Udgravninger på Bahrain. Dansk Arkæologisk Bahrain-Ekspeditions 2. udgravningskampagne


  • P.V. Glob


Temple, Bahrain, second excavation campaign, tempel, anden udgravnings kampagne, Barbar, Qala'a


The Danish Archeological, Bahrain-Expedition's Second Excavation Campaign.

When in May 1954 the humid heat of the desert and a shrunken purse brought the end of five months excavation on the little palm-crowned island of Bahrain in the Persian Gulf, the Danish Archeological Bahrain-Expedition had achieved such good results that a subsequent campaign was a foregone conclusion 1). The following expedition was moreover speedily ensured by the generous assistance and economic support of the Government of Bahrain, and by the receipt of considerable grants from the Carlsberg Foundation, the Bahrain Petroleum Company, and the Pennsylvania University Museum.

The archeological staff of the expedition was hospitably received 8th January 1955 in Manama, where headquarters were established from then until the termination of the work on 25th March. The Ruler of Bahrain, His Highness Sheikh Sulman bin Hamad Al-Khalifah, extended to us all possible support, through the medium of the members of the Government, in particular of Sir Charles Belgrave, K. B. E. In the villages our former workmen flocked around us, eager to be re-engaged for the work. In addition to the expedition's members from the previous year, T. G, Bibby, M. A., Kristian Jeppesen, M. A., and the present writer, the staff consisted of Mogens Krustrup, M. A. and Dr. Robert Dyson of the University Museum of Pennsylvania. A house was rented for accommodation, and meals were taken at the Guest House of Petroleum Concessions Limited, where David Cooper was our genial host.

While the previous year it had been necessary to start from scratch, and a month passed in examination of suitable sites for excavation, on this occasion we were able to turn immediately to the continuation of the investigations which we had already begun with such promise. The large-scale excavations of the temples at Barbar and of the prehistoric capital at Qala'a required, however, a certain amount of preparation including the obtaining of heavy equipment; in this the Director of Public Works, Mr. Lees, assisted us very generously. The ten days or so which were spent on this preparatory work were also used for an investigation of the southern desert where last year 12 sites were found containing flint artefacts from the Palæolithic and Neolithic Periods 2). This number was increased by the discovery of 15 new sites of the same character and age as those discovered earlier. They produced, however, a number of new artefact types, while at two locations under the cliff outcrops north of Jabal Ad-Dukhan the flint appears to have been washed out from stratified deposits. There was unfortunately no time to investigate these sites in greater detail, but excavations will be carried out there at the earliest opportunity.

The greatest effort was set in at Barbar, where the 3 metre high gravel mound, which had been heaped above the upper temple after it had been razed to the ground, was removed, baring the temple precincts in their full area of 3600 square metres 3). The first year's investigations had shown three temple buildings superimposed the one above the other, the earlier building having in each case been razed to the foundations before the next temple was built above them. The inner precincts of the two upper temples were, in the main, exposed (Fig. 2), and sections laid bare portions of the earliest temple (Fig. 3). In the western area of the site a stone setting unshaped blocks was moreover discovered, belonging to an even earlier phase of construction (Fig. 4 A).

The stages of construction of the temples were revealed by means of an east-west trench dug down to original ground level in the inner court of Temple III. Before Temple I was erected the topsoil had been removed from the area, and upon the virgin subsoil a mound of clean blueclay had been built up above a layer of sand which varied in thickness from two or three to 40 cms. This mound was thereafter covered with a layer of clean sand, the whole complex thus measuring a scant two metres in height. The earliest buildings, of which only part of a circular structure of uncut stone and gypsum mortar has yet been excavated, were erected on this mound, which was surrounded by a stone wall (Fig. 3, I) which appears to have enclosed a square area of 400 square metres. In contrast with the remains of the buildings from Temples II and III, which are of stone brought from the little rocky island of Jida off the northwest corner of Bahrain, this wall foundation is constructed of the local stone. In dismantling Temple I the central cult-complex was allowed to stand and the new altars erected directly above it. One of these altars was partly excavated, a very finely turned circular stone block, 0.43 metres high, 1.2 metres in diameter at the top and slightly less at the base, so that the shape was very sligthly conical. The upper surface of this circular altar is somewhat concave. It belonged to Temple II, and directly upon it rests the stone flagging around the altar complex in Temple III, the floor level of which lies about 2.8 metres above the ground surface. It is worthy of notice that, in the dismantling of the temples, the cult-complexes have in all three cases been left practically untouched.

When the mound upon which Temple I was erected was under construction there was deposited in the thick layer of clean clay, about 0.5 metres above ground level and in an east-west extension of about 5 metres, a score of pottery beakers (Fig. 5), together with a copper beaker and a narrow gold band 4). At another point lay a little heap of various objects of copper, including an axe-head and a copper disc 13 cms. broad 5). A similar axe-head was found in the stone-quarrying layer by the east side of Temple III. As the section trench in which these objects were found is only 9 metres long and 1 metre wide the temple mound may well conceal many more treasures. It is at least known with certainty that only a small part of the pottery beakers has been recovered, as several of these are still in situ in the trench walls both to the north and to the south. A further noteworthy feature is that, about 20 cms. above ground level and enclosed in the layer of clay, were very thin layers of virgin yellow subsand of the same type as that covering the flooring slabs around the altar complex of Temple III.

Belonging to Temple I are 8 large cut blocks of stone, set up in a double row by its western side (Fig. 4, I and VI). These blocks run under a ramp which leads up from the west to Temple II (Fig. 4, II), and this ramp is in turn topped by the outer wall of Temple III (Fig. 4, III), which was, however, removed before the picture was taken. These stone blocks, which measure about 1 metre in length and ½ metre in width and are about ½ metre high, have two square holes cut in the upper surface, and are thus of the same type as the blocks associated with Temple III which were the cause of the discovery of the temples, two of which have been exposed at the northern side of the site. They are of somewhat varying size and one of them has, in addition to the large square holes, a smaller circular hole in the centre of one side near the edge (Fig. 4, extreme right). Where these holes had not been emptied a sleeve of copper sheathing was found in them, the space between them and the stone being filled out with bitumen. Through the sheathing rows of copper nails had been driven, with the points inwards, and in many cases remains of wood could be seen between them. These nails have been found in large numbers everywhere on the site and have been previously mentioned 7), but their function is now for the first time clear. It is now obvious that they were used to secure wooden supports in the holes in the stones. These wooden supports have been of about 20 cms. in diameter. It is possible that the stone blocks have acted as foundations for human figures, wooden statues of gods, with one foot fitted into each of the holes. In the case of the block with the third hole it would be feasible to suppose that the figure held a staff or the like. On one of these blocks there was found on one side two figures of men crudely carved in a technique corresponding to that of the rock carvings of the Scandinavian Bronze Age (Fig. 7). The figures represent two people, the one with down­turned arms and the other with arms uplifted. It is possible that these figures show us two of the statues which originally stood upon the plinth-stones. Two further figures can he dimly traced on the same block to the right.

A discovery of the highest interest, probably belonging to Temple II, was made in the corner where two wall foundations met. It consisted of a magnificent ox-head of copper (Fig. 1) and a large collection of copper sheeting in strips 15-20 cms. broad, all pierced with many rows of copper nails. It lay in two layers within an area of half a square metre, and together with it was found a copper ring and a copper band of thicker metal, 4.5 cms. wide. All were buckled and folded. It is possible that these copper strips were originally attached to an object of wood on which the ox-head was mounted.

In Temple III various stones of considerable interest were discovered. During the first year an overturned stone with a rounded top and pierced with a circular hole was found behind the circular complex 8). A second stone of the same type, standing in situ, was found at the same time, and this year two further stones, one with a projection in the form of an animal's head, were found in line with it (Fig. 8). These stones, which were perhaps intended for tethering sacrificial beasts, as considerable marks of wear from the hole out to the edges would suggest, were compared with the ringstones of the lndus Culture 9). This comparison is further supported by a somewhat damaged stone of the same type in which the ring is emphasized by carving. It was found lying in the stone-quarrying layer (Fig. 9).

A stone of unusual character was found lying south of the altar stones in Temple III, and perhaps originally lay as a flooring stone behind them. It is a slab, 64 cms. long, 26 cms. wide and 20 cms. thick, with in the centre a cup mark, 12 cms. wide and 6.5 cms. deep, as well as a runnel about 2 cms. wide and 8 cms. deep, running out to one edge (Fig. 10). It is possible that a linga figure originally stood in this hole. Cup marks with the same significance were found during the first campaign on a stone block near Diraz, but this is now walled into a mosque and is no longer visible 10).

The earlier descriptions of the ground plans of the Barbar temples 11) can now be revised with greater confidence on a basis of the greater area uncovered. Final measurements cannot, however, yet be given (Fig. 6). The first temple, Temple I, was constructed on top of the artificial mound of clay and sand already described, and its surrounding wall appears to have formed a square with sides about 20 metres long. Outside this wall towards both south and east have lain various buildings, portions of the foundations of which have been uncovered. On the western side stood the 8 large plinth-blocks. Temple II has possessed an inner core, approximately square and measuring about 25 X 25 metres, with a wall thickness of somewhat over 2 metres. The wall was built of solid ashlar, finely cut rectangular blocks of stone of various sizes but carefully fitted together and joined with gypsum mortar (Figs. 2-3). Here too an outer complex is to be found, including to the west a stone-set ramp (Fig. 4). Stones from Temple II were undoubtedly used to a wide degree in the construction of the last temple, Temple III, the inner court of which is somewhat smaller, measuring about 16 X 16 metres. On the west side of this courtyard are remains of rooms, one measuring 3 metres in width, its length being impossible to determine on account of stone-robbing (Fig. 11). It is possible that this room originally belonged to Temple II. A massive wall of unshaped blocks set in gypsum plaster was constructed around Temple III. It appears to have formed a square of about 40 X 40 metres, externally measured, and beyond its north side stood the two large plinth-stones 12). The outer wall of Temple III was of imposing dimensions, being about 4 metres broad and faced with massive rectangular shaped blocks of stone, which are preserved for a considerable distance along the west side (Fig. 6).

On a basis of the objects discovered in Temple III during the first campaign this building was dated to the Third Millennium B. C. 13); but it is perhaps considerably earlier. A fragment of a stone vessel, ornamented with concentric circles, from the floor level of Temple III is of the same type as a vessel from a grave at Ur which must be dated to predynastic times 14). It is probable that the pottery also points to the same period. It is characterized by a hard reddish or red-brown surface covered with horizontal ridges (Fig. 12). This ware is completely dominant in all three temple levels and is undoubtedly the characteristic Bahrain ware of the period. It is still impossible to determine how long it has been in use, but it is probable that three specimens found in predynastic graves at Ur are exports from Bahrain 15), with the results that the temples at Barbar must be attributed to the period around 3000 B. C. or perhaps a few centuries earlier. A definite connection of the first importance between Bahrain and Ur at that period is thereby established, though only a direct comparison of the pottery can, of course, confirm with certainty such predynastic connections; a comparison of ordinary unspecialised pottery will always be inconclusive when based only on the published drawings.

Much remains to be done before the investigation of the Barbar temples and the critical examination of the material thereby obtained is complete, but it should be possible to finish the work in the field in the course of a couple of excavation campaigns. Near to the temples here described lies a smaller mound, which appears to be of the same period and perhaps belongs to the same complex. It must also be included in the investigation.

At Qala'a, the site of Bahrain's prehistoric capital, where considerable remains of buildings were identified last year 16), work was continued at two points. From the coast a section was dug in towards the centre of the "tell". Considerable remains of buildings were found and a long series of levels investigated from the lslamic settlement in the 16th century back to the Parthian period. The considerable collection of pottery from this excavation will provide an excellent basis for a chronological subdivision of this period.

The excavation of the large building in the centre of the "tell", which was commenced during the first campaign 15), was continued and a length of 8 metres was exposed without the end of the building being reached. Its breadth proved to be 3.7 metres. It had been divided secondarily into five smaller rooms. Between the two lowest floor levels were found sherds with ridged surface of the same type as those from the Barbar temples, which gives a probable date for the building to the end of the Fourth Millennium.

At two points in the building sondages were made right down to the bottom of the "tell" through about 2.5 metres of occupation deposits, revealing about ten occupation levels, in which a few sherds of painted pottery were found with patterns in black or red on a yellow ground. The pottery appears to show a connection with the lndus Culture, but only the discovery of further material can give a more certain basis for this view. The material at present obtained came from an area of only two square metres, but, as the levels are extremely rich in remains, future investigations should give a sufficiency of evidence. An important consideration is that these levels contain, in addition to pottery, considerable quantities of animal bones and charcoal. A further pointer towards the lndus Culture was a microlith core from one of the lowest levels 17).

A valuable encouragement in the often difficult and strenuous work of excavation was provided by the interest with which our work was regarded by all classes of the population. Visitors came constantly to our excavation sites and followed the progress of the investigations. Among others Dr. Henry Field, of Florida, U. S. A., took part for two days in the work at Barbar and Qala'a, as well as in reconnaissance of the desert area where the palæolithic sites are to be found. 25th January T. G. Bibby gave a survey of the earlier work of the expedition to the newly instituted "Bahrain Archæological and Historical Society", while the visit of that Society to our excavations the 18th March, under the leadership of the Society's President, Sheik Rashid bin Khalifah Al-Khalifah, and its Secretary, Mr. James H. D. Belgrave, who have both given the expedition constant and many-sided assistance, formed a pleasant termination to the excavation campaign. For their continual interest and their visits to our excavations we would like to thank His Highness Sheikh Sulman, the Ruler of Bahrain, the adviser to the government Sir Charles Belgrave, and the management of BAPCO, the vice-president E. A. Skinner, and the general manager, C. R. Barkhurst.

P. V. Glob





Glob, P. (1955). Udgravninger på Bahrain. Dansk Arkæologisk Bahrain-Ekspeditions 2. udgravningskampagne. Kuml, 5(5), 178–193. Hentet fra https://tidsskrift.dk/kuml/article/view/97204