No. xxxiv. The first is the picture of a musician seated upon a stool. He wears a long skirt with waistband, but the details of his dress and features call for no special comment. The instrument which he plays, however, is of interest, being[674] ‘a Tambur of pronounced Assyrian type, exactly similar to one of the time of Assur-nazir-pal,’ cir. 880 B.C.[675] There is a cord for attachment to the shoulder. It has been suggested, alternatively, that an instrument like a harp seen from the side is intended; the position of the musician’s hands, however, does not support this view, as with his left he clasps the stem of the instrument which he plays[676] with his right. On a small stone placed above this one there is the outline of an eagle or vulture.[677]

No. xxxv. The sculpture on the next stone shows a man who seems to be in relation to the musician towards whom he is turned. His dress is a short skirt, and he seems to have worn, in addition to the upturned shoes, some prominent ornament around his ankle. The hat is close-fitting and ribbed laterally. The beard and curl are as usual. His left arm is bent, and with his fingers he seems to touch his beard; the right hand is obliterated.

Nos. xxxvi.-xxxvii. The last two sculptures of this side may be taken together, for they represent the familiar oriental scene of a pair of goats standing upon their hind legs in order to reach the upper green shoots of a shrub at which they are nibbling.

The left-hand flanking wall to the approach of this main gateway seems to have been destroyed previous to excavation, and no sculptures are on record to form the counterpart to the group last described. In our description we have endeavoured to record the actual find-spot of each stone, but we are convinced, from an examination of the original monuments, and of the excavators’ photographs, that very few of them were found in the positions for which they were originally intended. Their varying sizes, the medley and lack of sequence of the subjects they represent, all contribute to support this point of view. Nor do we share with the excavators the opinion that all the sculptures of the outer gate of the town, which we described first, are necessarily older than those of the gate of the citadel; the former are more weathered, for they have been more exposed, and the latter are not, in our opinion, all contemporary with one another. We can distinguish three or four groups of subjects, which were probably ranged together, as at Eyuk and Sakje-Geuzi. Among these are the procession of mythological creatures and representations of the deities (possibly the king himself impersonating the gods[678]), the scenes of the chase, the musicians, and the Ceremonial Feast, all of which appear for the most part to be intermingled haphazard. It seems to us that the warrior-figure (No. ii.) might be as old as any sculpture of Boghaz-Keui, while others again might be as late as history sanctions. We believe them to be in the main the work of the tenth century B.C., and we regard the rearrangement to be due to one of the later restorations of the site, such as the local documents show must have been not infrequent, and possibly to the preparation of the buildings as summer palace of Esarhaddon in the seventh century B.C.

There are several further sculptures of considerable interest, including two pairs of massive lions which must have served as corner-jambs of doorways, like those of Sakje-Geuzi.[679] In this case, however, the lions are of enormous size. One pair is carved only in outline, while the work of the other is carried out in detail of admirable quality. Most striking of all, however, are two stone busts in relief, found near to the lions last mentioned.[680] Here we are face to face with that remarkable facial type and head-dress which characterises the sphinxes of Eyuk.[681] The stones are very weathered, but there can be little doubt, from the front and profile views which have been published, that they are each decorated with the bust and head of a woman. There can be distinguished the roundness of face, the high cheek-bones, the band across the forehead, the curving wig over the head, and finally (but faintly) the outward curl of the ends of the wig on each side of the throat, which are plain on the sculptures of Eyuk. The nature of the sphinx-bases, another series of sculptures of striking character, will become apparent in the description of the palace-portico at Sakje-Geuzi which follows.

Part III.—The Mounds and Palace-portico at Sakje-Geuzi.

We have already described the situation of the neighbouring site of Sakje-Geuzi, and the nature of some of its surface monuments. There are several prominent mounds in this locality: the sculptures mentioned and the palace ruins lately discovered[682] are connected with the smallest of these. Soundings made in the other mounds have made it clear that their nature is similar, and their growth collateral: in all probability they contain inscribed and sculptured monuments, the careful uncovering of which would contribute new pages, if not volumes, to our knowledge of oriental history. So far as excavation has proceeded on this site, it has been sufficient to determine the nature of the main fortifications, and to disclose within the walls the portico of a palace decorated with a frieze of sculptured slabs in their original positions and in fresh unweathered state. It is also demonstrated that here, at any rate, long ages of local development preceded the period which these striking monuments have rendered more conspicuous, though historically not more important. In the story of the decline and fall of the Hittite power, however, nothing could be more interesting than these sculptured monuments, with the increasing signs of Assyrian influence upon them, and the study of them becomes endowed with wider significance by comparison with those elsewhere. Not only can we measure, by the local differences and similarities to be seen in the works of Sinjerli and Sakje-Geuzi, the depth to which Assyrian feeling had already permeated the Hittite arts in the early centuries of the first millennium B.C.; but by comparing these again with those of Eyuk, we may realise how far certain features of architecture and religious symbolism were originally Hittite, and though here modified by close contact with the all-absorbing Assyrian power, remained on the farther side of the Taurus free from recognisable intrusion to the end.

In the small mound excavated at Sakje-Geuzi, the form of the main enclosure was found to be practically rectangular and four-sided, enclosing an area about a hundred feet long and eighty feet wide. A slight modification in form seems to have been made, either at the time of building the whole wall or later, where the north-western wall skirts the steep edge of the mound as it approaches the northern corner. No gateway was found, the lower side of the mound opposite the palace being almost denuded even to the foundations of the wall, which was found, in other places, six or eight feet below the surface. Nor did any outer rampart on the lower level correspond to the wide enclosure at Sinjerli. The wall was built of small stones revetted together by stouter facing blocks; these, though laid approximately in courses, were fitted together without much shaping and without mortar, as in Beuyuk Kaleh at Boghaz-Keui, and in the fortifications of Sinjerli. The wall was supported by external buttresses or mural towers, about thirteen feet wide, and projecting about three feet; these occur at intervals which decrease considerably around the steeper edge of the mound. The corners were similarly strengthened by rectangular turrets of the same projection. The wall was nearly twelve feet thick, and its foundations were proportionately deep and massive, as though destined to sustain a height of twenty feet or more, of which some thirteen feet remained preserved where the soil was deepest. The lowest courses of the foundations were built of large stones, suggestive of the masonry on Beuyuk Kaleh at Boghaz-Keui.[683]

Within the enclosure a series of superposed buildings on the higher ground gave token of successive ages of occupation, and partially covered the site of a palace, which was found at a depth of seven or eight feet below the surface. This has not been completely uncovered, but the details of the portico and the sculptures which adorned its façade, show that it was generally similar to the chief Hittite palace (of Aramaic times) at Sinjerli, which was still in use in the eighth century B.C. Probably most of the construction had been made in unburnt brick, which had largely been reduced to mud, leaving, as usual, little trace of original arrangement; but the sculptured slabs which had adorned the entrance remained standing in position, and enable the plan of the building to be traced. The doorway seems to have been divided and supported in the middle by a round column,[684] and to have been flanked on either hand by a square wing-tower, distantly suggestive of the Egyptian pylon. It was approached by two broad steps reaching from side to side, leading up to a platform or threshold paved with large flagstones. This formed the main threshold, twenty-four feet in width and seven feet in depth. The pavement was continuous in the wings only far enough to serve as foundations for the facing slabs of the corner towers: the ornamental pedestal just mentioned was also set upon it.

PLATE LXXVIII