Chapter 7
The church at the sanctuary site (Fig. 39).
The church at the sanctuary site at Karakuyu has been neither excavated nor described in detail. Fortunately it has withstood the centuries well enough. Although much collapsed and partially obscured by vegetation, the ground plan can be traced and some of the walls survive to over a metre in height. However, we did not measure the church with surveying instruments and consequently it is difficult to be sure of the floor levels and to judge relative levels precisely. The measurements given in the account that follows should be treated as approximate.
The basic plan of the church was basilican, with a nave flanked by aisles on either side, a narthex at the west and an apse at the east end. The total length is c. 22 metres and the width c. 13 metres. The main entrance was in the short north wall of the narthex, and three doorways probably connected the narthex with the two aisles and the nave of the church.
The masonry of the apse is much weathered, but it was clearly semicircular both inside and out. There are crumbs of disintegrated white mortar in the joints between the stones and at the base of the apse wall. The two lower courses are intact, and the third complete except at the north end; only three blocks of the fourth course remain. The height of the blocks is even; their widths vary, but the vertical junctures alternate course by course. The thickness of the wall is aconstant 1.10 m, formed by one thick or two thin blocks. The internal diameter of the apse is almost five metres and it measures about 3.50 m from front to back.
The nave and the aisles are c. 10.5 m long, but there was a space of about 4 m between the apse and the nave which was occupied by a transept (PLATE 133). The eastern end of the nave is defined at floor level by five long, narrow stones laid end to end. At the opposite ends of the two southernmost stones there are squared depressions, joined to one another by a shallow channel. Since the northerly of these two depressions falls short of the longitudinal axis of the church, and five rather than four stones form the foundation of the division, it seems that the chancel screen at the east end of the nave was in two sections, with a gap in the middle represented by the central stone of the five foundation blocks.
South of the area in front of the apse, at the east end of the south aisle, there was a room about a metre wider than the aisle itself. This may have been a baptistery and the provenance of the font, which now lies some metres east of the building. The east and south walls of this room incorporate several decorated stones which originally came from the temenos wall of the Mên sanctuary. This room was joined to the south aisle by a door c. 1.90 m wide, whose threshold stone, marked with a rectangular indentation at the north end, is in situ. Since the west edge of the stone has a raised lip, the door must have opened into the baptistery. The juncture between this room, the reserved area in front of the apse, and the nave is marked by a large, upright block. Adjoining this on the east is another block at floor level, and a second block at the same level lies 1.35 m further east. There may have been an arcaded division here rather than another door.
Fig. 39. The church at the sanctuary site.
The ground on which the church was built sloped towards the south. After levelling, the south wall was constructed against the high ground which remained. Perhaps because it would not be seen from the exterior, this wall was less neatly built than the others, at least in the lowest three courses. The stones are of various sizes, including large, well shaped blocks and many smaller pieces, some roughly or only partially shaped. The north outer wall is hardly preserved. Its lower, partially buried course can be traced, but not as far as the north-east corner, so that we do not know if there was a projection on this side to match that on the south side. We would not necessarily expect the ‘baptistery’ to be matched with a similar chamber opposite. On the other hand it would be normal if the ground plan of the church was symmetrical, taking the shape of a Latin cross. Approximately two metres of a course of a wall higher than floor level extends towards the apse from the north division between the central nave and the aisle. There was probably an entrance beyond this from the north aisle to the area in front of the apse.
133. Sanctuary church. View west towards narthex.
The west wall of the narthex is complete. Two courses survive above a bed-rock foundation. The upper consists of reused socle blocks from the footing of the walls of the hellenistic temple laid above two smaller blocks. There is a space c. 1.20 m wide in the centre of the wall but no obvious sign that this was a doorway, and the stones of the two courses on either side are not flush with one another. The centrality of the opening, however, and the fact that rooms or structures may have been added later to the west wall make it conceivable that this was asmall entrance to the narthex, perhaps inserted at a later date than the original construction. Paving stones remain in situ across the threshold and inside the main entrance in the short north wall of the narthex, and there is ahollow, probably for fixing a doorpost, at this end of the west wall.
The narthex was 2.70 m wide. Two courses of the wall dividing it from the nave survive. The lowest visible course, made from single blocks laid flat, forms the top of the foundations and is at floor level. Above is a row of double blocks laid on end, with a total width of 95 cm. At each end of the wall, on a line with the divisions between the nave and the aisles, is a tall, upright, reused block. The width of the larger of these, at the south end, is 1.35 m, covering the same width as the inner row of stones from the upper of the two wall courses and protruding about 70 cm into the body of the church. The other block protrudes only 52 cm. They defined the western ends of the nave-aisle divisions and presumably acted as supports for architraves at the end of the internal colonnades.
The 95 cm-wide blocks which divide the nave from the aisles are mostly of a finer, whiter limestone than those of the solid exterior walls and were laid on a foundation of coarse white mortar and small stones. At the west end of the northern division, 1.45 m from the tall slab which protrudes from the west wall, there is a single surviving white marble block at a higher level, with six holes in the upper side, which were probably designed for attaching a circular column or column base, although none of the column bases or fragments found elsewhere on the site is large enough to be compatible with the dimensions implied by the holes. The block is moulded on the south side, but its use in the church was probably secondary. A similar moulded block was incorporated into the division on the opposite side of the nave, but at floor level as part of the foundations. The foundation course of the internal divisions of the church is complete except for two or three gaps, but there is no surviving trace of the internal colonnades.
The floor of the narthex was probably completely paved with stone. Two rows of paving remain at the north end, one across the doorway and the second adjacent to it, just inside the building. Another paving stone is still in situ in the south part against the outer wall. There is a further slab in the doorway to the nave and the stone extending across the end of the south aisle is also probably part of the pavement. A row of stones can be seen against the south half of this wall which extends into the church as far as the end of the tall, upright block, implying that the west end of the nave was paved. On the other hand numerous rectangular tesserae, measuring c. 2 x 1 x 1 cm, mostly of pink and black stone or of white marble, but including some grey and white marble and pale brown and grey/white mottled stone, were found in the area of the apse and the east part of the nave. Traces of pink mortar adhering to the stone show how they were fixed to the floor. One small piece of green glass mosaic was also found at the north end of the narthex. Fragments of white and grey marble revetment are scattered throughout the church as well as pieces of coloured glass, perhaps from windows, near the apse. There are also large quantities of tile from the roof.
At least twenty-two limestone blocks belonging to an interior frieze are scattered about the church. Their decoration, in raised relief, includes stylised vegetation (PLATE 134), fruit, tendrils and trees, ducks, partridges, a Latin cross (PLATE 135), a single-handled cup or jug, and a chalice. Clearly this was a major decorative feature, which probably surmounted the internal colonnades and certainly ran round the top of the apse, since several of the blocks are curved. Their total length was more than 15 m and three blocks which have designs on two adjoining faces, must have been set at corners. Two probably belong to the ends of the apse. A third seems to show that the frieze was carried round the end of the south aisle.
Two large rectangular limestone blocks (1. height 0.45, width 0.65; depth 0.55, cross height 0.36; 2. height 0.42, width 0.86; depth 0.61, cross height 0.50), each of them bearing a Latin cross with slightly splayed arms, were found, respectively in the apse area and to the north of the church (PLATE 136). Despite their differing dimensions they may originally have been placed symmetrically at each end of the apse. A simple pilaster base of white marble, rounded on one short side and squared off at the other (48 x 53 cm), is in the north aisle, and a plain double column base of white limestone, 85 cm long with a 5 cm lip at the bottom, lies to the north of the apse. East of the apse is a displaced font, with a deep, quadrifoil-shaped basin, which had been converted from one of the large blocks of the angular building (house 1, above pp. 74–5) on the slope to the south of the church (137).
134. Sanctuary church. Frieze block showing vegetation.
135. Sanctuary church. Frieze block with Latin cross.
136. Sanctuary church. Latin cross.
The reuse of so much material from other buildings, including the temple and the temenos of Mên, implies that the church was constructed after the demise of the sanctuary. This had still been in use in the third, although there is no unambiguous evidence that it continued to function in the fourth century. The church, however, was built not long afterwards. The custom-made frieze, and especially the partridge motifs, should be dated at the latest to the fifth century, and probably earlier as the moulding of the blocks has a Roman, not a Byzantine, flavour. The two large blocks with simple Latin crosses may also imply an early date.
137. Sanctuary church. Quadrifoil font.
On the slope above the church to the south are traces of building activity which were probably associated with it. Two delapidated courses of a wall lead up the hill on the same line as the wall at the east end of the church. There are possible traces of a parallel wall running from the west end with a mass of field stones between them. There is also evidence for construction adjoining the narthex. Ramsay was of the opinion that there had been a monastery attached to the church. The remaining walls are not substantial enough to support this sugges-tion, although it cannot be ruled out. The remoteness of the site would certainly have been suitable for an ascetic community. The spring to the south-west of the church was still damp in 1982 and could have provided a continuous and convenient supply of water in antiquity. On the other hand, an alternative hypothesis would be to see these poorly constructed walls as evidence for the inhabitants of Antioch taking refuge during the disturbances of the seventh and eighth centuries, when they might have converted or taken over some of the old sanctuary buildings to create a small settlement in this relatively safe and isolated spot. The building added to the church could have been part of such a programme, and of a non-ecclesiastical nature.
The central church at Antioch (Fig. 40)
Hardly any building on the site at Antioch has been more obscurely reported than the central church, which lay roughly opposite the Tiberia platea and west of the cardo maximus, behind the row of small shops excavated by Robinson’s team (see above p. 104). Arundell identified its remains but it received no further attention until the 1924 excavation, when some clearing took place. Robinson referred to it as a church in the shape of a Latin cross and published a photograph in his preliminary report (PLATE 138).1 Today the only readily identifiable part of the building is the apse, which is largely intact, but a ground plan of the 1924 excavation made by Woodbridge has been preserved, and offers the best basis for a descriptive account. This appears to show that, apart from the apse and the nave, much of the building was only preserved at the level of foundations, which included the substructure of the dais in the north of the apse and two foundation walls crossing the chancel area from east to west. The plan appears to confirm the description of the church as taking the form of a Latin cross, but caution is in order. A sentence in Robinson’s diary for 5 July 1924 says that ‘the walls of the transepts were cleared off so that we were able to get the entire plan of the church, which is that of a Latin cross’, but the published photograph shows that digging barely extended beyond the walls which appear on the plan. The excavators, in other words, rather than attempt to lay bare the whole structure, simply followed the lines of the walls they had identified. Thus we cannot be certain simply on the basis of Woodbridge’s drawing whether the apse was free standing or enclosed within a larger rectangular ground plan, whether there were aisles on either side of the nave, or whether there was a genuine narthex at the west end of the church. Examination of the area in 1982 suggested that the answer to all three of these questions should be positive, and that the church was accordingly both much larger and more orthodox in external appearance than was thought in 1924. A photograph taken in 1924 while excavation was in progress shows walls of ashlar masonry east of the apse which may belong to the church (139).
Fig. 40. The central city church (Woodbridge).
Some light is thrown on the question by a short article of W.M. Ramsay, in which he reported the discovery in 1927 of a large iron seal depicting and naming three Diocletianic martyrs and saints, Neon, Nikon, and Heliodorus. Ramsay claimed that the church in the centre of the city had been erected on the site of the synagogue where St Paul had preached. He described this, on the basis of excavations carried out in 1927 which had revealed more than the main apse visible in 1924, as having two apses within the boundary walls of the church, one in the usual place, the other, smaller in size, set back from it and merging on the south side with the south wall of the church. Its north wall joined the main apse but there was a door communicating with the corridor behind the latter and the east wall of the church. He took the smaller apse, which contained stone foundations, ‘for a bench and a desk looking obliquely towards the congregation’, to mark the position where Paul had preached.2 The theory itself is implausible, for the true position where Paul had preached would certainly, if identifiable, have formed the centrepiece of any church. Ramsay’s archaeological talents were so scanty that one may even wonder whether he had seen a subsidiary apse at all, but his claim that there were exterior walls to the church beyond the area excavated by Robinson accords with our own observations. It is to be hoped that the uncertainties might be resolved by further digging in the future.
138. The central city church viewed from the west (Kelsey archive 7. 1301).
In the light of the many uncertainties it is best to confine further description to a few vital statistics. Part of two courses of the main apse survive, the lower complete and made from fifteen blocks, one certainly reused, of varying widths and between 0.98 and 1.02 deep, which form a smooth curve inside and out. They rest on foundations of smaller, squared stones, rubble and white mortar. The second course, of which only four blocks remain, was aligned above the first on the inside but set back between 6.5 and 10 cm on the outside. The internal width of the apse was 8.275 and the depth from the front to the east face was 5.96 metres. A single block which we would attribute to the enclosing east wall remains in position, about 1.5 m behind the rear of the apse.
139. Excavated area east of the apse of the city church (Kelsey archive 7. 1302).
The north wall of the church was 34.76 m long from the north-west corner to the east wall of the transept, and this allows a total measurement from the outer face of the west wall to the outer face of the wall enclosing the apse of 42.35 m. The two blocks at the north-east corner of the transept are 8.65 m from the apse. If the church was symmetrical we can calculate that it was about 25.58 m wide. One block survives from the west wall of the southern arm of the transept, giving a transept width of 5.8 m, which is confirmed by Woodbridge’s drawing. It appears from the photograph and from Woodbridge’s plan that two stylobates ran from the apse across the arms of the transept.
The two sides of the nave were marked by foundation stones, presumably for stylobates. These show that it was narrower by the width of a block than the diameter of the apse, measuring about 5.8 m, and its narrowness may have helped to emphasise the wider central part of the Latin cross as the focal point of the church. There are traces of the foundations of the chancel screen at the head of the nave.
Seven adjoining blocks at the north end of the west wall define what appear to be the limits of the narthex, but the division between the narthex and the body of the church is uncertain.
The date of the church is a matter for conjecture. Ramsay suggested the fourth century and Robinson’s photograph shows a frieze block bearing palmettes on either side of a simple Latin cross positioned in the north-east corner of the south aisle, in line with the stylobate of the transept, which was probably from an internal cornice (Fig. 41). The palmettes can reasonably be dated in the fourth century.3 There is little evidence for other decoration. A fragment of an oval column of pavonazetto marble lies in the nave, and there were two broken pieces of limestone pilasters in the north aisle. Several tesserae of green glass were found in the body of the church.
Fig. 41. Frieze block from the city church.
The large basilica (church of St Paul)
The only one of the three churches so far identified at Antioch which has been discussed in any detail is the large basilica at the western edge of the settlement (PLATE 140). This was identified by Arundell, who even published a ground plan, is referred to by several subsequent visitors, and was excavated by Robinson’s team. In recent years it has been the object of renewed attention from Mehmet Taşhalan, who has worked on the building between 1985 and 1995, and has recently produced an interim report on the excavations.4 The most important contribution to the understanding of the church is an article by E. Kitzinger on the mosaic floor uncovered by Robinson.5
140. View of the basilica taken from the central city church (Kelsey archive 7. 1334).
The structural remains indicate a building of regular basilican plan with nave, two aisles, a narthex and what was apparently a large forecourt. The orientation of the apse is almost 45o south of east. The final phase of the building, represented by the visible remains, had a floor 60 cm above the mosaic, implying that the church was in use for a considerable period of time.6 Further, the longitudinal axis of the mosaic lay south of that of the basilica in its later phase and later walls were built over it, particularly at the west end of the nave and perhaps also in the south colonnade. The excavation reports offer no details of the earlier building but according to Woodbridge’s plan it was also a basilica, little if at all smaller than the later church. The east end of the mosaic, apparently complete, stops short of the later apse, and the later dividing wall between the nave and narthex has removed its west end, so we should imagine that the walls of the church were displaced some metres towards the east at the time of later reconstruction. The incompleteness of the westernmost panel of the mosaic may even imply that the basilica was six metres longer than the later nave (Fig. 42).
141. South side of the mosaic floor of the basilica viewed from the east (Kelsey archive 7. 348).
The design of the mosaic was predominantly abstract. The body of the nave is covered by a succession of four square panels of equal proportions, each composed of a different geometrical design and separated from the adjacent panel by a frame which also encloses the whole (PLATE 141).7 To the east of this unified length of mosaic is a different arrangement of smaller panels, two on each side of a narrow band of decoration laid along the longitudinal axis of the church. This central band measured about 7.65 m long by 1.37 m wide and there are signs that it was separated from the panels on either side by a parapet. Four inscriptions were worked into this section. One of those in the central band and the more westerly of the two panels to the north record the names of donors, Eutychianus and Idomeneus, a church reader, who are said to have fulfilled their vows to God when Optimus was bishop, presumably by contributing to the cost of the flooring. The other two, both in the central band, are quotations from Psalm 42. 4, Then I will go unto the altar of God, and Unto God who gladdens my youth. The positions of these two texts confirm that the central band of mosaic represented a passageway by which the altar was approached. Kitzinger suggests that this is the earliest dated example of a solea or liturgical corridor in a church building.8
Fig. 42. The large basilica (Woodbridge).
The mosaic is dated by the reference to bishop Optimus, who represented Antioch at the Council of Constantinople in 381 and was also a correspondent of Basil of Caesarea in the preceding decade. Kitzinger reasonably assumes that the whole floor, being of unified design, was laid at the same time. It provides a valuable terminus ante quem for the construction of the building. It is reasonable to assume that the mosaic dates to the foundation of the church, whose simple lay-out would not be out of place in the late fourth century. This was the period when monumental urban churches first became common in Asia Minor, but in fact the Antioch basilica is one of only two dated churches in Asia Minor belonging to the fourth century.9 There is, quite simply, no clear evidence at all for the date of the later phases of the church, when it was enlarged and the apse was displaced to the east. Although the Yalvaç Museum contains a number of pieces of church ornament which may belong to the middle Byzantine period, none of these is linked to the basilica on the site of Antioch. The continued use of monumental ashlar masonry points rather to the fifth and sixth centuries.
The ground plan and dimensions of the church are not entirely clear. Woodbridge’s plan is published without a scale, while Arundell, who also published a simplified plan without indicating a scale, said that the building was 160’ long, excluding the portico, 80’ wide, with an apse 40’ wide. The north wall, complete at both ends, is 43.10 m long, and the distance to the internal face of the apse is just over 50 m, a little longer than Arundell’s calculation of 160’. Before Mehmet Taşhalan’s recent excavation a single course of very large rectangular blocks was visible above ground, laid on edge. Their average height is 1.10 m and they vary in length from 1.6 to 1.83 m and in depth from 65 to 85 cm. The excavation photographs show that the wall was originally two blocks thick, the inner line, apart from a single block towards the west end, having been removed since 1924. Both Arundell’s plan and his account of the basilica record two doorways in the north wall. Only two blocks of the south wall were visible in 1982/3, of the same grey limestone and of similar dimensions to those of the north wall. Two adjacent blocks survive to mark the south-east corner of the south aisle.
The apse is also composed of grey limestone blocks, one of which was obviously reused. It too was originally two blocks thick, with a thickness of between 1.45 and 1.71 metres. The inner face was circular, the outer polygonal with six edges (PLATE 142). The internal diameter of the apse is 10.82 m and the depth to the outer face is 9.23 m. Two protrusions, 43 cm long on either side of the apse, might have been supports for a screen separating it from the nave.
The nave was divided from the aisles on either side by a stylobate, originally surmounted by a colonnade. The stylobate blocks are of creamy white limestone and of roughly uniform shape and size. Three blocks of the north stylobate are in place. Two are adjoining and bear builders’ drafting marks to assist in the correct positioning of columns, giving interaxial spacing of 3.33 m.10 As the total length of the stylobate was 44.8 m, this would allow thirteen columns and two pilasters or half-columns at the end, as suggested on Woodbridge’s plan. The two adjoining stylobate blocks in the north row are 2.63 and 2.66 m long. The remaining length of the single block to the west of these is 2.02, and an example on the south side, also marked to receive a column, measures 2.70 m. On Woodbridge’s plan each stylobate comprised seventeen blocks, a calculation based on an average length of 2.64 m for each. The sides of the blocks bear markings at regular 5 cm intervals and the width of the upper side of each block is just short of one metre. This suggests that they may initially have been intended for use as architrave sections, each to be divided in two to produce a width of 50 cm.
The stylobate was laid on a bed of white mortared rubble, but the slight incline of the natural floor-level towards the south was compensated by the insertion of grey limestone blocks underneath this layer of stone and mortar.
The west end of the basilica is represented by two blocks in situ at the end of the north aisle. Arundell noted three doors in this wall, measuring 10', 15', and 10' respectively. Only that in the north aisle is still discernible, defined by two blocks 2.76 m apart. The north part of the threshold stone, with a squared hole for a door-post, also remains in its original place, and from this the actual width of the door itself can be estimated at a little over two metres. The door opened inwards into the aisle. Arundell remarked that the north and south sides of the basilica extended 55' beyond the narthex to form a portico. Nothing survives to confirm this. The colonnade shown on Woodbridge’s plan is conjectural and suggests that the details of the outer wall of the narthex were as obscure in 1924 as they were in 1982. The excavation diary for 11 August 1924 notes that during the exhumation of the continuation of the north wall several blocks were un-earthed, but only one seemed in place.
142. Basilica apse (Tasçlialan).
The proposal to identify the basilica as the church of St Paul has been made by Mehmet Taşhalan. The evidence for this is a large font, cut down from an altar of Roman date, now in the Yalvaç Museum (PLATE 143). It carries a roughly incised text, later than the sixth century AD, which reads (on top) [φ]ωνήK(υρίο)υ έπη τον ύδά[των]. (shaft, front) Ί(ησού)β X[ρίστο]ς Nήκα. (left side) ό αγηος Παυλος (right side) ό αγηος Γρηγόρηος.11 In 1914 the text was seen in a bath house near the main market area of Yalvaç, but Taşhalan was informed that it had originally been found at the site of the basilica. If this report is correct, the inscription at least shows that Paul’s association with Antioch was commemorated in the cult at the basilica, although it does not supply definitive evidence that the church was dedicated to him.
Since the above description of the large basilica, based on observations made in 1982/83, was compiled, the situation of the church has been radically transformed by excavations which have now been published. The plan of the church in its secondary (supposedly fifth-sixth century) phase is now revealed in more detail (Fig. 43). At the west end there was a double narthex, 27 m x 13.5 m, divided by an internal colonnade. There was an exterior door in the middle of the west wall and three doors, measuring respectively 2.8, 3.5 and 2.75 m, which connected with the body of the church. Clearance of the naves and aisles confirmed that these were separated by two rows of thirteen columns, resting on hexagonal bases. There were two doors in the south wall of the basilica, and three in the north, including a central main entrance 4 m wide, which led out to a paved courtyard, surrounded on the north and west by an L-shaped portico, largely constructed from spolia. Other structures adjoined the south side of the church, including a large rectangular cistern. While the new excavations have considerably clarified the plan of the church, no fresh evidence has come to light for the date of the construction periods.12
143. Font with inscription for St Paul (Yalvaç Museum).
Fig. 43. The large basilica (Tasçlialan).
There can be no doubt that the basilica was the seat of the metropolitan bishop at Antioch, the most senior ecclesiastical official in the province of Pisidia. It is larger than any other church yet recorded in Pisidia, including the two big Latincross basilicas at Sagalassus, and is one of the largest in early Christian Asia Minor, comparable in size with the biggest churches at Anabarzus and Corycus in Cilicia, and exceeded in that region only by the great basilica of Thecla at Meryemlik, built in the later fifth century.13 It thus confirms the conclusion to be drawn from other evidence, including the enlargement of the theatre, the building of a new agora and the construction of the city walls, that the population of Antioch was greater in the fourth century than at any earlier period of its history.
Notes
1 AJA 28 (1924) 443, 437 Fig. 1. The same photograph is also reproduced in The Michigan Alumnus 6 June 1925, 702.
2 W.M. Ramsay, ‘Anatolica Quaedam III. Neon, Nikon, and Heliodorus’, JHS 48 (1928) 51–3.
3 In the museum there is a column capital decorated with rather similar open palmettes on each of its four sides. It might have come from the church.
4 Taşhalan, 1997.
5 E. Kitzinger, ‘A fourth century mosaic floor in Pisidian Antioch’, Mélanges Mansel I (1974) 385–95. The inscriptions from the floor mosaic were published by Robinson, TAPA 57 (1926) 234 ff. with pl. 39 f.
6 Robinson in fact remarked that the basilica had been reconstructed at least three times, which may well be a correct inference from what could be observed during the excavation, although the evidence of the plan implies no more than two phases.
7 Two details are illustrated in Kitzinger’s article from Woodbridge’s drawings, figs. 60 and 61.
8 Mélanges Mansel, 394 n. 32.
9 Mitchell, Anatolia II, 67. The observation is owed to Dr H. Hellenkemper. The other precisely dated fourth century church in Asia Minor is that of St Babylas at Daphne (Syrian Antioch).
10 Kitzinger observed that two of the columns in the north stylobate seemed to have been in situ. In the photographs of 1924 a six-sided column base can be seen in the north aisle, east of the adjoining blocks.
11 Published by W.M. Calder, JRS 2 (1912) 98 no. 29 (incomplete), and B. Pace, ASAA 3 (1916–21) 55 no. 43. Height 0.85, width and depth of top 0.84, width and depth of shaft 0.68; letters very variable, 0.02–0.03. The corners of the shaft have been cut in a rough concave manner, so as to give a roughly octagonal shape. Inscriptions and reliefs have been incised or shallowly cut on to the four main faces of the shaft. On the font there is a relief of two winged beasts on either side of a tall cross, with a Maltese cross top left and a rosette top right. On the left there is a Latin cross. On the right there is a cross above aglobe inside a recessed arched niche with pilasters on either side. On the reverse, which is uninscribed, there is another Maltese cross.
12 Taşhalan 1997.
13 See Stephen Hill, The Early Byzantine Churches of Cilicia and Isauria (1996) for plans and the most recent discussion of these.