The Society For the Promotion of Wellenic Studies SUPPLEMENTARY PAPERS None ol: ECCLESIASTICAL SITES ISAURIA (CILICIA TRACHEA) BY ARTHUR. C. HEADLAM Frtiow. or ALL Sours’ Contecr, OxrorpD PUBLISHED BY THE COUNCIL, AND SOLD ON THEIR BEHALF BY MACMILLAN. AND CO., 29, BEDFORD STREET, STRAND, LONDON The Rights of Translation and Reproduction are Reserved MDCCCXCII DF cena % Mowe & —— Sf a 22 ECCLESIASTICAL SITES ISAURIA (CILICIA TRACHEA) The Society For the Promotion of Hellenic Studies SUPPLEMENTARY PAPERS No. 1 ECCLESIASTICAL SITES ISAURIA (CILICIA TRACHEA) BY ARTHUR C. HEADLAM Fetntow or Aut Souns’ Conuecr, Oxrorp PUBLISHED BY THE COUNCIL, AND SOLD ON THEIR BEHALF BY MACMILLAN AND CO., 29, BEDFORD STREET, STRAND, LONDON The Rights of Vranslation and Reproduction ave Reserved MDCCCXCIII AnD Sons, Limrren, ICHARD CLAY R GAY. NG BU AND LONDON CONTENTS PAGE ECCLESIASTICAL SITES IN ISAURIA— I—KOJA KALESSI . 9 Il—KESTEL, OR DA BAZAR. 20 I1I.—NOTES ON OTHER ECCLESIASTICAL REMAINS IN ASIA MINOR. 21 INSCRIPTIONS FROM ISAURIA 22 LIST OF ILLUS RATIONS Prate I, Fie. 1. A Generat Puan oF THE Site. Fie. 2. A Puan or tHe Cuurcu. Pratre II, Fre. 1. Lonerrupran Secrion or tHe Caurcr. Fie. Wesr Exrvation or tHE CHuRcH. Fig. 3. Cross Section or tHe Cau OH. FIGURES IN THE TEXT PAGE Fig. 1. View oF Doorway Marxep 0. on Phan . : 3 : a A 8 10 Fie. 2. Derain or Centra Parr or Linre, or rue Same Doorway . 3 3 11 Fie. 3. View or tHE West Exp or THE CuuRCcH . 4 3 : 3 2 11 Fic. 4. Derar, or Sour Doorway In West Watt. 4 3 A 13 Fic, 5. Deram or ORNAMENT oN THE LininG or THE SAME Door é : é f 4 A ‘ a 14 Fic. 6. Doorway at BaaLBec P : é fi p 5 A 5 : : : 3 e 3 15 BIGs fee ALN 10R Virw 1N Cuurca Looxinc Wexst é 2 é 5 16 Fie. 8. A Derare or One or tHE CarrraL igi Fic. 9 8.E. View or tae Cuurcm at Kusren. 5 4 4 2 20 Fic. 10. Puan or THe Cuurcn ar Kusren : 3 : F é 5 5 2 : fe 21 The Plates were drawn by Mr. Scuunrz and Mr. Barystey from our Plans and Photographs. Figs. 1, 3, , 7, 9, are reproductions of photographs. Figs. 2 and 8 were drawn by Mr. W. R. Leruany. Fig. 6 is from a drawing by Mr. R. Pa 5 SPre ESE STE J-H.S. SUPPL.II (1892) PL.I. * KOJA * KALESSI + - A> GENERAL - PLAN - OF -THE-SITE- ' SIE OO (ee EON PONE STEN ore + HIGH: ROCKY + PLATEAU + age ese SS Ate ‘ NY nate "my, is ing ee ei bay PMR i E Ine, Mts wn eee y : a “ny “Wy Peers: ITD ibiapin wi 7 gy Mii. Mages "Ny tere 1 ret, ae can eh ii ui Mt laggy tnt Mj i inl, ST tl ley N : 1), as itis aes exe E- ENTRANCE: cea i LTT HE Tr i ipsa i Ty TA ic aan Tia agama as 43 ( va ul ih “i iil a i eT Tec eee na oat HC mi in aE a ne 3a \ sf i" = 2 SM | fee ans or i! ml + STEEP. DESCENT: oTHien') 6 “FIG: 2- o/\ OTL © OIF 6 Tel 9 CinlUintefal > S[2fEEu Trovettret a __fe ie CP | PMeTRES f KELL PHOTO-LITHO.8 FURNIVAL ST HOLAORN EC CHURCH AT KOJA KALESSI, (ISAURIA) ee ED A IIE hy EIS Ti ESE = | } | i | CHURCH AT KOJA KALESSI, (ISAURIA.) Se eT? ECCLESIASTICAL SITES IN ISAURIA (CILICIA TRACHEA). | Tux purpose of this paper is to describe certain sites in the ancient Byzantine province of Isauria or Cilicia Trachea, the interest of which is mainly ecclesiastical. They were visited by a party consisting of Prof, W. H. Ramsay of Aberdeen, Mr. D. G. Hogarth of Magdalen College, Oxford, and myself during July, 1890. § 1.—Kosa Kae Koja Kalessi (called by Laborde Aladja Khan) is situated about five hours to the north of the town of Mut—the ancient Claud iopolis—high up on the north-eastern slope of the Calycadnus valley. It was visited in 1826 by the French traveller Laborde, who describes it in his Voyage en Orient. His remarks are interesting and his drawings ia ribed of very considerable artistic merit, but inaccurate anc untrustworthy. We spent part of Ju and 8, 1890, at the ruins, and it is the result of this visit which will be d in the following paper. It was not until some months after our retum that we found Laborde’s description ; until then we imagined we had been the first visitors to the site. route to Karaman for 4% hours. After passing over a low ridge the river Pirin] Su, which descends from Kestel, is crossed by a bridge about an hour from Mut. Then comes a low plain, then the village of Yapanli, and then and, At 24 The road from Mut follows the most direct the road begins to as 1ours from Mut a little to the left of the road is a spring called Ilidja, ‘warm,’ for though the water seems cool on a hot day, in cold weather it is discovered to be warm in temperature, and it has a slightly medicinal character. At 44 hours is a built fountain and a group of ruined buildings, including apparently a ruined khan and some other remains of an architectural character. Laborde’s descriptions of these buildings seem exaggerated, or they have been much destroyed since his time.? From here the ruins of Koja Kalessi are seen on the side of the mountain above, and a steep climb of half an hour up the hill side is required to reach them. Near here also must be some ruins which were reported to us as existing in the valley towards the village Maliya, half an hour to the north-west ; the native who had promised to guide us to them failed. The ruins of Koja Kalessi are situated about 3000 feet above the Calycadnus valley and 4000 feet above the sea, facing south and south-west and looking over the junction of the two great river valleys through which the two branches of the ancient Ca yeadnus run. They consist of a large monastery with buildings of various characters too much destroyed to be easily identified, and a church in very good preservation, They are built on a terr running nearly due e partly cut out of the side of the hill, with the ground falling away very steeply below and rising almost precipitously above. Just outside the entrance, which is at the west, and on the left hand, is a large cave, (see the general plan Plate I.), and immediately beyond this comes one of the gates of the monastery, of which there appear to have been several. The doorway marked O in the plan, ‘ace ast and west, 1 Laborde, Voyage en Orient, pp. 124, 126. Planches, ddifice, qui sera notre khan, et qui était une église. A 68, 69. quelle époque attribuer ces constructions solides, imposantes 2 Tb. p. 124. ‘En effet, non loin de 14 nous trouvons un _ par leur ensemble?’ monument qui doit avoir été une chapelle, et un vaste B 10 ECCLESIASTICAL SITES IN ISAURIA. of which a reproduction from a photograph is given (Fig. 1), is carved. On the inside of each of the gatepos still standing and is richly are two winged figures which seem to be undoubtedly the archangels Michacl and Gabriel; above in the lintel is a head of the Saviour surrounded by a circle and borne by six-winged figures. This is the (avroxpdtwp and the fieures on either side are the éfamrépuya, and it represents the usual ornament of the doors of Greek churches and monasteries.” Fic. 1.—Vriew or Doorway Marken 0 on Puan. This is certainly the right explanation of these carving: Christian, but they show a very marked resemblance to pre-Christian designs. work may be which he copied at Koni Christianity of pre-Chris and another example is given by Fellows. n designs is more marked in and they are undoubtedly In Laborde’s na very similar design representing two-winged figures bearing the sundisk The adoption by than in lat this site er buildings. The influence of the spirit of Christianity has not had time to make itself fully felt, CA cbté est un porte de six pieds de haut sur quatre de large, et le linteau richesse de conception, une abondance de détail et une délicatesse d’exécution vraiment surprenantes. montants se dressent deux a sur les tétes de deux vieill linteau est orné d’un ba: ailes étendues, ayant a ses cétés des aigles qui combattent, 3 Laborde’s description is fuller, ib. p. 124. dont les montants sont ornés avec une Sur les re} 3 ils reposent leurs pieds ards figurés en buste. Le s-relief qui représente un ange aux ici un lion, 14 un taureau: aux deux extrémités deux figures vénérables dhommes Ags sont séparées du reste de la composition par un arbre. pagnent ces bas-reliefs sont empreints des souvenirs encore Les ornements qui accom- assez purs de l’antiquité, et on les a répandus avec une profusion qui n’est ¢galée que par leur bonne exécution,’ And Revue Archéologique, iv. 1847, p. 174. ‘Des anges avec six ailes déployés, soutenant dans un médaillon Vimage du Christ.’ ECCLESIASTICAL SITES IN ISAURIA. 11 unchanged. This seems corroborative evidence of the and the old designs are still almos early character of the site we are de ibing.* , Tnside the gateway are the ruins of what was apparently a large reception hall of basilica form with an apsidal termination, Round it ran a gallery with a long row of windows on the south side, which perhaps marked a division into small cells. The remaining buildings are partly too much dilapidated, partly too much blocked by brushwood, for their purpose to be apparent —Derrarm or Cenrrat Parr or Livre, or tHe Same Doorway. ring the ground. Along the south side of the monastic To without excavation or at least cles enclosure ran a wall surmounted with columns on a balustrade with niches and a shrine. the north the monastery was protected by the steep slope of the hill, in which were tombs cut in the rock, some few with inscriptions. The whole length of the enclosure was nearly 600 feet, the breadth about 70 feet. te LS Rk = eres ES Fic. 3.—Vmw or tHe West Exp or tHe Caurcu av Kosa Kauesst. angular building At the extreme east end of the monastery is situated the church, a rect vation. th by 45 feet in breadth, the ruins of which are in very good preser about 70 feet in leng drawings, has kindly communicated the following notes. ‘The figures that uphold the central garland and medallion are six-winged, and of a kind—mostly wings and little 4 Fellows, Lycia, p. 224. In the small edition (1852), p. 377. Mr. W. RB. Lethaby, to whom I am indebted for several ECCLESIASTICAL SITES IN ISAURIA. The roof, which was probably of wood, has been destroyed, but all the stone work and the stone I ) , Nees vaulting over the apse still remain. In front of the west end was probably an open court similar to that described by Eusebius, colonnade traceable. with a instruction might advance.’ or cloister at any rate on the south side, where the remains are distinctly c HE c! 5 p * if ron * (x sro This, Eusebius tells us, was the furthest point to which those without any Christian Immediately in front of the church was a portico, the construction of which it is difficult to determine. By looking at the v built into the wall. probably inserted. that case it may be suggested southern end there was a staircase situated there. represents ta évSdtata mpomvda Mr. Schultz has pointec galleries in a monastery church the north side, which is at a higher level. There is one more point which is sugges be seen that a wall ran up to addition. Moreover there is a doorw: directly to the church and its disturb the seclusion of the monks. crowds from the districts round. would the capitals, and the niche, now half-buriec earlier body—that recalls those of the pendentives of S, Sophia. The idea seems to be derived from Isaiah vi. 2. ‘The angels on the jambs of the doorpost are almost certainly Michael and Gabriel—see De Vogué’s La Syrie Centrale, p. 90, compare also the figures which Wood found on the jambs of the so-called tomb of §. Luke at Ephesus (Wood’s Hphesus, p. 57), which would also be these guardian angels. Laborde (op. cit. p. 124) says they stand on busts. One of these is seen rather indistinctly in the reproduction from a photograph of the door (Fig. 1). These two busts with five on the posts and lintel make seven, and it is tempting to suggest that they are the planets, Michael and Gabriel standing on their own stars. The planets were often represented in this manner (see Wood’s Palmyra, plate xix., where we have on a ceiling busts representing the seven planets surrounded by the signs of the Zodiac). The position of Gabriel and Michael is that which they elevation (Pl. IL Fig. 2) it the doors in the west end is a line of four capitals, and at the These two indications seem to suggest tha Whatever the structure may have in Greece, that of St. Luke in Phocis near Helicon. to this gallery and the galleries of the church was probably also obtained So access is obtained to the galleri church of St. Luke from a corridor in the monastery, to which the cells open. gested by this portico. By one of the posts of the central doorway, and portion of the portico would thus be separated from the rest. opening to the outside of the monastery. this arrangement may have been to admit women, and others not connectec galleries without obliging them to go through the monastery and The church was probably a place of pi grimage, and sanctifiec by some memories which are now lost, and on certain days in the year would be visited by large To admit these into the church and keep e the object of these alterations in the portico. The question has been raised as to how far the portico was a later addition. the walls last mentioned were not part of the orig sort of portico and gallery outside the west end of the church w holes for rafters mentioned above are undoubtedly part of the ori structure and there is no sign that they were added later. The entrance to the church is gained by three doorways ; on the right of each is a in the débris that has fallen. ore-Christian custom, examples of which may be will be seen that a little above south corner a square projection Presumably these must have been intended as a support to some floor or roof, while higher up, above the great windows, are a row of holes into which rafters were the portico was of two stories. In hat the upper story was a gallery, roofed above but open, giving access to the internal galleries (see below) by means of The ground outside the church is considerably higher than that inside much as is shown in the plan—and suggests the ruins of a of the portico are some remains which might the two openings into the aisle. although not quite so considerable structure; while at the for ye accountec ry supposing that een, this portico of Eusebius,’ the mpovaos or @Ew-vapOn£ of a later church. out to me the existence of a somewhat simil: arrangement of Entrance rom the rock on in the monastery plan it wil shat the southern looking at the This wall was probably a later The object of with the community, he monastery secluded Tt is clear that ginal design, but it is equally clear that some intended from the beginning, for nal sarved These niches again are the reproduction of an een in some of the drawings in De Vogué.” occupy in Byzantine iconography, viz.: on either side of the entrance as guardians of the door,’ Mr. Lethaby goes on to suggest that the central bust, that of Christ, is treated as the sun in pre-Christian buildi: ‘The association of the sun and moon with the door was quite universal, compare the temple of Baal Zamin (De Vogué) which had a bust of the sun over the doorway.’ 5 Kus. H. #. X. iv. 39 (ed. Heinichen), eiow 8% maped- Déver mudGv, odk BOs ediixev dvdyvors Kal aver rots Tool téy %ov eriBaiver dyiwv, SiaraBdy 8% mdeiorov baov 7d peradh tod te veb Kad av xpirov ciodduv rérrapor piv répié eyxapoios Karexdopnoe oroais, cis rerpdyovsy m1 oyipa Tepuppdéas rdv rérov...pérov albpioy ibier els ryy T0% ofpavod xdroww. Olt, Save AL, * De Vogué, La Syrie Centrale, pl. ECCLESIASTICAL SITES IN ISAURIA. 13 Of the three doorways, that to the north is plain; that in the centre, which is larger than the others, is richly carved with traditional and palmette ornament. That to the south is still richer (Fig. 4). Outside, on the doorposts and lintel, the ornament consists of vine-leaves with grapes and birds; inside, between the posts, of fishes arranged in a very striking and effective pattern Three such doors are specially mentioned by Eusebius in his d r early church.’ He states that the centre was always larger than the other two. ption of an yAY IN West WALL. The ornament of this south doorway is full of interest, and reflects the character of the site. Fishes are, of course, a well-known Christian when, partly from a sense of rey definite representations of Ch mbol, but a Christian symbol of the early centuries, rom a desire for secrecy, the Church abstained from jan ideas. The vine-leaf, the grapes, and the birds plucking the grapes, are all conventional pre-Christian ornaments, but adopted here, as in the catacombs and in the Roman mosaics, as admirably representing in a symbolical manner Christian teaching. One of the great doors in the temple at Baalbec,” ornament figured by De Vogué,” and doors in erence, partly 8 Eus. H. #. X. iv. 41, adda yap Kai tiv tovtwy Oday a drawing that he made of the temple doorway at Baalbec rapapeupdpevos, theloow et paAdov Trois evdotdtw mporthos (Fig. 6). He also points out to me how common the vine tas énl rov ved wapddous dvarertapevas erole, tro pev Tals Wi a Jewish ornament. ‘On the tombs of the kings HAtou Bodais abfis tpeis ridas bp’ &v KaraHels TAevpdv, Gv 76 and on the tomb of Jehoshaphat, both ascribed to the Toy Tas Tap Exdrepa peydbe te Kai ware. wheoverreiv 7H Herodian period, and in other tombs near Jerusalem, the péon Xapirdpevos, Taparyypact te xahkod oidypodérots kal vine and grapes are carved, and in the two first named mouiApacw dvayhigos diadhepovrws airy gadpivas dodv examples the style and character of the work suggests its Bac. rary Tovs Sopupdpovs trélevée. having been executed by Greek artists.’ ® R. Wood, Ruins of Baalbec, plate xxxii. London, 10 De Vogué, op. cit. Pl. 3. 1757. Mr. R. Phené Spiers has kindly put at my disposal c \ ; , SE Ne REED —- 14 EC Diocletian’s palace at Spalatro," may be cited as doorways be seen that although the design is the same the treatment is different; the later work is free bolder, if less refined, and beginning to break Byzantine rather than Roman, or perhaps Roman Fic. The church itself consists of a nave, t apse, and side chambers opening ito the aisles tower. The following points seem to be of s (1) The first two bays of the nave, as unde a low D one another and from the supporting the roof. These arches are interes the church are with those of northern Syria, bs of stone su square Space churches are roofed with large sla unlike northern Syria, was a well the arches crossing the nave and taking the pl preserved and are a sign of the district from But the division between t in the columns are marks which im he two bays o from one another and from the space under the far we have been able to illustrate, we shall find ; besides the la never to enter into the church, there were two di to whom different places were assigned within adapted to these divisions.” (2) One of the most stri that this much seems evident, tha’ columns supporting arches which divide the nave from the is a second somewhat resembling the triforium bolder and more effective.“ This probably impli 1 R. Adam, Ruins of the Palace of the Emperor Dio- cletian at Spalatro, 1764, Pl. 45. 22 De Vogué, op. cit. Pl. 7. 3 Hus. H. #. X dflav 708 mavtds Aaod Sufper tas Svvdpets (1) ols pay rv eéwOer . iv. 63, dixaoovvn Sfra oixoSopiv Kar’ aitd pdvoy mrepibpdrrov mepiBodor,...(2) ofs 8& ras ext tov zéwbev (4) rods 8° jon dudi tov BaotAcov olkov éxarépwler rapaletyruow, ere otkoy émirpéruv eto dSov 3) dddws 88 mpdros ois Gppt tiv addiy é terpaydvov kloow trearipite LESIASTICAL SITES IN wo aisles, a bema or - ‘ing as showing that the architectural fioured and described by De Vogué.” pported on arches crossing the nave. -wooded count rply that two screens ran across the church, separating ISAURIA. instances of the type from which these Christian and their ornament have been developed. But if these be compared together, it will and The work is away from the early conventions. passing into Byzantine. sanctuary ending in an and sanctuary. In the centre of the building is ecial interest :— also of the aisles, are divided the central tower by arches which tructurally from ted im affinities of There the But Taurus, A stone roof therefore was unnecessary, but s were still of the beams of the older basilica which the tradition of the church came. 1es. Below the bays whom so the nave was not confined to these are central tower. If we turn to Eusebius, that the language he uses is by no means clear, but rge multitude who on ritual grounds were fferent classes, the ‘ audientes’ and the ‘ substrati,’ it. This church then shows signs of having been supposed king features of the architecture of the church is the double tier of sles. Above the main row of pillars arches of an English cathedral, but very much es that there was a gallery of wood built over the pev karnxovpévors kal év avéy kal mpoxoTy Kabeotaras, ob py Tei cow Kel agp as aay Gordian Pamalas a quar SteCevypevovs. 1 Cf. Laborde’s description, op. cit. p. 126, ‘La srieur galerie ornée de colonnes et percée de fenétres, donne a cet intérieur un air grandiose et aérien qui présente a Varchéologue de l’ornementation et la grandeur vraiment antique de l'appareil, un spectacle nouveau et séduisant.’ ECCLESIASTICAL SITES IN ISAURIA. 15 aisles. This is corroborated by the fact that the main windows in the south wall are very high up and are adapted for lighting the gallery and not the aisles below. Entrance to these gal eries could be gained either from the top of the rock which forms the north wall or by a staircase in the portico, as is explained above. Such galleries appear to have been almost universal in early churches and were generally used for women. They are called the tmepéa and are still yery common in the Hast.” } (3) The aisles are divided below from the nave by a low stone wall which may have served for seats, and by a screen, as may be judged by the marks in the columns. The aisles themselves terminate in a small apse or recess through which was an opening into the chamber beyond. A somewhat similar termination to the aisles, with a doorway into chambers beyond, may be seen in the fifth century church of the Red Monastery, near Sohag in Upper Egypt. ; 2 He oF SS a Fopss-del: Fic. 6.—Doorway av Baapec. (4) The nave properly so called was formed by the space under the central tower, and is approximately square. This corresponds with the description of a Christian church given in a letter of the Emperors Valentinus and Theodosius, who speak of the three parts of the church, the holy @vovacrrpsa, the ederiptov tod daod, which is square, and the space beyond these to the outer doorway of the church.” ; The tower is supported by arches resting on four large columns, the capitals of which are elaborately carved with birds worked into the foliage ; one of these columns has fallen. It was 15 Of. Greg. Naz. Carm. de Templ. Anast. i. 1, xvi. 20, 16 Labb, Cone. t. 3, p. 1236, éxt yap tév jperépov Katpav pp. 844-845. of pdvov 72 Bela Bvovacripia Kal 7d eberiiptoy 700 Aaod 73 rerpd- ai 8° dp! da’ Symp\av reyeov ebxoopov dxov} Gyvad rapbevixal KNivov dip? éxOoydpors. wvov rolywv epi Boy rexeopevoy cis daopdderay owredely Tov Xx! p 1) TECXEG mpooherysyrov GeoriLouer, add’ el Tu Kal mepatrepov Tovrov avyxdvet, dxpt Tov Tedevtalwv GupBv Tis ekkAyotas. 16 ECCLESIASTICAL SITES IN ISAURTA. one story higher than the body of the church and was probably surmounted by a low-pitched roof of wood; the angles of the corners were rounded above, and the roof was partly supported by small columns springing from very elaborately carved brackets. In the south- east angle the design was of bunches of grapes, in the south-west of rams’ heads, in the north-west a basket containing fruit. In the north-east angle the ornament is more conventional. The question has been raised, Was the tower surmounted by a pitched roof as described above or by adome? It has been suggested in favour of the latter alternative that the transition from the quadrilateral to the octagonal form in the upper part of the tower would naturally seem to be the halfway step toa dome. Many buildings in Anatolia at the present day, especially Turkish baths, Fic. 7.—Intertor View 1x Cuurcn Looxine Wesv. lave a square tower from which rises an octagonal base, and from this again a dome. But the angles are not sufficiently cut off to make the upper part of the dome octagonal. The splayed angles are not structural pendentives leading up toa dome. The upper part of the tower is and appears square but with the corners slightly rounded, and not octagonal. Against the dome the ‘ollowing two points seem to be conclusive. There is no sign of débris in the church sufficient to 1ave formed a dome, and the upper courses of masonry remain intact ; if there had been a dome above which had fallen in it must have carried with it some of the masonry of the tower. (5) Another feature of interest is the horseshoe character of the arches, This appears to iave been a local peculiarity, and examples may be seen in the drawings of Maden Shehr in Lycaonia made by Laborde (Pl. 66). ECCLESIASTICAL SITES IN ISAURIA. 17 (6) The chancel and apse do not call for much remark. The latter is entirely internal, as is he case in the majority of the oldest known churches, for example in the very old ones in Africa j Reparatus near Orleansville, in those of the Red and White Monasteries in Upper Egypt, which certainly date from the fifth century, and in the oldest among the churches of old Cairo » but ic of later churches—the three chapels. In the. present church the place of the two side chapels is taken by chambers such as are found also in the Red and White Monasteries, in a small church the ground plan of which exists among the ruins of the island of Philae, and in some of those figured by De Vogué. At the extreme east of the apse was a raised stone stand for the bishop’s chair, while all round ran a stone ledge marking the seats of the oresbyters. The north wall of the church was formed by the rock against which it was built. Taking the church as a whole it seems to represent an almost unique design. At first sight > might be judged to be a basilica of the older type, but it differs from this by the presence of the tower. It has not the continuous row of equal columns which marks the ordinary basilica. But it is equally removed from the later Byzantine type of church, and seems to represent.a transition 2 ct these last have a characteris acd is ca RS ee | | ie Fic. 8.—A Derarn or One or THE CaPrrats. from the older oblong basilica to the later Byzantine church with the central dome. The ornament is equally interesting—it is very rich, very well preserved, and sculptured with boldness and freedom, although perhaps wanting in delicacy. With doorw: here is nothing which can be called distinc ively Christian, the vine and the fish, although rapidly adopted by Christianity, being of course found as pre-Christian designs. The only ‘ormed Byzantine type is the capitals of the h might, but need not necessarily, be as he exception of a cross on the central + portion of the monument which seems to represent a columns, which have all the characteristics of work whic! late as the time of Justinian.” The beautiful character of the masonry throughout this building ought also to be specially The regular courses of squared and closely fitted stones which are built without mortar as noted. pen joint is left of about two inches in good Greek times ; the manner in which over every lintel an o deep and the full width of the opening below in order to keep all direct weight off the centre of the stone and so prevent it from couching ; the cutting and fitting of the arch stones ; the pillars and capitals all show careful and excellent craftsmanship. 17 On these capitals compare Laborde, p. 126, ‘ornés de * Nous avons dans ces montagnes l’exemple le plus frappant chapiteaux d’un grand caractére et dont le dessin marque d’une fusion secrete de V/architecture antique avec l’archi- une fusion de reminiscences antiques et de style byzantine.’ , de And on the general character of the architecture, /.c., tecture orientale et byzantine, et le lien, caché auss celle-ci avec V’architecture arabe.’ D We have next to re (1) We have first a spot marked in the plan It had beer the of that name. asserted that he—the son of in the consulship of Gada’ aiph 5 help us, but un: ortunately it may be the fourteenth. the second in the fourteenth. of this paper that the balance probability Tarasis first became the word zapapovapios the fact monastery on t. prove that this makes it proba ier one. (2) Our second source o ear. cal ? ing this monastery Apadna some of the buildings may we is no other ecclesiastical site in If this site be Apadna, t to the time of Justinian a chure to what extent we do not know. of the monastery to a period as KCCLESIASTICAL SITES IN ISAURIA. as the Tomb of Tarasis Mr. n previously copied by Labo a fat: which is translated either custodian of the church or the numbers have never been inserted. in the list of consuls, in 366 and she rea epitaph of an inhabitant of the monaste 1er of the same name: bai us. The name We might is uncertain. 461. ding will of probability is in a monk on the spot in the year 461. t he spot in the year 461, and by consequence a church. was the church, it may haye been a later addition to the structure; but it ble that the present church was then standing, for there are no signs of an evidence is much more uncertain. but it is in the Byzantine province o 1 have been added by him and are wor he notice of Procopius proves that there a h and monastery in this place, and that t under Justinian. The epitaph of Tarasis early as the first half of the fifth 2 of some alterations in the por could be built during that perio either before the middle of the Let us turn t the beginning of the to temples, west and eas century and not la: The plan of end—and_ previous er.” the church absence of a dome a later date might Without sp inclined at present be dated, as the e seems to pre 18 The epitaph is edited below, p. 24. ey "Ioaupla...dvevedoaro. always at the west end, in the fifth the « ico, the church in my opinion shi addition—it represents a single design and a single date. until the time of Justinian Isauria was in a disturbed condition, and it is unlike o the evidence given by the str fifth century for it hes east and west, ze. with this date ; but all the other evidence points in my opinion perhaps be implied pitaph of Tar: % Procopius, de aedifictis, v. 9, povacripioy ets 7d "Amddvas * In the fourth century the altar seems to have been OWS no From the midd d. We are reduced to t fifth century or during he conclusion tha: he time ucture. It could all churches were, as far as we know, Ne 1 was based on a mistranslation very century. er to such evidence as we have of the date of the church. y called Tar Hogarth copied the epitaph of a monk rde, and under the auspices of the late Into that question Dr. Hatch had been made to play some part in ec iastical controve we need not now enter, for his use of the epitap natural to any scholar who was not < squainted with the study of inscriptions. important as giving a date. cut the inscription himself in his own At asis.!° For us it is ifetime and —was presbyter and mapapovdpuos (a word iff) from a certain year of Space is left for the date of his death and Gadalaiphus or Dagalaiphus 1ope that the year of the ind It may be the tenth year of The first Dagalaiphus was consul in the tenth year of It she indiction uis age, but occurs twice iction would she indiction, an indiction, be seen by referring to the discussion given at the end avour of the reading 14, anc that in all However we translate hat he held such an office proves that there was already a It does not of course thy of traces sign 0 not the a built 1 50 ac represents the transition from the older clude a church of the time of Justinian. by the capitals of the pillars in the the ¢ later domed church of Justinian. The absence of an external apse, sanctuary, the apsidal terminations to the sides, all seem to point to the earlier date. does the style of ornament and the a sence of more definite Christian symbolism. Wit be asilican form to Procopius says that Justinian restored a monastery at Apadna in Isauria. There is of course no certain evidence for Isauria, its size and character show that it must haye been a place of importance before the time of Justinian, him, and there the province which can vie with it, so far as we are aware. ready existed previous rere was a restoration, the foundation h the exception restoration or e of the fifth century y that a church ; the ¢ of Justinian. hurech was built earlier than ar at the east ke the heathen ate in the fifth the hambers of the So too Again the On the other hand nave, eaking dogmatically—and I am certainly not qualified to do so—I am to hold, until further light be thrown on the subject, that the church should sts, to the first half of the fifth century. and the position at the east end became the normal one. s to be in 417. change took place ties there cited. st known and dated example of the later custom See Kraus, Realencyklopidie der christlichen Alterthtimer, s.v. Orientirung, and the authori- ECCLESIASTICAL SITES IN ISAURIA. 19 Let us end by reconstructing as far as we can the ustory of the site. ae The beginning of Monasticism in Asia Minor dates from the year 357 or 358, when Basil of Caesarea returned from a visit to Egypt, Syria, and Mesopotamia full of zeal for the asceticism which was everywhere springing up, and became the vigorous promoter of the new ideal of life in his own country. Under him the coenobitic life Ges tuntavorihineerdl into Asia Minor, and in the next few years monasteries spread everywhere throughout the country. But a more special influence may have operated in promoting its spread in Isauria. Tn the year 375 Gregory of Nazianzus disappeared suddenly from public life and retired to Seleuceia, where he lived in seclusion, attached to the shrine of St. Thec i—already famous—for three yea Tt is possible that his influence may have helped in the foundation of monasteries in Isauria. However that may be, some time probably not long after the rise of Monasticism in Asia Minor, the place called Koja Kalessi became the resort of hermits. The sites of Christian monasteries in the ear y Church seem to have been determined by two main causes. Hither they sueceeded to a heathen temple, and a locality already sacred in popular tradition (to this type the monastery of Daphni near Athens belongs, which succeeded to a temple of Apollo), or else they grew up in spots which had been made famous by the sanctity of Christian hermits, or tradition. Such sites are Mount Athos and the monaste of St. Anthony and St. Paul by the Red Sea. These latter would of course be built in places removed as far as possible from any signs of the old religion. To this latter cl. Koja Kalessi belongs. The place is quite unlike any which would be chosen for a heathen temple, there are no traces of any remains of an earlier building, the church is in no place built of material which has been used up a second time, all the workmanship bélongs to the period of the church itself. The place was chosen then as a resort of hermits, We know something, from an often quoted letter of Basil the Great, of the motives which inspired the Christian hermit in his search after a place of retirement. He has described to us his own retreat in Pontus. It was a place exactly corresponding to his taste. He tells us of the lofty mountain and the beech woods and the cool and transparent streams and the view on the great plains below. ‘What need to tell of the exhalations from the earth, or the breezes from the river? another might admire the multitude of flowers and singing birds. Leisure he has none for such thoughts. But the chief praise of the place is that it nurtures what to him is the greatest produce of all—quietness.’” Solitude and beautifu scenery the hermit and the monk always seemed to long for; for the refining influence of their religion seems to have awakened among the Christian solitaries of the fourth and following centuries that love of nature which has since become a commonplace of the literary world, and solitude and beautiful scenery he will find in the elevated terrace of Koja Kalessi. The view from the mountain. plateau over two great river valleys and the wooded slopes of Taurus, with the gorge of the Calycadnus below, is of extreme beauty. A spring close at hand supplied water, and the cave now outside the monastery gate would form the first home of the vermit, like the caves of St. Anthony and St. Paul in the desert by the Red Sea. Probably he fame of some hermit made the place famous and a monastery grew up on the spot. If B he results of our previous investigation be correct, the monastery and church were built in he first half of the fifth century. In the troublous times that followed when the mountains of Isauria were the scene of a great deal of wild turbulence, the monastery fell into decay, and this may be the reason why the epitaph of Tarasis was never completed. It may even ye that he left the monastery in his lifetime. Repairs were made by Justinian and to his time may perhaps belong the big reception hall with its doorway which seems to be of a later sriod; but with the Persian and Saracen invasions of the seventh century it would again ye deserted, and probably has so remained to the present day. Its remote situation and pe the absence of any population near have preserved it undestroyed to our own time. a1 aparov pev jAOov cis SerevKeay puyas the shrine of the Virgin Thecla, and become one of the tov rapbevava Tis dowipyou Kopys attendant priests. In that case the use of the word is @ék)as. interesting. The principal temple in Seleucia was of Greg. Naz. Carm. ii. 1, xi. 547, pp. 702, 703. Athena, and the worship, the name, and the cult could all be t The word zap$evdy is ordinarily translated ‘nunnery,’ nsferred to the Christian virgin. and with that meaning it is used in later times. But it is See Basil, Hpp. I. unlikely that Gregory would go to a nunnery. On the translation, //istorical Sketches, ii. p. 59. other hand it is probable that he would attach himself to I have made use of Newman’s S The second si mountains from Mut and it is near this the east. It appea which are given be summit of the p about 10 hours. T in Byzantine times erecting a strong pl 2,.—KusteL or Da Bazar [Cororis Da Bazar are the ruins of a considerable town. described by the Rev. E. Davis.* We approached it from Karaman, from which it is distant called at various dates Coropissus, Hieropolis, and Sebilia. he crossed the mountains from Karaman to Selefki, just before his deatl ECCLESIASTICAL SITES IN ISAURIA. Suprita-Himroporis |. e to be described is not of great importance. There are two roads over the to Karaman: the first goes up the northern branch of the Calycadnus valley, that Koja Kalessi is situated ; the alternative route is somewhat further to rs to pass near the site called Sinabich (Dalisandos), ims iptions from ow, and then enters the mountains; a few hours before it reaches the it passes near the Kestel Yaila, and below this at a site now called This route was traversed and the ruins have been hese ruins have been identified by Prof. Ramsay with a town which was Tt was of Roman foundation ; it became the centre of a bishopric, and the materials were used for ace of the kingdom of Lower Armenia, visited by Frederic Barbarossa when 24 Ds later than the time not thought it necessar topography of the d The ruins are considerable, but for the most part destitu one almost undecipherable inscription on the spot. those of a church, of which a plan and photograph are here given. having been built on foundations dating from an early period, perhaps those of the temple that it succeeded, and was surrounded by a large peribolus wall. the west end was a structural narthex. consisted of a nay external, the aisles in side chambers. seems certainly later and of a more conventional type than t 23 Life in Asiatic Turkey, by BH. J. Davis, p. 24 See Ramsay, Historical Geography, p. 366, 369. I have ry to go into further detail on the strict, as the geographical results of Fic. 9.—S.E. Vinw or tHE Cuurcn at Kesren. e of inter t, and we only found The most conspicuous and best preserved are It has the appearance of Although probably early—being not of Justinian—it has no features of great interest. The plan is simple. At From this the church was entered by three doors. It e with two aisles. The nave ends in a bema and apse which is partly There was no ornament of any interest. The chureh vat at Koja Kalessi. our journey have been described by Mr. Hogarth in the Proceedings of the Royal Geographical Society, Suppl. Papers, 1892. ECCLESIASTICAL SITES IN ISAURIA. 21 = only remains to express the obligations which the Asia Minor Exploration Fund is under to MM. Schultz and Mr. Barnsley for the trouble they have taken in enabling us to produce a satisfactory account of the church at Koja Kalessi. We are indebted to them for the plans, and - a ee ' 1 Settles 1 1 i 1 1 1 1 1 LV VJ; Wf Uy YY) y aE KESTEL-(COROPISSUS) GY Y Sitnnaltmute i if x" trp Fie. 10,—Pian or tHe Cuurcu at Keser. on their own. I have for bestowing more labour on other persons’ work than many would do also to thank them for many hints and illustrations which they have given me regarding the architecture and the analogy to other buildings. § 3.—Norss on orHeR Eccrestastican Remarns in Asta Professor Ramsay has at my request drawn up the following notes « Yemains in Asia Minor, which may serve to indicate how much work stil his direction. I. Phrygia 1.—There are among the Phrygian mountains a consic ias all the appearance of being a monastery. One of the churches besi These remains are scattered over the country of the rock monumen Inn and Seidilar. Near Bayat, on the rock at Assar Kumbert Kale, ainting remained in 1881. tS. Minor. concerning Heclesiastical remains to be done in erable number of rock- cut churches and other remains of hermits, and one curious group of chambers near Seidilar which de Ayaz Inn is, though small, an almost perfect specimen of a domed church of perhaps the sixth or seventh century. especially near Ayaz part of an old church Some of these would cer e.g. at Kolossae, Apameia, Hierapolis, ete. much interest ; but all would require excavation before anything could 2.—There are many churches of which the ground-plan and part ‘tainly prove to be of of the walls remain ; be learned about them. E I. Cappadocia. Dere are well known and have often been described. yield much information. I have not which I have seen, exist at Gelvere preserves the name of Karbala, the traces and relics of him. It claims is a large, prosperous, and wonderft so far as we saw them during a hu examination however might perhaps Matchan, the late Byzantine spent a few hours there in 1882. i Several days would have been needec number of the graffitti, but left many churches. observed several of th (1) Bactdeds Ovadévtse xaxds érivoas (exrotnoas) 67m anéSoxas THY éxd[n]olav 7d (i.e. Tots KaKodd£ous ’Apcavois 2), (2) K,BYAOOE TOCTHCE KAICHYZATO OIAPIANYTP HMCPAC ¥IKYNY CHINA/TYCIE KAICIA ECCLESIASTICAL SITES IN ISAURIA. " 2 + al avstharal ls The enormous groups of rock-cuttings at Utch Hissar and Soghanli They are curious, but do not appear to Similar groups, Gelvere seen them, and can only judge by report. in the territory of Nazianzos and at Matchan. patrimonial estate of Gregory of Nazianzos ; and is full of to possess also the head of St. John the Evangelist. It the rock-cuttings are quaint but, ily picturesque village : More careful uried visit, little can be learned from them. prove instructive. bishopric Marsavy, lies three hours east of Nev Sheher. 1 ut chambers, tombs, and ffitti in red paint. considerable There are here numberless rock latter which contained numerous ¢ I copied a specimens fe e to examine the place proper y unattempted ; the following may serve as xaxodoév ’Apiavos xat Bovrobé- (2) Tole|s THS é- krilolals nvEato NG of ’Apiavd pis Epas. ov Kuvu (KoLvot) yy avtos (avtois) i é- Kricla. (4) 6 88 dytos Bacldrevos cuvdtas Tovs Xpirtiaves IT]. Lycaonia and Isauria contain t seen. Their name is legion. (1) Bin Bir Kilisse (Thousand of churches, none however of higl dilapidation. (2) Kilisra, about seven miles rock-cut churches and tombs here are the seen. architecture, and so well preserved. The mo One very tiny church was a per 1e best preserved Christian monuments which I have ost interesting are : and One churches), the ancient Barata, where about a score 1 architectural value, are standing in various degrees of 1-west from Khatyn Serai (the ancient Lystra). The most beautiful and the best preserved that I have fect gem, so tiny and so complete in its miniature nort. (3) It is quite certain that many others remain to be discovered in these districts. (4) At Olba, about a mile down the ravine which runs towards the sea from the fortifi- cations of Oura, there is a monaste returning from a three days’ excursion to Corycos; but were so tired, and so ut energy, that we did not climb up the steep side of the ravine to examine. same kind of situation as Koja Kal a mountain, The buildings seemed to be in excellent preservation. INSCRIPTIONS The following inscriptions were to whom I am also indebted for the (1) Mut: in the courtyard of the school (Mekteb), opposite the castle and mosque. W. M. R. sarcophagus. Tall thin letters. We saw it when erly devoid of It occupied the essi, being perched on a long narrow ledge in the face of ary high up on the eastern precipice. From CiniciA TRACHEA (ISAuRTA). copied for the most part by Prof. Ramsay and Mr. Hogarth, greater part of the work required in editing them. On a to ECCLESIASTICAL SITES IN ISAURIA. SANP CTE TITOEmNAOYIOEAMMOECSENME ETMOLEIAYTwoKAIANTWNIACWTHPIAITHCYABlo TIAPATT EAAEITEXAHAENIETEPWTEGHNAIENT W TAD W TOYTW EANAETIC TIEIPACEITOYS EINAIOYTOCANENENKEIEICTOIEPONTHCTIOAIAAOCAGHNAC AHNAPIAXEIAIAKAITWKAAYAIOTIOAEITWNAHAAW AHNAPIXEIAIA (sic) at AW y i Xaipere . Tiros Praovwos Anpoobévns erole: adT@ kai Avtovig Swrnpids 7h cupBlo- Tapayyéehret Te wndevi éErépo TeOivar ev TH Tahy TovT@, édv bE Tus Teipdcet TOD Oeivar obtos dvevévees eis TO (epov Ths TloAudbos ’AOnvas Snvapia yelia Kai TH Kravdvororetav Sijum Snvdpra yelua. This inscription confirms one of Leake’s many brilliant conjectures. He first placed Claudiopolis at Mut, and he has been followed by all or almost all subsequent writers, but no epigraphic evidence of the fact was known previously. We may add that we purchased a copper coin of ‘Claudiopolis’ of the reign of Hadrian at Karaman in July, 1890. The similarity of fabric and style to those of other Cilician coins and the place of provenance are considered. by Messrs. Wroth and Head of the British Museum sati: ory proof that Professor Ramsay’s suggestion was right, and that it is the Cilician Claudiopolis to which the coin must be ascribed. No other coin of this town is known as yet, and our solitary example is now in the British Museum. (2) Mut: high in the outer wall of the castle, facmg the mosque, read with a glass. WevicgRs Tpoactio All [A]et OAYMTTIO Odvprri@ HTIOAIS ) TONS From this and the preceding inscription we obtain the names of two of the temples of Claudiopolis, one of Athena Polias presumably within the city, the other of Zeus Olympius, before the city gates. In both cases Greek names are applied to gods, who were probably native to the country and akin in type to Zeus Asbamaios of Tyana and Ma of Comana. (3) Mut: in a Turbe beside the mosque. AIOCKOYPIAHETPO Avocxoupléns Tpol pi IWKAINEIKHTHTY Elo «al Nelen th yu- NAIKIKAIADMNEINO vaxt cal Aopvelve (sic) AAAWJAENIE ZONE! aro &é [odd]evi Fou e- 5 NAIHALILEITOTAMEIO, vat i) ddcer TH Tape [Kp ? The name in 1, 2 is uncertain, as the edges of the stone are broken, perhaps Tpo[ ¢iu|o ; but probably, if the stone were complete, it would be found that line 1 extended further than the rest on both sides, and that the reading is Adp.| Avocxoupiéys Tpo |... . eavt |p nai#Necen (4) Mut: seven minutes along the Selefke road on the left hand side, on a large sarcophagus. AIIC . OIAES “Ams ’Odia E . . . [xate-] CKEYACENTWAAPNAI oxevacer Tv Napval ea EAYTHEKAITWANAI éavTh te Kal TO avd[pi KAITWITATPIAMAMHI kat 76 marpi dma pnr[pr 5 KAITWPFENITHLOO'!¢ Kal TO yéve Ths Od[ias] AAWAEMHAENIE////ONETT////BAAIN aro be pndevt eEov eriBariv. ECCLESIASTICAL SITES IN ISAURIA. and 31, “IvSas Movtavou rod 4 TO yeu THs ’Odias: of. the mscriptions of Sinabich Nos. 1: yévov, and on the female name see the note introducing No. 26 sq. Prof. Ramsay suggests Avp. Lopéa and line 5 rhs Lopéas. If this restoration is right the inscription is doubtless Christian. (5) Mut: in the Regie office. AYPHAIAEAAMICHMH Adpndtia Eppion pj- THPEAYTHCKAI Tp éavTH Kal TOICTEKNOICAYT > tois téxvos avt(js) KAIAAHAENIETEP WD) kat pn(d)evi érépm EZONEINAIETIEAABA é£ov eivac érepBa- AEINIEIAEMHAWCI Aely: ef bé pr) SHoe TWTAMEIWXBOKAI TO Tapelo (Snvapia) BP’ Kai THTTOAEITAAYTA TH Mode TH avTa. (6) Mut: in the Regie office, belonging to Demetrius. Late letters. 2WNXEPECOEYPPENOM ENO! Below, on a segment of a circle, in the centre of which is a fragment of the Christian monogram £& : WNENOAAEKITET.. . Tp ?|pov x(al)pecO(ar) edpp(o)vo(d)wevor. . . Tpiflov évddbe kite m[pecP. The proper name may end in -per or -¢ev. Perhaps we ought to read edppovovpevors (the ¢ being added as an afterthought and placed wrongly, so that it has become attached to the N), wishes good fortune to the ‘discreet,’ ze. the Christians. The restriction of the salutation to co-religionists belongs to the fourth century or later (a date which is also demanded by the monogram ¥%, and the title z[peo@irepos|, if the restoration is right)—see J.H.S. 1883, p. 406 sq. (7) Koja Kalessi (Apadna?). In a tomb marked in the plan. +ENOAAEKATAKITE +tév0dde KatdKcte TAPACICAICTENOMENOC Tadpacis Sis yevouevos TIPECBSKAITIAPAMONAPIOC mpecB(vTepos) Kal mapapovapios TIAPOIKHCACENTWTOTW Tapoikncas ev TO TOTM 5 TOYTWATIOYTIATIACTAAA TovT@ and bratias Vada- AAITITIOYINASIXEWCINAS Ala]/mzov ivd(uxTi@vos) 66’ (2) gws ivd(cxTLdvos) YTATIAC ZHCACTA tratlas Znoas ta TIANTAETH Tata éT™ Copied by D. G. H. See also Laborde, Revue Archéologique, iv. (1847) p. 175 and CU.G. iv. 9259. 2. TENOMENOC . ||. 6. AAITITIOY INAs 1X The inscription seems to have been cut by Tarasis during his lifetime with blank spaces to be filled in on his death; this was never done. Tdpacis| The name seems to have been very common in this district, cf Sterrett, Wolfe Expedition, 59, 60, 97, etc. ; in 97 a Tarasis commemorates his brother a deacon Tapacis Nados | éméatnoev oTAnv S - , ~ | n | [Anvadpe aerO@ | avtoo Siaxdvo | F Tépacis Sis| Tarasis the son of Tar 3, see Ramsay, Oaford Magazine, Vol. ix. No. 24 (1891), p. 411, and of. Reinach, Lpigraphie Grecque, p. 508. ‘This idiom is so common, especially in inscriptions of Asia Minor, that it is hardly necessary to give instances. It is commonest ECCLESIASTICAL SITES IN ISAURTA, 25 » erhaps in the form 4’; this occurs for example more than ten times in a long list of names edited by Sterrett, Wolfe Expedition, No, 366; Ramsay, Jowrnal of Hellenic Studies, 1888, js BS, . For 8s and spis and cerpdeis, which also occur, see Sterrett, Wolfe Expedition, 555, 609, Avp. Mapxos Madpxov Sis, and Ramsay, ‘Cities and Bishoprics of Phrygia, J.H.S. 1883, 3 ip, 398, No, 14, Adp. EiSoueveds TeTpaxess. ; a But the inscription and the interpretation of these words have had a curious history. Firs b published by Laborde in 1847 and by Boeckh in the fourth volume of the Corpus, it remained unnoticed until Dr. Hatch, with the thoroughness which he always displayed in collecting matter to illustrate his arguments, made use of it to prove the possibility of reordination, taking 8s with yevouevos. He refers to it Dict. Ch. Ant. Vol. ii. p. 1520, Bampton Lectures (ed. 3) p. 187, n. 51, and his use of the passage is accepted by Harnack (Die Gesellschaftsverfussung der Christlichen Kirchen im Alterthum, p. 234, n. 13), who however does seem somewhat startled by this isolated example of reordination in the fifth century, at a time when the usage was admittedly fixed. Mr. Gore (Lhe Church and the Ministry, ed. 1, p. 189, n. 1) points out that the two expressions mpeoBvrepos and mapapyovapvos. should be taken together, and that the words would merely imply that ‘Tarasis was twice appointed presbyter and residentiary of this church.’ As against Hatch and Harnack this would probably have been preferable, but the interpretation given above, which is certainly right, removes all ifficulty. uri mapauovaptos| There is some doubt about the meaning of this word. In the Second Canon of Chalcedon it oceurs as a various reading for pocpordpis, and is coupled with oixovenos (the church steward) and é«Sccos (‘ defensor’), In one of the Councils of Constantinople 1e Tapapovapiou are coupled with the Eevoddyor. In Conc, Nic. ii. 5. 7 we have et ris érloxoros Ls tee Now a x ¢ = TpoBarrotto ert NpPHwacen olKovojov 7 ExOtKov ri) Tapapovdptov ) dNMs Twa TOV KANpov. All. these Es passages—although none is decisive—seem somewhat to support the interpretation given by Beveridge of ‘ villicus’ or ‘ bailiff.’ On the other hand one of the Latin interpreters of the canon of Chalcedon translates ‘mansionarius,’ 7c. a resident officer who had charge of the church and its utensils (Du Cange, ‘Custos Ecclesiae *); and that meaning suits the probable derivation of the word and is not inconsistent with the passages quoted above. See Du Cange, sub voc. ; Bingham, Ant. 1. 360; Bright, Notes on the Canons of the First Four Councils, p. 129. ivStereavos 18] The final question remains of the date of the inscription. We have to guide us two facts, the name of the consul and the indiction, but neither is free from ambiguity. The name Gadalaiphus oceurs twice in the list of consuls, in 366 and in 461, in the form Dagalaiphus or Dagalaifus. Now taking the system of indictions which began on Sept. 1, which was the earliest in use, the year 366 corresponds to the years 9—10 of the indiction, and considering how late the indiction began, more probably to the year 9; the year 461 to the years 14—15. The reading of the inscription is unfortunately doubtful. The ¢’ is certain, the next letter much damaged; it may be merely a mark of contraction or it may be a 8. Laborde seems so to have read it, and Mr Hogarth, who copied the inscription, has no doubt that this was the reading. In the former case the date will be 366, in the latter 461, and the balance of probability distinctly inclines to the latter. (8) Koja Kalessi. +NEWNAKA Néoy ’Axaf. (9) +K IAI //31///¥ Ill] KAfns.... Of No. 14 below. (10) +WTWKO WKW NOM oy The name Kévev, which is very common in Lycia and Isauria, is probable here. (11) Da Bazar (Coropissus) : stone built into east wall of unfinished mosque south of the ancient wall of the city. Letters large of late date, and much defaced, ECCLESIASTICAL SITES IN ISAURIA. TeOC «... IPTHUCAECTIOINHC. .. Ths Secrrotvns [0- EMEAIO . OENTAPAIIAPITINO ewedlo[ O]év [a ]apa Mapirivo[u ’Ia- (?) TP. IKAEOCC. A, TVHMWNOA tplo]xAeos z[od] (5)[e]omo(Tov) 7udv Pr. - - ETOYC . KE Vv éTous Ke’ The lettering of the inscription is so much worn that the reading and interpretation are very doubtful. érous xe’ probably gives a date, and means the twenty-fifth year of the Seorérys, either a bishop or emperor; and $d’ may represent the era of the province. Sinabich (Datisanpos). The twenty-four inscriptions which follow were copied by Mr. Hogarth from sarcophagi on a hill called Sinabich which rises not far from the base of the northern wall of the Calyeadnus valley, on the left bank of the Pirinj Su, some six miles north-east of Mut. The hill is a striking one, isolated on three sides, and crowned by a high cliff; up the south slope winds an ancient road-way, which conducts to a plateau on the top of the hill, strewn with ruins of a Byzantine city. The remains are all poor and without character, except those of the Sacra via, which runs along the north and east of the plateau: here are a great number of sarcophagi, almost all inscribed with dedications in late lettering, of which Mr. Hogarth copied as many as he could between 4 p.m. and sunset on the day of his visit; a few, however, were left for future epigraphists to decipher, and these, we believe, have been copied by Messrs. Heberdey and Wilhelm, who confirm our conjecture that the site is that of the Byzantine bishopric Dalisandos. That Dalisandos must have been situated somewhere here is proved by Prof. Ramsay (Historical Geography, p. 366, of. 495), and the character of the site exactly corresponds with the description given by Basil of Seleuceia quoted by him. (12) INTACICBAAIOYE “Tvyaots Badiou é- TIOIHCACAYTWKE moinca [élaure ae TOICTEKNOICAAAO tols Téxvots GANO- TPIONAEMHAENIE Tpiov dé under é- ZONETIENTEOH £ov érevteOh- NEHATIOAWCITWHICKWXXIAIA ve }) drrodHo TS hloxw (Snvdpia) vida. The grammar seems to need dddozpép : unless the composer understood émevreOjvar as an active infinitive. The latter alternative is necessitated by dmoddéce, as otherwise the unlawfully Mi buried man would be the subject to it. (13) AOQHNAIOCMENEAAOY “A@nvaios Meveddov ETITOIHCENEAYTWKAITH érrolnoey éavt@ Kal TH CYNAIKIKAITOICTEKNOIC yuvatki Kai roils Téxvo.s MNHMHC XAPIN Lins xapw. Seribbled to the left of the above: INAACMON "Ivéas Mov- TANOYTOYE Tavou TOD yé- NOY vou. With the name "IrSas compare “Ivins, which is mentioned by Eust. Epiph. fr. 6 (Miiller, Frag. Hist. Gir. iv. p. 141) as the name of an Isaurian; see also Nos. 20 and 31 below. "Iv5as was probably the stonemason who had prepared the sarcophagus, and he belonged to the family or guild of Montanus (¢f. the inscription from Mut above, No. 4). For inscriptions giving the name of the artist of J. H. S. xii. (1891), ‘Inscriptions from Western Cilicia,’ p- 261, réyvn @pacéov, and No. 42, p. 262, ‘Epuddiros érroder. ECCLESIASTICAL SITES IN ISAURIA. 27 The name Montanus, although specially connected with Phrygia and with the well-known heretic, is found elsewhere in Asia Minor ; ef. Dict. Ch. Biog. iii. p. 935. The names ’A@nvaios and Mevéraos suggest a hellenizing tendency. A similar passion for Hellenic names is shown in INosmls3) 16, (14) + TATTICKIA€OYCEAYTH Tarrls Kinéous éavrh AAAWAEMHAENIEZON Gd Se pndevi eEdv ECT WHAAONONKAEONIKON gota 1) povov Kredvixoy TONYIONOCANAEETIENOH Tov vidv ds av 88 érévOn ATIOAWCITWOHICKWX aroddot TS hick (Snvapia) TIENTAKOCIA TEVTAKOTLA (15) AOPMICTIACKAAAA//// Aoppls Tacna Aad[icavdeds O- NIKIECTHCEAYTWKE (H)a(n). eornce adres Ké THAAHTPIKETEKNW Th pytpl Ke téxve@ [Ke AAEAPOYTEKNWKE adehpod Téxvp Ke EFTONOICA .. e€////ATH éyyovors dpa e(or)v ’ATy NHAAOYAAAWAE MAH ynuou (?) ddAA@ 88 pen- AENIEZONTEOITAE! HA devi €£dv TeO(Fvat) 7) a- TOAWCITWHICKWXXIAIA moddat TO plop (Snvapia) xia. The reading AadicavSeds, although by no means improbable, has of course no authority. Avrod and airg for éavrod and éavré are common: perhaps in all cases avrod and avro® should be read, but these more correct forms are not usually printed by epigraphists, and it is certain that the Greek writing of central and eastern Asia Minor is particularly inaccurate in the pronouns and pronominal adjectives. (16) MAYPABHNOAOTOC M. Adp. *A@nvd8or0s ICIAWPOYEAYTWICIHW *TowWdpou éavté [Kat] “loide- PWKAIMENEAAWTOICTEKN pe kal Meverd ois réxv- OICKAIETTONOICAPCECINAAAW ots Kab éyyovors dpoeow, GA AEMHAENIEZONECTINHATIOAWCIT//////// 8¢ pndevt Fou éorw i) droddor 716 Ta- MEIWXBOKATHTIOAIXBH Hel [Snvapia) Bh! Kal rh wore (Snvdpia) Bb All the names here are foreign. Cf No. 13, where we have an Athenaeus the son of a Menelaus. (17) ICIAWPOCTIAPICTIOYOKAI IaiSwpos Haplomou 6 Ka IPAICETIOIHCENEAYTW "Ipods érrolnoen Eaves KAITATATHIYNAIKI xa Tara TH yuvackt AYTOYKAITOINC///////C abrod Kat robs [réxvo]is AYTOYMNHMHEXA avTod pujuns ya- PIN piv. On this inscription, see on No. 19. Professor Ramsay suggests that the Persian name Tapéoxas (Plut. Artaw. 12) may have been transformed by an error of the engraver, or by aot: ; popular pronunciation, into Iapsoras. (18) TIAYAOCTIAYAOY Tladndos Iavnov ETIOIHCENEAYTH érolnoey éauTh MNHMHCXAPIN puns xapw. There can be little or no doubt that this inscription is Christian: éav7é should be read. ECCLESIASTICAL SITES IN ISAURTA. (19) ICLAWPOCKPICAMOOYOKAI "TolSmpos Kpicapoov 6 Kal IPAIC EK TWN Ipdis ex Tav IAIWN THN iSlov tiv COPON M N H copov pv7j- MSE X A PIN Ls xapw. In No. 17 we have an Isidorus son of Parispas, also called Indis. Was Irdis the native name and Isidorus the assumed foreign name? Compare Ramsay in Kuhn’s Zeitsch. fiir Vergl. Sprachf. 1886, p. 391. The difference in the size of the lettering may imply that this was one of many sarcophagi made according to pattern with a space left for the name, and that the name was filled in afterwards, in this case in smaller letters owing to its unusual length. (20) INAOYCEOKAI *Ivdous €' 6 kab TIAYAOCBATH Tladros Ba rH EAYTOYIYNEKI éavtod yuveri CWPOLYNHE coppocdyns TEKAIEYNOIAL Te Kal edvolas XAPIN xdpw, Ivdovs| Cf: Indas, 13 and 31; on the double name cf. Nos. 17 and 19, and the remarks there. Here the change would be due to the influence of Christianity. Ave there other instances known of the use of ¢ in this way? It means the fifth of the name; & is commoner. *IvSous is here masculine, AvAdous 1 (J-ELS. 1891, p. 266) is feminine ; probably the accentua- tion differs according to gender, though how the accent should be placed is uncertain, (2 1) MMMM ove OH "Etov]s on’ THLETTAPXEIAL Ths erapxelas T. TOMTTWNIOLIO T. Tloprevios *Lo- YAIANOLKAINALBIA udtavos kat Nas BiA- ALOCETIOIHCLANEAY Mos érolnoay éav- TOICKAITOILTEKNO Tols Kal Tois TéKVO- IZCMNHMHEX APINOP 1s punpns Xap op- KIZWTHNCEAHNHN Kibo Thy cedjvny. Erov|s on tis émapyelas| Cf, No. 27 below. The era of Dalisandos is probably either 72 (the date of its incorporation into Cil ia) or some year in the reign of Pius (137—161) when the three Eparchies, Cilicia, Isauria, and Lycaonia were instituted. With Nas cf. Nav rv yuvaixa adrod, J.H.S. xii. (1891), ‘Inscriptions from Cilicia,’ p. 229; Nav Aaxpatovs thy éavtod yuvacea, tb. p. 262; cf: also Mas tov éauras dv8pa, ib. p. 261; Aas, 2b. p. 262; and Ba above, No. 20. Tés occurs perhaps in No. 28, but is masculine. The name *Evas in Pisidia is doubtless the same with prothetic ¢; it is always feminine. Similarly we have Bas and ”Afa. Enas has been maltreated by MM. Duchesne and OCollignon, B.C.H. vol. ii. p. 603, No. 15, and by MM. Cousin and Diehl, B.C.H. 1889, p. 342, who read évdrn yuvaret in place of "Eva Th yuvatKt. (22) K.. 1EMOPOAICIOIME K[AA]is "Agpodiodou pe- TAKYPIOYANAPOCE Ta Kuplou avdpos é- TIOICENEAYTH Trove EauTh [Ke TEKNOICMNHM-XAPIN Téxvols uYENS XapLV. Kids “Agpodioiov is Prof. Ramsay’s conjecture: on the name see No. 29. Here we must doubtless understand wife of Aphrodisios. Cf No. 31 below. 1 With Isaurian Awddovs compare Thracian Ad\Aes Frag. Hist. Gr. V. i. 36 ; Bithynian Aides, Athen. Mittheit, 1887, p. 183; also (at Cyzicos?) Athen, Mittheil. 1885, p. 20. ECCLESIASTICAL SITES IN ISAURTA. 29 CALOCIOYAIOCKEAEP OYETPANOCEAYTW KAIOKTAOYIATHTY NAIKIKET////////T .. NOIC MNH/////////////H/ V///ETEOHAEAAAOC Datos TotAvos Kénep overpavos éavTe nar ’Oxraovia rH yu- vaixl Ké tT[Lois] 7Lé« vous priluns xapwv drArw] (2) [ovsen) fv éorw reOjv-] av.] éré0n 8€ GdXos. rv F sei 5 her The last three words were added, probably during the lifetime of Celer, to indicate an exception made to the prohibition. (24) ZHNOPANHEKAIAIOMH Znvoparns cai Acopuj- AHLCAAEADO! MEPOL dns adeXpol pépos KAI TATIC MELOL kat Tarts né(p)os ETTOIHCANEAYTOIC émoinaay éavtois MNHAAHELXAPIN Hunens xapw. Two brothers, Zenophanes and Diomedes, made one part, a sister, Tatis, the other. The tomb was erected at joint expense and perhaps contained two divisions. (29) IAII//PNIKOC OYIFEPIOC (26) TTWAPICMONTANOY OKAITIATPODIAOCETT OIHCENEAYTWMN MHCX APIN K(Ae6)pexos Ovuyéptos. Tlwdpis Movrdvov 6 «at Marpodiros éx- olnoev éauT@ pv(y)- PS xapw. For the termination -8pus cp. Mayydpis, J.H.S. 1891, p. 264. The remaining seven inscri instances of descent traced through the mother. certain, and prove the existence corroborative evidence. assigned to various causes (see Pa Philol. xix. 37, p. 88), but so many instances occu in this part of Cilicia coincided with that of the Lycians as recorded by Herodotus. When ‘sin, tions have been grouped together because they all appear to be The first two, Nos. 27 and 28, seem quite of the custom; the others are suticiently strong to give gle instances occur of descent through the mother it may be on and Hicks, Inscriptions of Cos, p. 256, Hogarth in Journ. ing together suggest that the local custom We may compare also No. 4 from Mut. The exceptions are to be explained by the gradual hellenization of the people and their customs; they might also arise from the influence of Christianity. Bp. Lightfoot (Phil. p. 55) draws attention to a similar custom prevailing in Macedonia; he ascribes it to the elevated social ETOYE OBTHLCETIAP XEIAL CIAAI//ENH CIOLOKAIKAEO NEIKOCKAINENHE IAMBIOYHKAITATAL ETTOIHCANEAYTOIC MNHMHLX APIN osition of women. érous 08’ Tis émrap- yelas, Stra[s Nlevy- alos 6 Kat Kired- vexos Kat Nevno[ts "Tap Riou 4 cat Tatas érroinaay éavTois LUnENs Xap. It is clear that the name Nenesis (the restoration of which in the second line seems A Lawes Ue? I roy Si srefore trace 1 y is ap. certain) is feminine, and that the descent of Silas is therefore tracc d through his mother. TTY . TAC TYPANNIETIOIHCENAIAY TW KE THTYNEKIAYTOYKETE NOIC MNHMHC XAPIN (28) (A)t[p.] Tas? Tépavve éroincey aiav- TO Ke TH yuvert adrod Ke ré[K- vows mununs xapw. ECCLESIASTICAL SITES IN ISAURIA, The name Tyrannis is proved by No. 30 below to be feminine. It occurs in the form Tyranis, CLG. 4971, b, Add. ’Arefdvdpov, untps TupaviSos, and in the form Tyrannis, C.1.G. II. 3796. In both cases it is feminine. This may be counted therefore with No. 27 as a clear instance of descent traced through the mother. (29) IPAICKIAAIOCTHNCOPON “Tpdis KidAtos thy copov EAYTWKATECKEYACEN éavT® KatecKevacer, With Kéadwos compare Kercous, No. 14. We have perhaps a similar uncertainty m Tyranis and Tyrannis (Nos. 28, 30). Such variations show how uncertain the native Isaurians, who can have been only very slightly acquainted with Greek, were in regard to the hellenization of the native names. Killis is shown by the termination to be feminine. No. 22 cannot, however, be quoted in corroboration of this, as the reading is too conjectural. (30) TYPANICTPOKO Tupav(v)is Tpoxd- NAI voe éroinaey ETIOIHCEN éauTh py7- EA pens xapw. YTHMNH MHC XAPIN In this and the three following inscriptions the name Trokondis occurs in the genitive in three different forms (Tpoxdvd:, Tpoxdvdv, TpoxdrSe). As the masculine form TpoxévSas is of frequent occurrence, it seems natural to assume that the name feminine. These instances of descent traced through the mother are not so certain as the two first given, and without them the analogy of Indis might be used to prove that the name ified in taking it as feminine in accordance with its termination, and in using these instances as corroborative evidence. as its termination suggests, 1s Trokondis was masculine; but when we have two certain instances we are ju (31) MAYPOYABABCICTPOKON M. Adp. OtaBaBots Tpoxdy- AINKETATICOYABABCEWC dw Ke Tarls OtaBaBoews KATACKEYACANEAYTOICTH- KxatacKkevacay éauTots TI- NOHIKHNMNHMHCX APIN v Onenv jvypns xapw INAACMONTANOY "Ivéas Movtavou TOYreéngoy Tod yévou. Tpéxovdiw is the genitive of the mother’s name. On the sporadic appearance of the suffix » in genitive and dative cases, see Ramsay in Kuhn’s Zeitsch. fiir Vergl. Sprachf. 1886, p. 386. Taris OtaBaBoews must here be translated Tatis the wife of Ouababsis : the expression is a variation of Tarts » yuu) abrod. So Kidus "Adpodiciov, in No. 22, seems to mean Kilis the wife of Aphrodisius. » : sn oh Oe Ivdas x.7.d. are cut in a seratehy style below the rest of the inscription. (82) TATALCTPOKON//PI 11 ECTH Taras Tpoxor| diy] éorn- CENTHBYD ATPINEC cev TH Ovyatpt Neoip MNHMHCXAPINA////////WAE pjuns xdpw a[rr]o 88 MHE ZONTWENBAAINHOLAN By e&dv Te évBariy, 7) ds av TIAPATAYTATIOIHLZHATIO Tapa TadTa ToUjon aTro- AWLEITWOILKWAHNAPIA dace 7H pick Syvapia AICXIAIATIENTAKOLIA Sioy ida Trevtakdova. Neon] Cf. Pape, s.v. = SSeS =4 ECCLESIASTICAL SITES IN ISAURIA. 31 (33) EPMOKPATHLTPO “Eppoxparns Tpo- KONAEIECTH-E Kovder éaTn we IAIPEIONTHEYN//// Gl yyletov rH yuv[ac- KIKAITO/////////////N Kb Kab Toles Téx|p- Hill, IMIS MAA EN//// ous. Gdd@] 8 pndev[? ME ZONAT W/////////////I/I/|// py e&dr iro: [oprt- ZWTHNECEAHN////N fo thy cedij[v nv CHAE OMI On the mention of the moon in the imprecation cf: JES. xii. (1891), p. 281 (‘ Inseriptions from Western Cilicia’), and Sterrett, Epigr. Journey, No. 31, also No. 21 above. ‘ On nro off JS. xii. (1891), p. 228, ‘Inscriptions from Western Cilicia, No. 4. ; Sterrett, Epigr. Journey, No. 31; and Ramsay in Kuhn’s Zt. f. vergl. Sprachf. 1887, p. 386, who suggests that iro is a reminiscence of the native Phrygian form. It is noteworthy that jr» occurs usually in that imprecatory formula, which in Phrygia and Lyecaonia is often expressed (even in Greek inscriptions) in the native language so as to be better understood by the people. ie The above inscriptions, together with those published by Mr. Hicks in the Jowrnal of the FHellenic Society (‘Inseriptions from Western Cilicia,’ 1891, p. 258) nearly complete the epigraphic results of our expedition as far as Isauria was concerned. It may be convenient here to add one or two notes on these previously published. In publishing the inscriptions on the great tower at the Hieron of Olba, My. Hicks dates the first ‘at least as early as 100 3B. Professor Ramsay had the n when copying the inscription, that the reign of Augustus was a probable date for it, and thinks that it perhaps gives the name of the dynast of the Teucrid family who ruled at Olba after Aba was deposed: the tower would in that case be built about 30-17 Bo. It has not been noticed that this tower is represented on some autonomous coins of Olba, of which there are specimens in the British Museum. Professor Ramsay notices also a few disc impress: ‘epancies between his reading and that of the published text in the case of the inscriptions in the Corycian Temple. He unfortunately had not his notebook beside him when he looked over the proofs, as they did not reach him until he had started for the East. The following may here be mentioned as of some importance. They occur in the graffitti on the stone marked (f) where he reads: 40-1. M. Adp. Movravos Sis 6 Ke *Avard[d les B. 42, M. Adp. K——s’Ackanra f’. 43-4, M. Adp. “Ep| woyévn 2\s 8 6 we [Novy Tleyus x. 45. M. Advp.——os ’AOnv0Swpos, vids——rod Ke——. I 41 a common contraction of ’Avarodwos, in 44 of Novvéyis is used. =— a SE A IDE LO LANES RAS NED ESET Ricuarp Cray And Sons, Lumirep, LONDON AND BUNGAY.