It was, then, as it seems, the Chaldæans who invented the vault;[10] the want of timber compelled them in early times to contrive to defend themselves at once against the heavy rains and the ardour of a torrid sun; the creation of the vault was in their case instinctive and spontaneous. They raised, two or three thousand years before our era, vaults and domes like those which are built to this day by the rudest masons at Mosoul or Bagdad. No doubt the present state of the Chaldæan ruins and the insufficient explorations which have been undertaken among them do not enable us to say whether these Proto-Chaldæans knew every kind of vault, as the Assyrians did in the age of the Sargonids, or the Babylonians at the epoch of Nebuchadnezzar; but the remarkable perfection observed in their monumental structures, and in the very manufacture of the bricks, are so many arguments in favour of the inference that the palaces and houses of the Chaldæans in the time of Gudea were surmounted, for the most part, by semicircular vaults or by cupolas, as were later, according to Strabo,[11] the houses of the Babylonians. The vaults supported a terrace formed of clay; this layer of earth would be less thick over rooms roofed only with a ceiling of palm-beams and reed-matting. The ascent was by staircases, an example of which seems to have been found in the palace at Tello.[12]
While clearing away the material accumulated between the courts A and B, the workmen employed by the French explorer came into contact (at the point H) with a structure of baked brickwork, which proves that the Chaldæans at the remotest epoch had already invented one of the most interesting and characteristic elements of their architecture—the zikkurat or staged tower. The lower layers in the palace of Gudea alone exist, and are composed of two solid masses in stages one above the other. In its present condition the upper terrace is a mass 26 ft. square, 13 ft. less on all sides than the lower stage; perhaps there still exists a third and lower step, which has not been reached by the soundings, which are imperfect at this point. The zikkurat of Tello was not in any case so lofty or so important a structure as those of the Ninevite palaces or those represented by the ruins of Babil or Birs Nimroud at Babylon. It was even much less considerable than that which Taylor observed at Abu Shahrein, and which was equally old. These towers always had, from the first, seven stages, each painted of a different colour, and connected with the worship of the sun (Samas), the moon (Sin), and the five planets of the astronomical system of the Chaldæans.
The disposition of the royal apartments showed a striking analogy with that which we shall meet with again later in the palaces of Nineveh; there were the convenience and comfort which we find in the palaces of modern oriental sovereigns. To the Chaldæans again we must give the credit of having invented that architectural arrangement which springs from the necessities of oriental life, and is so well fitted to its needs that for four thousand years it has never varied. There were in the palace of Gudea three interior courts (A, B, C, [fig. 2]), round each of which the rooms radiated, and from which they received air and light. Each of these three groups had its own entrance, and communicated with the next group only by a single passage easy to guard or to close. The group of chambers situated in the northern angle (C) was especially isolated and removed from the others; it was the hareem or women’s apartments. At the eastern angle (B) were the rooms composing the seraglio or selamlik—that is to say, the part of the palace inhabited by the king and his officers; there was the saloon for official receptions, of which we have given the dimensions. This part of the royal dwelling communicated on one side with a state courtyard, measuring 55 ft. 8 in. by 68 ft. 9 in., and on the other with the outside by means of a smaller room serving as an antechamber; beside the door opening on the façade, boxes or recesses had been arranged in which the guards were posted. The third group of chambers, on the south-east (A), formed the Khan—that is to say, the dependencies of the palace, the kitchens, the slaves’ lodgings, and the stables.
All the rooms were paved with bricks; they very rarely led into one another, and had an opening looking on to the court. The largest of the doorways, that which opened into the state saloon, was of the unusual breadth of 6 ft. 6 in.; it was probably a folding door. Under each of the principal doors there was a great threshold of marble or alabaster, sometimes covered with an inscription and placed on a bed of bitumen and crushed bricks; under this concrete, finally, cylinders of precious stone and talismanic amulets were generally found.
The leaves of the door turned on pivots, the point of which rested in a cavity hollowed out for this purpose in a great block of diorite. M. de Sarzec brought to the Louvre a large number of these natural blocks, which were found buried in the pavement so as only to rise an inch or two above the surface. On the smooth surface of each of them it is seen that the socket, hollowed out in the form of a conical cup, has undergone an incessant friction; round the hole an inscription, sometimes circular, was engraved ([fig. 5]). To prevent the wooden pivot of the doors from wearing out too rapidly, it was enveloped in a metal sheath, which took the form of a funnel, and which was fixed to the wood by means of nails. One of these bronze cups has been found at Tello, still in place on the socket.[13]
Fig. 5.—Socket for pivot of door, from Tello (Louvre).