeila Plain in early July. There was this unending sand-laden wind; visibility never exceeded 400 yards; the temperature I have already mentioned; and the atmosphere was so dry that mouth and throat became so parched that no amount of liquid made any effect – one inevitably reflects on the wisdom of the Touareg veil in the Sahara. Fortunately, they reached the mountains in a week and were able to restore their moral and health in the fresher atmosphere and green surroundings due to recent rains. In the interior – I am speaking of the boundary zone – we were at altitudes varying between 600m at the eastern end and 1500m in the Borama area. Our base camp was at Borama for just over two years. There the air is good and the scenery pleasant, and the average annual rainfall is about 20 inches. Round about Christmas, the night temperature falls to very near freezing point. The coldest place we met was Jijiga, where in early January 1935 water in chaguls hanging outside our tents froze at night – to the amazement of many of our staff who had never seen ice before.

During the winter of 1932-33 Mr. Farquharson, Geologist to the Somaliland Government was attached to the Commission and made two reconnaissances in the mountainous country to be crossed by the boundary. His reports were far too technical for my comprehension, but may be summarized by saying that he had a very interesting time but found nothing of any value. In the botanical field, Major Godfrey-Faussett collected a number of specimens of grasses during 1932. In September of that year, the Director of Kew Gardens sent out Mr. Gillett, who, based on the Commission, worked a considerable area round Hargeisa and Borama before going on to Harar in the middle of February. After that, Major Godding carried on the work. Captain Taylor had developed an interest in reptiles when serving with the Anglo-Italian Boundary Commission. Before coming out again he studied the subject at the Natural History Museum, to which he has now sent a collection of about two thousand specimens. Mr. Curle made a collection of ethnological specimens, with special relation to the Esa tribe, for the British Museum. Burton, in his ‘First Footsteps in East Africa,’ describes a ruined town, Abasa, which he visited in 1854. Several others have been known for some time past, but no knowledge of their origin was forthcoming. The Somalis can contribute nothing. Mr. Curle, who is the son of the well-known Scottish archaeologist, made a series of investigations, assisted by Captain Taylor, at no fewer than fourteen sites within reach of Borama, and their finds have been sent to the British Museum. These included coins which have been identified as of Kait Bey, Sultan of Egypt 1467-97, and this leads one to the conclusion that these towns probably belong to the period of the Adal Empire. In the western part of the country, by which I mean the mountains and the lava area, graves are often quite elaborate, the mound of the grave itself being decorated with bands of colored pebbles; a neat low wall, about 18 inches high, surrounds a clean-swept space, generally more or less rectangular. Miniature alcoves in this wall, as often as not covered with a flat stone, each represent a hut, that is to say, a wife. We met a variety of shapes, some possibly of some age and suggesting pre-Islamic origins or at least influences. Side by side with these one finds surface burials, a hollow cairn being raised over the body. In the neighborhood of the French frontier, we met a number of enclosures of unknown age. The Somali knows nothing about them. One is tempted to describe them as halting places for caravans, but for the fact that some of them are not near tracks.

The British Section was a large party. At its maximum it included besides myself:

Major B. T. Godfrey-Faussett, M.C., R.E.

Mr. C. H. F. Plowman, O.B.E., Political Officer.

Squadron-Leader V. S. E. Lindop, R.A.F., Air Liaison Officer.

Major T. H. Twigg, R.A.M.C., Medical Officer.

Captain R. H. R. Taylor, R.A.

Mr. R. A. Farquharson, Geologist.

Mr. J. B. Gillett, Botanist.

Mechanicist Staff-Sergeant D. G. P. Shelley, R.A.S.C., in charge of our motor transport.

Lance-Sergeant E. V. Elkins, Royal Corps of Signals, wireless, clerical, and store duties.

Corporal J. B. Marshall, R.E.

Corporal R. Griffiths, R.E.

Sapper S. L. Terry, R.E.

Mr. Plowman was replaced by Mr. A. T. Curle; Major Twigg by Major H. C. Godding, M.C., R.A.M.C.; and Staff-Sergeant Shelley by Corporal A. V. Wells, R.A.F. The Air Survey Flight, commanded by Flight-Lieutenant F. G. Cator, R.A.F., consisted of twenty senior N.C.O.’s and other ranks. Our native staff varied according to the work in hand, but on average totaled around about 100 to 120. We were provided with an escort by the Somaliland Camel Corps, and for part of the time by the Somaliland Police, the total strength varying according to circumstances.

The Ethiopian Section was headed by Pitaurari Tasama Bante, one of the Emperor’s more immediate following, who had been in charge of H.R.H. the Duke of Gloucester’s shooting trip to the Arusi early in 1931. I have nothing but the pleasantest memories of my dealings with this typical Ethiopian gentleman of the intelligent “old school,” though I acutely felt the lack of a language common to us both. He had as assistant Lij Zaude Balaine, a young man who had spent some twelve years studying engineering in France; and also for the last year of our work Ato Lorenzo Taezaz, who had the advantage of a first-class legal training at Montpellier University and speaks most fluent French and Italian. Their technical staff were all Europeans of varying nationalities. Starting with two Frenchmen, two Dutchmen, and one Hungarian, they later had two Germans, one Czecho-Slovak, and one Bulgar. In addition, they had two Syrians in charge of their mechanical transport. Our relations with the Ethiopian Section, and indeed with the Ethiopian authorities in general, were throughout of the best. To venture into the fringe of the political field, the essence of the work of a boundary commission is the improvement of frontier relations. In this case, it is useless to deny that at the start there was considerable room for improvement, due primarily to ignorance or out-of-date ideas on both sides. I cannot say whether we or the Ethiopian Section worked the harder on this aspect; but if we succeeded in having as good an effect on the Ethiopian side as Fitaurari Tasama and his staff achieved on the British, we have every reason to be satisfied. We certainly received a great many kindnesses and a lot of help from the emperor and all grades of Ethiopian officialdom with whom we came in contact.

In Somaliland, we were of course very heavily indebted to the Governor, Sir Arthur Lawrance, for many acts of hospitality and personal kindness, in addition to our excellent official relations. I will not attempt the invidious task of mentioning individually the heads of departments, administrative officers, and officers of the Somaliland Camel Corps, from whom we received so much help, often at the cost of considerable additional work to themselves, and so much hospitality. Our visit to Addis Ababa was spent at the Legation and proved a most delightful contrast to the preceding years, as I am sure those who know Sir Sidney and Lady Barton will readily imagine. Nothing could have been more helpful than the attitude of the Royal Air Force at Aden. They lent us stores, notably a wireless transmitting set which proved of untold value; they met all sorts of needs either from stock or by local purchase, thereby saving us the delay and freight charges involved in ordering from Home; I have already said how they carried our hospital cases to Aden; and there was also really excellent co-operation in connection with the air survey work. I simply cannot express my gratitude to Air Commodore 0. T. Boyd, O.B.E., M.C., A.F.C., and his successor, Air Commodore C. F. A. Portal, D.S.O., M.C., and to their officers for the way in which they eased our path, particularly since it was all done from the kindness of heart – I had no call on their services apart from the photography.

Title: The British Somaliland-Ethiopia Boundary.

Publisher: Royal Geographical Society, London

Publication Date: 1936

About the Geographical Journal

The Geographical Journal is a quarterly peer-reviewed academic journal of the Royal Geographical Society (with the Institute of British Geographers). It publishes papers covering research on all aspects of geography. It also publishes shorter Commentary papers and Review Essays. Since 2001, The Geographical Journal has been published in collaboration with Wiley-Blackwell. The journal was established in 1831 as the Journal of the Royal Geographical Society of London. Prior to 2000, The Geographical Journal published society news alongside articles and it continues to publish the proceedings of the society’s annual general meeting and presidential address in the September issue.

This site uses Akismet to reduce spam. Learn how your comment data is processed.