ISSN: 0041-4255
e-ISSN: 2791-6472

AYŞIL TÜKEL

INTRODUCTION

This article includes the documentation and comparative study on Alara Han near Alanya[1] which is a XIII. Century Seljukid Han. The field study on the building was done in August 1966. The measurements were taken in relation to a refence line established along all the walls of the building and by using the triangulation method. The photographic documentation was done at three different seasons: Spring 1966, Summer 1966, and Winter 1967 through 178 negatives, enabling the study of the effects of seasonal changes on the building. The measured survey was drawn in 1 /50 and the restitution drawings in 1/100 scale using the measured drawing standards established by the Department of Restoration, Faculty of M.E.T.U. Architecture.

Here it will be helpful to state some facts about the caravanserai and its particular development before discussing our subject, to be able to place it in the general development of Seljuk caravanserai institutions: Caravanserai, in its simplest implication, is a roadside institution to serve the caravans. But it was much more than that: It served as military barracks for the armies on the move[2], provided hospitality for the visiting sovereigns[3], served as prison[4] or place of refuge[5], and, when it was on a secondary route or when the route lost its function, it served as a zaviye[6] or any other religious institution.

There are various hypotheses on the origin of this multi-purpose building: Müller accepts the Roman castrum as its origin which was altered and adopted to the function and the needs of a caravanserai, noting the similarity between the central courts and corner towers of Persion caravanserais and the Roman castrum[7]. Erdmann relates the building to Sasanian origin giving the example of a certain caravanserai in Kasr-el-Hair-el-Garb, in Jourdan dating from the Eight Century A.D.[8] The Byzantine “kseneiodon” of which only the name remains is also said to be the antecedant of caravanserai[9]. Due to its development in Asia Minor it is even claimed to be a Anatolian-Seljukid invention[10]. A building type which shows so many variations, especially its examples in Asia Minor, is more likely to have more than one origin. This becomes even more clear if we consider the complex functions of a caravanserai and the use of the words "han”, “zaviye”, “ribat”, “hankâh”, “imaret” synonymously for this multi-functioned building type. However, leaving the question of origin aside, it would be reasonable to return to the ribats of Central Asia to trace the development of caravanserai. In Transoxonia, before the tenth century, we find the “ribat” or “rabat”, which served as a caravanserai as well, used as a means of defense against the invading Turks[11]. It is only normal that the same system was adopted by the Moslem Turks who overtook and used them as frontier forts for their troops ready for “cihad”. These were little forts, fortified by a wall which enclosed buildings, stables, kitchens, baths, and mescids[12] ; and they were financed by endowments bequested by the State or the rich. When the frontier expanded they lost their military function and were used for commercial purposes[13].

The Seljuks, who were already equipped with the tradition of caravanserai, brought it along from Central Asia to the Near East and developed this tradition into a state policy[14]. Anatolia, due to her position was always in the chain of commercial traffic, however, most of the trade in the Twelveth Century was transit in character, centered in such cities as Konya, Sinop, Kayseri[15]. Commerce was in the hands of Greek and Armenian merchants and Lesser Armenia played the most important role due to its strategy[16]. The Seljuks Sultans, starting with Kılıç Arslan II, realized the economic strategy of the country and oriented their policy and conquests accordingly : The state helped the resettlement of important ports such as Sinop, Antalya, and Alanya, encouraged big Anatolian merchants by easing the entrance and the exit of goods in and out of the country[17], applied state insurance[18], provided the security of the caravan roads[19], and made commercial treaties with other countries[20]. Partly due to these favorable conditions and partly due to the fact that the Syrian caravan routes w'cre no longer safe because of the Crusaders, Anatolia became the most important trade center in the Middle East at the beginning of the thirteenth century and continued to be so until the Mongol invasion.

The effect of this development can be witnessed in Anatolia by the fine network of caravan routes and the number of hans on them, between the dates of 1204 and 1288/9 and most of them belonging to the period between 1204-1246[21], and lying on the three main arteries of caravan traffic; that is :

1— The roads connecting the East to the West,

2— The roads connecting the North to the South,

3— The roads connecting the South-East to Istanbul[22]

Our subject Alara Han, constructed when the caravan traffic was at its prime, lies on the roads connecting the North to the South. It was an important artery due to military activities first directed to the Black Sea and then to the Mediterranean. It connected the capitol Konya to the port of Alanya, which had also become the royal winter resort after its conquest by Alaaddin Keykubad I. in 1221. This road was divided into four sections as[23] :

1 — Sinop-Ankara

2 — Ankara-Konya

3 — Konya-Antalya/Alanya

a) Via Beyşehir

b) Via Hatunhisar

4 — Antalya-Alanya

Starting from Alanya, Şarafsa Han[24] at about 15 km. distance was the first and Alara Han was the second, and Manavgat was the third stop. Here the road in the direction of Konya branched off. Köprü Suyu Han provided the stop between Manavgat and Antalya[25]. Thus Alara Han, due to its location, was actually on two routes, the one to Konya and the one to Antalya.

As rightly stated by Erdmann “while the picturesquely situated castle is mentioned by the travellers, the Han received less attention”[26]. No document concerning Alara Han could be found in the archives of the Vakıflar General Directorate[27]. Although our earliest source on the area is Evliya Çelebi in the XVII. Century, he only mentions the Castle, not the Han, as being in a very ruinous condition[28].

Erdmann mentions that O. Van Richter has visited the Han in 1816, that Tschihatscheff talks about an antique building in the location of the Han, and that Herberdy-Wilhelm refer to it as an handsome building of the Seljuk period giving reference to Lanckoronsky[29].

Although Sarre includes “Alara” in the map in front of his book[30], there in no mention of the Han. Riefstahl[31], who has not seen the Han refers to it as “the important Seljuk Han in Alara” as he discusses the caravan road between Alanya and Konya.

Yetkin, Ozergin, Kuban and Rice also mention it in their publications[32]. Works which include material on the Han start with the book of Fikri Erten[33] on Antalya. It includes a brief description, the date, and a sketch plan of the building (Fig. 18), noting that is has its own “speciality”. A touristic publication on Antalya[34] includes a completely wrong description of the building, written most probably without seeing it. Konyalı[35] briefly mentions the building, giving the inscription but not a description, only noting the uniqueness of the plan and the selective use of material.

The two works which deal with the Han more than the others are those of Lloyd-Rice and Erdmann. Lloyd-Rice[36] are the first to publish an architectural measured plan (Fig. 20) and a brief description, as well as some photographs. Erdmann’s[37] documentation is quite complete including description, photographs, and a plan (Fig. 19), which has some mistakes because it is a version of Erten’s plan corrected with the aid of his own notes and the drawings of Lloyd [38].

Although Alara Han has been mentioned and documented by several sources, the documentation on it is incomplete, not even including complete measured drawings. Nor the research necessary to place it in the panoroma of “Han Architecture” is yet done.

GENERAL DESCRIPTION

Situation :

Alara Han is within the boundaries of Alanya, Antalya. It lies on the left bank of Alara Çayı which reaches the sea about 30 km. from Alanya in the direction of Antalya. The Han is ca. 10 km. inland towards the North. The nearest villages to it are Alara (ca. 3 km to the South), Çakallar, (ca. one km. to the West), and Karakaya (ca. two km. to the East), on the opposite bank of the water.

At the present the Han is isolated, not lying on any roads. There exists a rough, narrow road passing through Okurcalar and Alara villages on which only a jeep can fare. Another possible approach is the service road of the irrigation canal which was under construction in 1966. The road passes through Çengelköy which is 45 km. from Alanya on the Alanya-Antalya highway and follows the canal ending at the head of it which is right opposite Alara Han, on the other bank of the water. There is no bridge on Alara Çayı to link the two banks, but, during the Summer, it is possible to cross the water since it only reaches waist-high[39].

The Han is picturesquely situated among the pine trees and the river (Pl. 2). The remains of other historic structures exist in the vicinity. Alara Castle[40], about 2 km. to the North, encircles a small steep hill and several ruined structures with its baileys (Fig. 1, Pl. 1). It was taken over after the conquest of Alanya and rebuilt and elaborated at the same time that the Han was constructed[41]. Right at the foot of the Castle, on the watershore, there are the remains of a domed structure which are said to belong to a bath by the villagers[42]. The remaining stone piers of the wooden Kemal Bridge stand between the Han and the Castle where the road coming from the direction of the Han passes to the other bank since the rocky cliff of the Castle does not give passage. At the head of the canal there are some stone ruins which give a slight indication of an arch form, and which therefore, may belong to another bridge[43]. Most probably several bridges were built on this section of Alara Çayı at different periods because the river frequently overflowed and changed its course, demolishing the bridges on it as well.

The rectangular Han structure is built on the narrow track of land between the river and the small hill to the East of it with the long sides of the rectangle laid parallel to the water; and the Han itself is oriented to the North (Pl. 6). The river passes about 50 m. from the west façade. At the present there is a cotton field in front of the building and a cemetery to the South of it. There is no possibility to date the cemetery since the loam deposited by the water has completely hurried the graves, only, looking from the right bank one can see that several strata of burial exists one over the other. The Han itself is embedded in the loam which has also penetrated into the interior of it, reaching a height of one and a half meters at certain sections[44].

Exterior :

Externally the Han measures 37.80 m. at its north and 49.90 m. at its west sides. Its three façades are constructed in cut stone rectangular almost isodomic masonry', whereas the east façade, which more or less leans to the hill, is constructed in rubble stone, (Fig. 2).

The north façade has the segmental arched portal opening in the center measuring 3.52 m. (Fig. 4a b, Pl. 2, 7). It retains its original height with some of its battlements still standing.

The two abutments on each side of the portal are rectangular in plan measuring 2.80 m. in width and 1.90 m. in depth. The one to the east of the portal is the more complete one reaching one course higher than the portal. It has two small windows located at 5.85 m. from the threshold of the portal, one on the north and the other one on the west side of it. These ogee archtopped openings light the landing of the staircase inside this abutment thus giving it the function of a tower more than an abutment. The section of the façade to the east of it is a solid wall only pierced with two slit windows. The section to the West of the portal has crumbled dowm including the upper part of the west abutment[45]. The crumble includes the west corner and even part of the west façade exposing the section of the superstructure of the spaces behind.

The west façade is abutted by three triangular buttresses dividing the wall into four sections (Fig. 5 a, b, Pl. 3). These sections are pierced by seven windows all together. The first three sections, starting from the North, have two windows each, and the last one has a single one. Only the uppermost stone of the first window from the North remains, due to the crumble. There are six gutters on this façade located in the course below the last. Again starting from the North, they are located thus: one on the south side of the first abutment, two in the second and one in the third sections, and one on the north side of the third buttress. No gutters exist neither in the first nor the last sections (the last two courses are missing in the last one).

The south façade is treated the same, only, being shorter, it has two buttresses and three sections (Fig. 6a, b, Pl. 4). The sections, starting from the southwest corner, have one, two, and two windows respectively; and three gutters, two in the second and one in the third section. There are no gutters in the buttresses.

The cast façade is not treated as a façade since it was not meant to be seen due to the already mentioned situation of the building (Fig. 7a, b Pl. 5). It is entirely built of rubble stone except the frames of five windows which are in cut stone. Three of these windows are near the south corner and the other two near the north comer but they are not visible from the exterior because they are completely embedded in the loam and the earth and stones fallen from the hill behind.

Interior :

The portal, (Pl. 11) upon entering, recesses forming a space 4.14 m. wide and 1.10 m. deep, and the back of the portal is spanned by another segmental arch similar to but higher than the outer one. The door[46] of the Han was fitted here and bolted by a big wooden bolt, as the two holes on the sides walls indicate. These holes are .20 m. high and .22 m. wide; the one on the west wall is as long as the bolt itself whereas the corresponding one on the opposite wall is only deep enough to secure the end of the bolt.

The portal leads to an open court which is the main distribution area (Pl. it, 12 Fig. 8) : An inner portal on the axis of the main one opens to the core which contains the guests’ quarters. The north corridor runs in fron of this portal in the east-west direction leading to the service rooms lined along the north wall of the corridor, and to the galleries. The two rooms directly opening to the court are the fountain on the east and the “mescit” on the west sides.

Fountain :

The fountain on the East is an open eyvan facing the court (Fig. 8, Pl. 13). It is the only specially treated and decorated space of the building. It measures 4.45 m. on its south, and 3.46 m. on its east sides, and is a step higher than the level of the court[47]. The floor is paved with large flagstones, little of which is visible under the loam bed. The fountain basin is built into the east wall with the tub placed in front of it. Apparently, in recent years, treasure hunters did excavations in this eyvan, digging under the tub and breaking its sides as well as ruining the pavement around it[48]. A staircase on the north wall leads to the roof of the building. (Pl. 14, Fig. 8). At the top of seven steps there is an opening .82 m. wide and 2.80 m. high (at its highest point), which is spanned by a false ogeé arch (Pl. 14, 15) caned into two stone blocks meeting at the midpoint. The stair continues beyond this opening winding in the west tower ending on the top level of the vault of the eyvan.

The superstructure of the eyvan is a star vault and the open end of the vault facing the courtyard is scaled with an arch that starts at the springing point of the vault, above a profiled console projecting .17 m. from the wall most of which arch has collapsed (Pl. 16, 17 Fig. 15), An interesting structural detail is used to pass from the non-square plan to a square one in the superstructure before the vault starts (Fig. 16, Pl. 14). This passage is achieved on the north wall, at the level of the last step of the stair, through a decorated frieze, as we may call it, by which the wall is recessed .42 m. to acquire the dimensions of a square 3.90 m. X 3.90 m.[49].

Room 2 :

The other room on the east side which gives its back to the north wall is entered from the corridor through a .90 m. wide arched opening. It measures 4.03 m. by 2.17 m. and is spanned by a pointed barrel vault in the east-west direction. Light enters through a slit window on the north wall which corresponds to the second one from the east on the west façade.

Room 5 :

The Room 3 on the west side of the court, opposite the fountain eyvan, is entered through a segmental-arched-opening adjacent to the north wall. It measures 3.40 m. by 3.35 m. The crumble in the north wall includes the north-west corner of the room. At the present it is filled with mud carried by the river up to 1.20 m. w'ith trees growing in the damp loam bed. There is a .40 m. w'idc niche on the north wall and the upper part of another niche, .70 m. wide and spanned by a flattish five centered arch, is seen in the south wall. (Fig. 2, 3, Pl. 20) The dimensions of it suggest that it starts from the floor level.

The room is barrel vaulted in the cast-west direction. There are no windows, it is illuminated through a rectangular oculus at the highest point of the vault, .10 m. away from the north wall.

Room 4 :

It is the room next to the Room 3 and entered from the west side of the north corridor. Its north wall has entirely crumbled. It is very similar to its correspondent on the cast side, Room 2.

Inner Portal :

The inner portal is treated exactly like the outer one (Pl. 7, 22). Its 2.72 m. opening is spanned by a .47 m. wide segmental arch. The recess between the two arches accomodating the door, including the bolting system is similar to the main portal. It indicates that the inner portal was also closed after dark as a second means of security.

Core :

The inner portal opens to the core of the Han reserved to the travellers. It is a combination of four rooms and three eyvans lined alternately on either side of a long corridor (Pl. 21 a, b, 22 Fig. 8, 13, 14). Apparently, this alternating scheme, starting and ending with a room, was to differentiate the spaces of living and sleeping. The doors of the rooms are located nearer the eyvan next to them, only the rooms on the south end have their doors near the eyvan before them. The size of the rooms are approximately 2,95 m. X 3.15 m. and the eyvans measure 3.23 m. X 3.04 m.

The corridor measures 26.48 m. on the east and 4.59 m. on its south sides, and it resembles an open ccourtyard. It is constructed in cut stone on the long sides whereas the short sides as well as the interiors of the rooms and eyvans are in rubble stone.

The rooms are entered through the usual arched openings. They are all barrel vaulted but, unlike the earlier plan drawings, they show variations in the direction of vaulting (Fig. 2, 13, 19, 20) : All the rooms on the west wing are vaulted in the east-west direction. On the cast wing, the first and third rooms are vaulted in the north-south direction whereas the second and the fourth rooms are vaulted in the cast-west direction, like the rooms on the west wing. All the eyvans are barrel vaulted in the north-south direction and are scaled off with cut stone arches, like the fountain eyvan. All the rooms as well the eyvans have a small window on their rear wall opening to the stables, which, most probably enabled the merchants to control their animals and goods and communicate with their servants. Besides, every room has an oculus in the vault.

Careful provision is taken for lighting the core ; There are four candle brackets along each long wall, placed at regular intervals, and one on the rear wall of each eyvan. There existed one on the south wall of which only the part within the wall remains. There is a cavity to the east of this broken bracket which is .60 m. long, .25 m. high, and a few centimeters deep. The regularity of the hole in the irregularly coursed rubble wall leads one to think that it once contained a cut stone piece intentionally placed in the middle of the wall which could either be an inscription or a reused stone put there for its decorative value[50].

There is no indication of floor height, the material of flooring of the court, nor of the rooms and the eyvans.

Galleries :

Besides the north corridor (Pl. 24, Fig. 12), there are two concentric galleries which envelop the core on the remaining sides. The inner ring of galleries constitutes the stables and the outer ring functions as the corridors serving the stables. The east (Pl. 27) and west (Pl. 25) corridors which run in the north-south direction start from the north exterior wall and meet the south corridor (Fig. 9, 10, n, 13, 14) which runs from the east exterior wall to the west one. The wall common between the stables and the corridors are pierced with eight arched arcades. Along the east and west corridors, starting from the north end, the first openings lead to the north corridor, the next five to the side stables (Pl. 26, 29), the seventh to the south stable (Pl. 28). At the south galleries, the outer first open to the side corridors and the inner six to the south stable (Pl. 30)[51].

All the galleries are barrel vaulted in the narrow direction. However they show variation with the varied use of bracing arches on the superstructure. The west stable is strengthened by three bracing arches in cut stone which correspond to the first, third, and fifth pillars of the arcade. They start over brackets just below the springing point of the vault. The vault of the east stable is supported by five arches which project from the west wall and form part of the pillar on the east side. The south stable has three such arches, the first two from the east are similar to those of the east stable and the third one, located at the next pillar, is similar to those of the west stable.

The galleries are lighted through the slit windows along the exterior walls as well as oculi in the vaults (Pl. 32, Fig. 17). Provision for artificial lighting is taken by the placement of candle brackets on the two sides of the arcades, only along the east arcade they are existing only on the stable side of it.

There are seven slit windows (two crumbled) and six oculi in the west corridor, five oculi in the west stable, five windows seven oculi in the cast corridor, six oculi in the south corridor, and six oculi in the south stable.

There are two candle brackets one on cither side of the north corridor, eight on the pillars of the west corridor, fourteen in the west stable (two on the north and south sides, four on the east wall and five on the west pillars), nine along the west side of the east corridor, ten in the east stable (two on the north and south sides, and six along the west wall and none on the pillars), seven on the pillars of the south corridor, and ten in the south stable (five along the north, wall, three along the south pillars, and one on each of the remaining sides).

DESCRIPTION AND COMPARATIVE STUDY OF THE INDIVIDUAL ELEMENTS OF THE BUILDING[52].

Portal :

The Portal of Alara Han is located in a recess created by the two abutment towers. It is spanned by a .62 m. wide segmental arch which is flush with the wall (Pl. 2, 7, 8, 9, Fig. 4 a, b). The springer stones of the arch are joined to the impost stones by means of a tongue- and-groove joint, the tongue projecting from the springer. Above the arch there is the inverted U shaped, profiled, and projecting frame of the inscription. It is 2.10 m. high, 2.40 m. long and .40 m. thick. The lower ends of the sides are finished in stylised lion heads, the jaws of which are now broken. The upper part of the frame is closed by a semicircular arch which has the same profile with the frame. The arch is smaller in diameter than the width of the frame so that it starts after the corners of the frame turn. It is missing now, but a voussoir of it was found by us in the loam bed under the arch of the portal (Pl, 10).

The portals of Seljuk buildings, especially those of the Hans are the most richly decorated elements. Although there are several types, the most frequent type of portal is outlined by a rectangular frame projecting from the main body and consisting of several ornamented borders. The portal receeds into a deep niche which is framed by an arch on the plane of the portal. The arched entrance lies on the rear wall of this niched recession. The passage between the two arches on the two different planes is obtained by a corbelled stallactitc half - dome. Many of the caravanserais have portals of this type or its variations, with or without the framed border. Aksaray Sultan Han, Kayseri Sultan Han, Altinapa, Akhan near Goncah can be given as a few examples.

The portal of Alara Han is neither similar to the prototype nor to its variations. It does have projecting sides but they do not create the recessed preperation for the portal. Although in Erdmann’s plan the space in between the towers is shown as closed[53], (Fig. 19) in reality there is no hint or trace to indicate any connection between the two. Besides, the presence of the window on the west side of the east tower proves (Pl. 2, 16, 17) that these tower-abutments are conceived as open in-between and not as preparations for a portal. The segmental arch of the portal is flush with the rest of the façade and is completely free of decoration. The portal is emphasized only by the raised height of the section between the two abutments.

The projected frame over the portal is quite an unusual treatment, because, usually the inscription is placed as a band under the stalactite half dome or around the outer arch of the portal. Among the few Hans with flat portals, we can mention the mescit entrance of Şarafsa Han and the portal of Eshab-i Kehf Han.

Inscription :

The inscription within the frame over the portal consisted of four marble pieces, but at the present the part within the arch is missing and the remaining three blocks are in situ (Pl. 9) [54]. It is written in a rather bold “Seljuk nesihi”. After several adjectives to glorify the Sultan who constructed the Han, Alaaddin Keykubad bin Keyhüsrev, the date is given as 629 H. 1231/2 A.D. The inscription was mentioned by Erten[55] and Konyalı[56] and the original content was published by Lloyd[57]. Erdmann[58] refers to the translations of Lloyd. The text of the inscription mentions the Sultan as the donor. Among all the Hans[59] there are only six which mention the Sultan as the donor. They are : Evdir Han by izzettin Keykavus, Aksaray Sultan Han and Alara Han by Alâaddin Keykubad I, incir Han and Şarafsa Han by Giyasettin Keykavus II, and Kırkgöz Han by izzettin Keykavus II[60]. That is Alara Han is constructed with the royal command at the same date with Aksaray Sultan Han.

Accordingly, we can conclude that, only these six royal buildings are the real “Sultan Han” s whereas the “Sultan Han”s are called so only due to their plan and size.

Abutments :

There are two kinds of abutments on the cutstone walls of the Han; the ones on the north wall are rectangular and the rest are triangular in plan (Pl. 2, 3, 4). The use of abutments are very frequent in Seljuk architecture, specially in buildings which are somewhat fortified, like the Hans. Usually several kinds are used in the same building; some are even ornamented, like those of Aksaray Sultan Han. The rectangular type is more common than the triangular one; its use in Sadettin Han, Durak Han, Karatay Han, Susuz Han can be given as examples. The triangular type is a less common one, being used only in Ertokuş Han, Çardak Han, San Han near Avanos, Şarafsa Han and Kargı Han. The use of this type of abutment besides the Hans could only be found in Tiirbes to which we can give the Gömeç Hatun Türbesi in Konya, Emir Yavtaç Türbesi in Kırşehir, and Kureyş Baba Türbesi in Boyalıköy as examples.

Roof:

The roof of the Han is reached through the stairs starting from the fountain eyvan and continuing in the west tower (Pl. 15). The landing on top of the stairs open to the top of the vault of the fountain eyvan. No indication of a pavement is seen over the vault, but merely the stones of the vault which are roughly finished (Pl. 19, 31). It is defined by a rubble stone wall which is only .50 m. high at the present. There is the opening of a possible doorway on the south wall leading to the roof. The roof cover is earth, levelling the difference between the different heights of the vaults. The fact that this levelling is intentional is proven by the frame stones of the oculi : the frames of the oculi of the lower vaults are of higher stones and those of the higher vaults are of lower stones to obtain the same level. The roof of the building is utilized as threshing grounds at the present. Most of the oculi are clogged due to this reason.

Not enough hans retain even their superstructure to be able to find about their roof covers. However, as far we can deduce from the remaining Hans as well as other buildings, they were either covered with earth, like Alara and Şarafsa, Hans, or paved with stone, like Sultan Han near Aksaray and San Han near Avanos; they might even be covered with bricks as can be seen in the Karatay Mescidi in Konya.

Battlements :

The battlements over the exterior walls of the building are built in rubble stone. Several of them remain over various parts of the building. They are all 1.10 m. high and .75 m. thick but their length varies from .90 m. — 1.20 m. They all were originally rendered white, as the remaining traces indicate (Pl. 33).

Unfortunately very few Hans retain their battlements to make a just comparison, however, judging from other types of buildings as well as Hans it would not be unjust to say that, usually, the buildings constructed in cut stone have cut stone battlements, as the example of Obruk Han, and if they are constructed with rubble stone, the same material is used for the battlements as well, Şarafsa Han being an cxampie of this type[61]. But this must not have been a rule because we know that the choice of materials depended on the importance attached to the building as well as the financial conditions, therefore an ashlar faced building could easily have rubble stone battlements, keeping in mind the fact that the rubble parts were rendered anyway.

Masons' Marks:

There are masons’ marks on most of the cut stones of the Han (Pl. 34 a, b). The same marks were encountered on the stones of Alara Castle which indicate that the same masons worked in the construction of the two structures [62].

The marks used in all Seljuk buildings are all very similar to each other being simple geometric lines or forms, and “the same marks can be observed in different periods and in different regions”[63] They are thought to belong to the guild of “Ahi Dervishes” which was involved in construction. However the appearencc of masons’ marks in Anatolia dates back at least to the Greek period. These marks were used by the workers or the quarry contractors to control the quality and the quantity of the blocks cut by a single or a group of stone masons[64]. There is no reason not to believe that the same use continued through the ages. The use of the same marks in different places docs not only, and necessarily, account for the fact that the groups of workers travelled from one job site to the other. The use of the same marks in different regions[65] can also explained by the fact that they were done with the same instruments used to shape the stones, which, due to their size, limited the use of a great variety of forms[66]. Hence the repetition of the marks is due to the possibility and the simplicity of the solutions. The fact that the same marks appear in buildings constructed at the same date at different places can be explained by this approach.

Use of Materials and Construction :

The little porous limestone of the area is the basic material of construction. It is used in cut stone rectangular masonry on the façades (excluding the east one), in the fountain eyvan, on the elevations of the courtyard and the core, the arcades of the galleries, the bracing arches, and the opening frames. The rest of the building is in rubble stone with a rare use of broken bricks in the superstructure.

The cut stone is not a mere facing but an indispensible element during the construction stage : The first course of stones was laid on both sides of the wall thickness desired, the rubble fill was placed in-between, and over it the lime mortar, which sometimes contained broken bricks, was poured. Therefore the cut stone faces acted ?s formwork during the construction and each stone course marked a stage of construction, as is detectable from the sections of the walls where the dressed stones are missing. The technique ’s the same when cut stone is used only on one side of the wall or even if the whole wall is contructed in rubble stone. This technique is no different than the Roman “opus caemcnticium” which was started to be used in I. Century B.C.

The type and the condition of the foundations cannot be detected in the present condition of the building. The exterior wall of the building are 1.60 m. thick whereas the inner walls are in varying thicknesses such as .68 m., .76 m., .82 m., .93 m., 1.02 m., 1.28 m.. The superstructure is all in rubble stone excluding the cut stone vault of the fo untain eyvan. All the arches and the vaults are two centered pointed however with different profiles. They are all spanned in the narrow direction, only in cases where the width and the length of the room are almost equal, changes in direction are seen. Segmental arches are used in the portals (Pl. 7, 22) and over the door opening of the mescit. False arches are found over the slit windows, and over the door arch in the east tower (Pl. 2, 3, 4. Fig. 4, 5, 6).

Most probably all the rubble parts of the building were once rendered, the evidence of which can be still seen on the cast façade on some of the battlements and in some parts of the galleries. The structural and constructional characteristics are true for all the Seljuk architecture : The use of cut stone in the important and seen parts of the building, as well as the important rooms, like the türbe, of the kiosk mescid, and in important structural elements; the use of rubble stone, rendered white, on the less important parts. However the ratio of cut and rubble stone depends on the importance of the building, that is its financement, besides the availability of the building material.

Condition of Fabric:

Alara Han has not had any alterations, additions, or repairs after the second superstructure of the core was constructed. Even in this state it is one of the structurally best preserved Hans. In fact it is among the very few which retain so much of its superstructure and original elements.

The south-east comer of the building has shifted from the vertical and tilted towards the outside. The whole structure shows a slight inclination towards the river. These two observable failures, as well as the crumbled condition of the north wall may be due to ground water, or the failure of the foundations due the extra load of the accumulated loam. The damage caused by the vegetation is clearly seen (Pl. 2, 11, 21). Weeds, bushes, even trees grow on the earth fill of the roof, on the loam bed, and in the mortar between the stones. The roots of the vegetation expand the joints between the cut stones or break the bond between the cut stones and the rubble fill, eventually causing the stones to fall. Examples to this kind of destruction are the fallen arch of the fountain eyvan (Pl. 11 ), various cracks on the façade (Pl. 21), and some cracks on the vaults. The loam bed which keeps the lower parts of the building moist causes the deterioration of the limestone. The stone used in the building does not show any deterioration due to aging. The only physical defect which is observed is in the crushed impost stone of the portal arch (Pl. 8). The mortar is deteriorated losing its binding quality, thus allowing penetration of water into the structure (Pl. 25, 27, 29). It is caused by the seepage of water retained by the earth on the roof causing a chemical reaction in the limestone. As a result calcium carbonate in the form of incrustation is deposited on the vaults, the arcades, and the walls[67].

Lighting :

The closed parts of all Seljuk Hans are very poorly lighted and Alara Han is especially dim due to its concentric planning. The sources of natural light are : the court in front of the portal, the slit windows along the exterior walls, and the oculi in the vaults, all of which are the common sources of light for all Seljuk building planned closed to the exterior.

There are twenty-one slit windows, three of which have crumbled (Pl. 2,29). On the façades, they are quite narrow openings each topped by a slightly pointed arch carved into the lintel stone. On the exterior they are, .84 m - .94 m. heigh to the top of the arch and .14 m. - .16 m. and .14 m. - 16. m. wide. They taper towards the interior and become, .78 m. - .84 m. high and .40 m. - .42 m. wide openings at an average height of 2.40 m. from the interior (original) floor height.

The oculi which add upto fourty seven in all the spaces of the Han are at the crown height of the vaults, but not in a regular location, as can be observed from the plan (Fig. 2, 3). They are square or nearly square rectangles, with the exception of the octagonal oculus of the fountain eyvan, and vary in dimensions from .25 m. - .50 m. They are framed with cut stone frames projecting over the vault heights. Oculi are frequently used especially in long, vaulted spaces, like the stables of the Hans. Çiftlik Han, Çakıllı Han, Hafiz Han can be mentioned among the Hans and the Arsenal of Alanya among other buildings.

Provision for artificial lighting has also been taken, in the form of brackets for candles or oil lamps which add up to 62 in number (Pl. 32, Fig. 17). The brackets are in cut stone, projecting, .21 m. from the wall, .22 m. long, and .23 m. wide and a triangular cavity on the upper surface holds the candle or the lamp. The front of each bracket is carved in the form of a lion head with an almost human expression. This detail is unique in Seljuk architecture as far as wc know. A detail for lighthing is only seen in the mescid of Ishakli Han in the form of a simple cavity.

The use of lion, as well as other animal figures, is frequent in Seljuk art, in the ornamentation of various parts of the building, in the form of gutters etc. We can give the example the Medrese of Huand Hatun in Kayseri for this use of the lion motive but these lions have gargoyle-like expressions. Konyalı mentions a similar lion head in the castle of Alara which we could not locate, however, the gutters in Niğde Alaaddin Mosque and a gutter facing the courtyard of Çarşı Camii in Siirt have exactly the same expression with those in Alara Han.

Services :

It is apparent that the amount of services that a Han provided for the travellers depended on the size and importance of the Han and the road it is on as well as the richness of its Foundation Charter. The few Foundation Charters that have reached us and belonging to the big Hans show that money was put aside for heating and cooking[69]. However, wc can rarely denote the exact location of the kitchen even in these Hans, although it is generally accepted that the rooms on the sides are the portal were for administration and services. Very few have areas especially denoted for bath and W. C. and fountain. In big Hans, like Sultan Han near Aksaray, Sultan Han near Kayseri and Ağzı Kara Han, these spaces are among these surrounding the courtyard. In Ertokuş Han it is said to be under the kiosk mescit.[70] In general, however, these services were located in a service building at a little distance from it. At present only very few of these service buildings remain, as that of Kızılören Han.[71] The ruins said to be a bath at the foot of the Alara Castle may very probably belong to such a service building serving both the Castle and the Han.

The Fountain :

The fountain occupies a room by itself in Alara Han (Pl. 11, 13, 14, 15, 16. Fig. 2, 15). The water container on the east walfis 1.63 m. above the pavement level, 1.02 m. long, .42 m. wide, and .35 m. high. It has an opening in the middle, which is .50 m. long, .40 m. high, is as wide as the basin itself, and is crowned with a half dome. The inner face of the half dome is carved with flutes radiating from the lower center towards the sides thus creating a shellikc form. There are remnants of four bulging forms, about. .10 m. in diameter, two of which are above the basin near the upper corners and the other are located .10 m. below the basin, corresponding to the upper ones. A tub carved from a single block of limestone is placed directly under the basin. In its outer dimensions, it is 1.75 m. long, .90 m. wide and .30 m. thick. It is not possible to determine the original height since the sides broken.

The source of fountain can cither be the river or a spring on the hill at the east of the Han. The existence of the built-in-basin suggests the existence of a spring source, otherwise, it had to be filled with carried water, which seems less reasonable. A third possibility is an information given by a peasant who had worked in the cleaning of the loam in the Han, undertaken some ten years ago by Antalya Vakıflar Müdürlüğü. He mentioned the existence of a waterpipe which started from the fountain chamber, and passed under the chambers of the core, going out from the south-west corner of the building. The slope of the land in that direction towards the river is rather sweet, but whether it is sweet enough to enable the flow of the river water in the direction of the Han is quite questionable. The water-pipe must be for drainage rather than to feed the fountain. Sondages for the water pipes must be done as well as some others in the direction of the hill to look for a water conduit from a possible spring.

Certainly all Hans were provided with a water source, be it a well, a spring in and outside the Han, or it was located right beside a river: Altmapa Han had a spring, Dokuzun Derbent Han had a well, and Çekcreksu Han was at a distance of 30 m. from Çekereksu. Alara Han most probably had two sources, the river and a spring. In a few Hans the fountain is a separate space beside the other services or there is only fountain and not the other services within the Han, like the example of Alara. San Han near Avanos also has a fountain eyvan facing the court, Kızılören Han has a fountain on the façade, most probably under the mescid[72], to the left of the portal.

Star Vault:

The vault of the fountain eyvan (Pl. 16,19, Fig. 2,15) is a variation of the so called “star vault” which, in principle, is an elaborated cloister vault: it is built over an exact square and finishes with an octagonal keystone. The corners of the octagon are congruous to the corners of the square and the midpoints of the sides. The diagonals of the square, broken into three lines created by two planes converging towards the middle, meet at the corners of the octagon. The monolythic keystone is perforated with an octagonal oculus in the middle and it is decorated with triangles in different planes.

Decorated vaults are widely used in Seljuk buildings; sometimes different types are used in a single structure, as the case is in Divriği Ulu Camii. The use of this specific kind of vault is found in two other Hans: in Karatay Han it is used over the “türbe”, and in Sultan Han near Kayseri it is used over the space in front of the entrance portal. In other buildings besides Hans it is used in Sivas Gök Medrese, over the entrance eyvan, in Erzurum Çifte Minardi Medrese, over the side eyvans. In Divriği Ulu Camii, various examples of the star vault is used in the superstructure[73]. All these examples are in fine cut stone and constructed without any jointing, and all are built over spaces of importance, as in our example in . The star vault of Alara Han is the only one among them which is open at the center, due to the function of the space it covers.

Stairs :

There is a single staircase in Alara Han, the one in the fountain eyvan leading to the roof (Pl. 15). The first part of it is .82 m. wide composed of seven steps which are partially built into the thickness of the wall, only .35 m. of it cantilevering from the wall. The riser height is .34 m. and the tread width is .32 m. The cantilevering sides of the steps are carved, so that an embossed diagonal and border, on the two sides forming the riser and the tread emphasize the outline of each step. The rest of the stairs continue in the east abutment in winding form.

In Seljuk Hans stairs are commonly used either to climb to the roof or to the mescid when it is on the first floor, be it over the portal or in the form of a kiosk mcscid. The ones leading to the roof are humbler than those of the mescids, like that of Karatay Han and San Han near Avanos. Ağzıkara Han, Kayseri Sultan Han, Ishakh Han have ornamental staircases in their kiosk mescids. The stairs of Alara Han have the sculptural character of the later examples being located in the only sculpturally treated and ornamented space of the Han.

Mescid :

The mescid of the Han is Room 3, to the west of the main portal. The already described niche on the south wall is the only one in the Han and it is oriented to the kibla direction. This fact denotes the function of this room as the mescit. Lloyd – Rice[74], who probably did not notice this niche, claim that the mescid is the small space over the east tower. Looking at the plan included in their book, (Fig. 20) such a possibility seems probable because the stairs leading to the roof are shown as built into the thickness of the wall. In fact, the winding part of the stair is partially built into the thickness of the tower and partially the wall, leaving only a landing, therefore, there is not space enough to accomadate a mescit there (Fig. 3. Pl. 31). The room in front of this landing, which is already mentioned, has a single opening in the south direction, leading to the roof, exactly in the location that a mihrab would be, leaving the possibility out of question.

A special space denoted to praying does not exist in every Seljuk Han, nor is there a specific location in the plan in the cases in which it exists. Some of the bigger Hans have seperatc structures, the kiosk mcscids, in the middle of the courtyard. These small structures raised on four arches have great sculptural value. The mcscids at the Aksaray Sultan Han, Kayseri Sultan Han, Jshakh Han, and Ağzıkara Han are of this type. In some cases, a structure similar to a kiosk mescid is incorporated to the main building, like the mescid of Kizilören Han which is located outside, to the left of the main portal but entered from the interior. In several examples, the mescid is situated over the main portal, like Sadettin and Altinapa Hans (the entrance to the latter is from the exterior through the stairs on the left of the portal). In one group, the mescid occupies one of the rooms on either side of the portal, to which Alara Han belongs; in Erlokuş Han it is again to the right of the portal, the domed room after the two rooms next to the portal. Yet in another group it is placed to the left of the inner portal, as in Ak Han near Goncah and Durak Han. Şarafsa Han is an example of its own with an independent mescid constructed adjoining the Han with its independent portal. In a few cases there is a seperatc mosque structure near the Han, like Eshab-i Kehf Han.

These examples also show that the location of the mescid depends on the planning of the particular building and not on an accepted location. However the existence of a mescid indicates to a special consideration for a specific function in the stage of planning and construction.

The Core Corridor :

The corridor has always been considered as open by those who described or studied Alara Han up to now : Lloyd-Rice mention it as an “open corridor[75]”, and Erdmann[76], in his description uses the word “court” but docs not mention it as being closed. There are several obvious traces in the structure which proves the corridor as being closed one[77]. In fact it was closed by two different systems at two different periods : its original cover was a barrel vault in the east-west direction, the remnants of which can be seen on the north wall of the corridor, over the portal (Pl. 22). The originality of this superstructure is backed by the fact that the facing stones of the south wall of the front court, to the back of this wall extend higher than the sides, thus facing the end of the vault (Pl. 12). Also enough remains on the south wall of the corridor to trace the springing point and start of the curve of the vault (PL 21 a, b). The indications for the second cover can be detected over both of the long walls of the corridor (Pl. 21 a, b). The traces over the cast wall, near the portal, are especially clear (Pl. 23, Fig. 8). Here we see the remains of a series of end walls supporting narrow vaults. This superstructure does not start regularly over the last course of cut stones; at some sections it starts on the course below the last. This proves that it was constructed after the first roof cover had crumbled down destroying the upper courses of the face stones as well. No trouble was taken to even the courses before constructing a new cover. The form of the remains cither suggest a scries of barrel vaults in the narrow direction or a single vault with lunettes in the long direction. No traces of arches to support each small barrel vault is seen[78], therefore one comes to accept the vault with lunettes as the more probable possibility.

It is very hard to decide the period of the second superstructure because it is built with rubble stone very similar in character to the original rubble stone masonry. However, vault with lunettes is very peculiar to Seljuk architecture, therefore it would be safer to assume it to belong to a later period.

Plan Type :

The only classification of Anatolian Seljuk caravanserais has been done by Erdmann[79] who classified them according to the existence of a “hall”, its direction of vaulting, the existence of a court, and its relation to the hall. Alara Han belongs to the group “C”, that is “Hans Without Halls”, together with Evdir, Kargı, Kırkgöz, and Eshab-i Kehf Hans. Among these belonging to the same group, Alara Han, Eshab-i Kehf Han[80] (Fig. 21 a, b.), and another Han which does not exist in Erdmanns catalogue, Mama Hatun Caravanserai in Tercan[81] (Fg. 22), form a group by themselves due to their concentric planning.

A simple comparison among the three would further classify the position of Alara Han :

Location — Mama Hatun Caravanserai in Tercan, lies on the East-West Road, Alara Han on the North-South Road, whereas Eshab-i Kehf Han does not lie on any of the trade routes, therefore does not serve to commercial ends. It makes part of a group of buildings consisting of a mosque, ribat, and a han which served the pilgrims who came to visit the sacred cave in the mosque. We can conclude that the plan type is not due to a regional nor to a commercial influence.

Date — Alara Han is the only one among the three Hans which can be surely dated 1231/2, due to its inscription panel. The Mausoleum across Mama Hatun Caravanserai is dated to late 12th Century or to early 13th Century[82]. It can be thought that the Han is built around this date but most probably later. The group of buildings in Eshab-i Kehf were built by the Governor of Maraş, Nusrat al-din Hasan bin Ibrahim who was kiled in 1234; the mosque near the Han is dated H. 612 (1215 A.D.) which means that the Han was built at a date about the construction of the mosque and not later than 1234. All the three Hans, according to our assumptions, are constructed in the first half of the thirteenth century, and Alara Han is the latest among them.

Size— Mama Hatun is the largest, since it measures 50 m. X 50 m., Alara Han 50 m. X 39 m., and Eshab-ı Kehf Han, 28 m. X 20 m. The sizes may have been indirectly influenced by the importance of the road it lies on, and the traffic on it. Eshab-i Kehf Han is the smallest due to the its special function, whereas, Mama Hatun Caravanserai, being the one on the more important trade route, is the largest and double storeyed as well.

Planning — All the three are similar in the concentric planning around the core, the stables in the periphery and the rooms in the center. In Alara Han the core is encircled on three sides, in Mama Hatun on the lateral sides, and in Eshab-i Kehf on one lateral side omy. In Mama Hatun there are rooms on the three sides of the core with the eyvans located near the west of the core. The center of the core is an open court as the size of it indicates. The same is true for Eshab-i Kehf although its court is humbler in scale. Thus Alara Han is the only one among them with a closed core corridor.

Connection Between The Stables and The Core — The communication in Eshab-i Kehf Han is provided from two of its eyvans through doors, corresponding to the windows of Alara Han. In Mama Hatun no connection exists on the ground floor but there is a vertical connccention between the rooms on the first floor and the stables on the ground floor[83]. The connection between the rooms and the stables might have also served for the heating of the rooms as well as its obvious function of communication, if we consider the long existing arrangement for the heating of the village house, with the stables placed under the houses.

Constuction — Both Alara Han and Eshab-ı Kehf Han are constructed on a hill side, leading to it. Their construction and material use is similar. Mama Hatun Caravanserai has more cut stone used in the construction which shows that it was better financed than the others, due to its more important location.

Examples Outside Anatolia :

In fact, the discussed type of Han is quite rare among all caravanserais, including those outside of Anatolia. As a result of a research through all the documented Hans only two more Hans, with similar planning to these three in Anatolia, could be traced. The first is “Taşrabat”, (Fig. 23) on the road between Kashgar and Narysoje[84]. It is a stone building with a big stone dome covering the court, reminding of Anatolian Danishmcndid Medreses. It is though to be a ncstorian convent by Diez[85], with Buddist influence in its planning. The only similarity between Taşrabat and Alara Han is in the concentric planning of one section of the building.

The other example is a ribat in “Senchas”, Persia[86] (Fig. 24). It is similar to Mama Hatun Caravanserai in its plan layout with rooms lined around a large central court and the stables on the lateral sides, and two storey rooms on the portal side. It is constructed in rubble stone, unlike the usually brick Persian ribats.

Both of these buildings are undated, therefore whether these buildings are the forerunners of the examples in Anatolia is impossible to decide. If we add up the possible Christian, nestorian, and buddist influences the Turks might have had to the already discussed complex problem of function and origin of caravanserais it becomes even harder to answer the question. But it will be more true to accept that the examples in Anatolia are more likely to be influenced by the East than accepting the development of this plan type as “paraphasc over the South Anatolian Court-Han with the strongly reduced court”[87]. as Erdmann states, that is, due to climatic conditions.

In all these examples in and outside of Anatolia one factor is common to all : the special care and importance given to single rooms, to the extend of differentiation between the spaces of living and sleeping. Such a treatment leads one to think that the rooms are the more important elements of the plan; that is, they were used for durations longer than a normal traveller could afford to stay. This actually is true in the case of Eshab-i Kehf which does not serve to commercial ends anyway.

In fact Eshab-i Kehf and Taşrabat are resembled to medreses by Özgüç-Akok[88] and Diez[89]. Kuban thinks that these buildings might have caused the development of “zaviye”s. But the complexity of the functions of Hans and our scarce knowledge on Seljuk life makes one refrain from pointing out a specific factor which determines this particular plan type.

Judging from the comparison, Alara Han has a very distinguished plan type, most developed even among the examples of its own group with all the spaces differentiated according to their uses. It further differentiates within its own group because of its closed core corridor instead of the open court in the others.

Footnotes

  1. The material included in the article is a summary of part of the study done for a restoration thesis on Alara Han. It was prepared as the final requirement for the “Scuola di Perfezionamento per lo Studio ed il Restaura del Monumcnto”, Faculty of Architecture, University of Rome, under the direction of Prof. Guglielmo De Angelis D’Ossat and was presented in July 1967.
  2. Osman Turan, “Selçuklu Kervansarayları”, T. T. K. Belleten, vol. X, Ankara, Temmuz 1945, no. 39, p. 490. The famous vizier of the Great Seljuks, Nizam-ül-Mülk speaks of the construction of caravanserais as one of the duties of the State, same as building bridges, roads, etc. and points to their military function saying that they should be used as sleeping places when the army is on the move. In fact Karatay Han was used for this purpose by the Seljuk armies on their cast-west movement.
  3. Turan, ibid. Karatay Han had also provided hospitality to the Mamluk Sultan Baybars and his army.
  4. Ibid. Ibn-i Bibi relates us that the Seljuk army was prisoned in the Hans after İzzettin Keykavus lost the battle to the Eyyubids.
  5. Aksarayli Kerimeddin Mahmud, Müsamerat-al Ahyar, (M. Nuri Gençosman - F. N. Uzluk), Ankara, 1043,3. 336, Aksarayî mentions a certain Turkish Bey, Ilyas, who revolted to the Mongol ruler trencin and took refuge in Aksaray Sultan Han.
  6. Turan, op. cit. p. 491.
  7. Karl Müller, Die Karavanserai in Vorderen Orient, Berlin, 1920, p. 64.
  8. Kurt Erdmann, “Bericht Über Den Stand Der Arbeiten Über Das Anatolische Karavansaray Des 13. Jahrhunderts”, / Atti del Secondo Congresso Intemazionale di Arte Turca, Venezia, 26-29 Settembre, p. 75.
  9. Dogan Kuban, Anadolu Türk Mimarisi'nin Kaynak ve Sorunları, İstanbul Teknik Üniversitesi, İst. 1965, p. 159.
  10. Erdmann, ibid. Erdmann also, refuses this hypo thesis.
  11. Kuban, op. cit. p. 157- 158. Kuban also relates from Barthold that in the X. Century there was one in each village and their number reached up to a thousand only around Bukhara.
  12. Fuat Köprülü, ‘Ribat’, Tarih Dergisi, Istanbul Üniversitesi, vol. I, Ankara 1942, p. 268.
  13. Turan, op. cit.
  14. See footnote 2.
  15. Tamara Talbot Rice, The Seljuks in Asia Minor, London, 1956, p. 103.
  16. G. Şahabettin Tekindağ, “Alaüddin ve Halefleri zamanında Selçuklu Küçük Ermenistan Hudutları”, Tarih Dergisi, İstanbul Üniversitesi, Vol. I., p. 30.
  17. Turan, op. cit. p. 473.
  18. Turan, ibid.
  19. Turan, ibid.
  20. Gıyasettin Keyhüsrev had made a treaty with the Venetians after the conquest of Antalya (Kemal Özergin, Anadolu Selçukluları Çağında Anadolu Yolları, İstanbul Üniversitesi Tarih Bölümü basılmamış Doktora tezi, no. 2797), We also learn from an Italian source that Alaaddin had made a commercial treaty on the Syrian traffic with the Venetians in 1219 (Giuseppe Stefani, L'Assicurazione a Venezia dalle Origini alla fine della Serenissima, Bologna, 1956, vol. 2. p. 6 m).
  21. The first dated Han is Kızılören Han (1204 A. D.); if we exclude the now non-existent Han near Aksaray erected by Kılıç Arslan II. mentioned by Turan (p. 34). The last Seljuk Han is that of Cebül Mücahit in Çay (Özergin, op. cit.). Erdmann (Erdmann, op. cit. p. 33) thinks that Argıt Han is the first Han which is mentioned by Turan, dating it to 598 H. or 1201/2 A. D..
  22. Özergin, op. cit. The centers of commerce surrounding Anatolia in the XIII. Century A. D. were thus distributed : / a) At the East, Tebriz in Iran, / b) At the South-East; Bağdat in Irak and Haleb in Syria. / c) on the Mediterranean : the ports of Ayaş, Alaiye, and Antalya, / d) On the Aegean : the ports of Ayaslug, Izmir and Focia, / e) At the North-West; on the water passages Istanbul, / f) On the Black Sea : the ports of Sinop, Samsun, and Trebizond.
  23. Özergin, ibid.
  24. It is also called Şarapsu, Şerefsa and Şarafsa.
  25. Özergin, ibid.
  26. Kurt Erdmann, Das Anatolische Karavansaray Des 13. Jahrhunderts, Berlin 1961, Band I, p. 184.
  27. İ. Hakkı Konyalı, Alanya Tarihi, Istanbul 1946, p. 271. Konyali mentions a certain “il yazıcı defteri” (governmental account book), registered in the Prime Ministery Archives in Istanbul, no : 166 But this source only mentions the village and the castle.
  28. Konyali ibid.
  29. Erdmann, ibid.
  30. F. Sarre, Reise in Klein Asien, Berlin, 1895.
  31. Rudolf Riefstahl, Cenubi Garbi Anadolu Türk Mimarisi, (Tercüme Cezmi Berktin), İstanbul, 1941, p. 50.
  32. Suut Kemal Yetkin, “Selçuklu Kervansaraylarının özellikleri” Milletler Arası I. Türk San'atları Kongresi, Kongreye Sunular Raporlar, T. T. K. Bas., Ankara 1959, p. 401. Suut Kemal Yetkin, İslam Mimarisi, Ankara 1965, 3. edition, p. 134 Kuban, op. cit. p. 159. Rice, op. cit. p. 296. Özergin, op. cit. özergin, “Anadolu Selçuklu Kervansarayları”, Tarih Dergisi - İstanbul Üniversitesi, vol. 20., pp. 144-145. However, since all these sources are at least published after Erten, they most probably used the earlier documentations as sources.
  33. Fikri Erten, Antalya Vilayeti Tarihi, İstanbul, 1940, p. 79.
  34. Turistik Antalya, Basın ve Yayın Genel Müdürlüğü Neşriyatı, Ankara p. 115.
  35. Konyalı, op. cit. pp. 367 - 371.
  36. Seton Lloyd-D. Storm Rice, Alanya, (Tercüme: Nermin Sinemoğlu), T. T. K. B. Ankara, 1964, pp. 51-53, 74 - 75. Levha XI a, c.
  37. Erdmann, op. cit. Band I, pp. 184- 185, Bailage Blatt 9, Tafel XXXII, Band II Abb. 343 - 348.
  38. The contraversial points of these sources will be discussed in the text later. The mistakes in the measured drawings can be observed by comparing them.
  39. We used the latter approach due to die sleeping and transportation facilities offered to us by the contractors of the canal during our twelve day stay on the site. Here I would like to thank Hamdi Görkay for his hospitality and my father A. Rıza Tükel for his help in taking the measurements.
  40. Turistik Antalya, ibid.
  41. Lloyd-Rice, op. cit. p. 5 For a detailed account see İbn-i Bibi, Anadolu Selçuklu Devleti Tarihi, (Nuri Gençosman — F. N. Uzluk), Ankara, 1941, p. 99,
  42. Erten, ibid. Erten mentions a bath but locates it within the castle which seem quite unreasonable.
  43. Erdmann, op. cit. p. 184. Herberdy-Wilhelm describe the Han as seen through a characteristic wild valley and “through the daring arched bridge which has thick walls and beyond the bridge there is the castle on the top of the hill. On the hill, in a wonderful forest you see the Han”.
  44. Photographs taken in August 1966 and March 1967 show that the loam deposited in one season is as high as 20 - 30 cm. The loam deposited in the Han had been once cleaned more than ten years ago by the Antalya Vakıflar Müdürlüğü.
  45. Although the original height of these abutments İs not definite, the east one seems to be complete lacking only its battlements.
  46. We do not know how the door was since no timber millwork has remained in any part of the building.
  47. The actual height of the step is concealed by the loam bed, but it must be ca. 30 m. high, as can be deducted from the section drawings.
  48. Lloyd - Rice, op. cit. Pl. XI a shows the tub intact. Since the English original of this book was published in 1958, the damage must have been done after this date.
  49. Such a treatment is unencountered in any other Seljuk building.
  50. Konyalı, op. cit. p. 371, mentions the use of re-used stones in the construction of the Han. Although searched carefully, no such stone could be located. This missing stone, if it were a re-used piece, must have been the only one.
  51. Lloyd-Rice, op. cit. fig. 23, In their plan, the pillars of the east gallery and the center pillar of the south gallery are shown as having T plans. At the present several pillars seem to have a T plan due to the fact that the loam bed has coveied them even above the springing line of the arches. Plan measurements taken by a horizontal reference line showed that these pillars are rectangular in plan.
  52. All the publications on Seljuk art have been used in this section as well as personal first-hand observations, therefore, instead of giving reference to each source mentioning the monuments or items concerned, only some very specific ones will be noted.
  53. Erdmann, op. cit. Band II, Tafel XXXII.
  54. According to an information given by a peasant, the missing part of the inscription is on the wall of the mosque in Alara village. This information is unchecked.
  55. Erten, op. cit. p. 79.
  56. Konyalı, op. cit. p. 371.
  57. Lloyd-Rice, op. cit. p. 74. / “1-263,... En büyük.?, Sultanların büyük Sultanı, milletleri boyunlarının maliki, Arab ve Acem Sultanlarının Efendisi, Hak Sultam, cihanın beldelerinin fatihi, karanın / 4 — denizin, Rum’un Şam’ın, Ermeninin, Frenç’in sultanı, / 5 — ‘Ala’ ud-dunyâ vad-din Kaykubâd b. Kayhusrav b. / 6 — Kılıç Arslan, “mü’minlerin, emirlerin burhanı, tarih, sene 629”.
  58. Erdmann, op. cit. p. 187.
  59. Ermann has catalogued 59 Hans with plans, 8 buildings which were built on the remains of a Han, 8 which were heard but not explored by him. Besides he traced 11 hans belonging to an unknown date. (Erdmann, op. cit.) özerging has able to trace 35 more Hans, and noted 19 more which he heard of but did not study. Besides, he mentions 32 places W’hich have names relating to “Han” (Özergin, Anadolu Selçuklu Kervansarayları pp. 144- 170) He, in his thesis, says that there were market places around the Hans which later developed into settlements. We can accept that there arc at least 120 Hans existing in our day. Among these less than thirty are inscribed.
  60. Erdmann, “Bericht Über Den Stand....”, p. 76.
  61. Only Alara, Şarafsa, and Obruk Hans still have their battlements.
  62. See footnote 41.
  63. Kuban op. cit. p. 101.
  64. For more information see : Roland Martin, Manuele d'Architecture Greque, Paris 1965, p. 223.
  65. The repetition of the same marks can be observed in the documentation given in the following sources : / Albert Gabriel, Monuments Turc d'Anatolie, Paris, 1931-34. Albert Gabriel, Voyage Archaeologique Dand la Turquie. Orietale, Paris 1940. Kurt Erdmann, Das Anatolische
  66. There are very few curvilinear masons’ marks which are very simple.
  67. This inscrustation has helped the structure stand up by acting as a strong bond in place of the mortar which had lost ist binding quality.
  68. Konyalı, op. cit. p. 370.
  69. See the following sources : Osman Turan, “Şemsettin Altınapa Vakfiyesi ve Hayatı”, Belleten, T. T. K. Ankara, vol. 42, 1947, pp. 197 - 236. Osman Turan, “Mübarizeddin Ertokuş ve Vakfiyesi”, Belleten, T. T. K. Ankara, 1947, vol. 43, PP· 415 - 430. Osman Turan, “Celâlettin Karatay, Vakıfları ve Vakfiyeleri", Belleten, T. T. K., Ankara, 1948, vol. 45, pp. 17- 170.
  70. Erdmann, op. cit. p. 51.
  71. Erdmann, op. cit. p. 48.
  72. Ibid.
  73. Later examples of the star vault are seen in Mardin Isa Bey Medresesi, Mardin Kasımiye Medresesi, the passage on the north side of the courtyard of Mardin Ulu Camii, Mardin Bab-as-sör Camii, silâhtar Mustafa Paşa Hanın Malatya.
  74. Lloyd-Rice, op. cit. p. 51.
  75. Ibid.
  76. Erdmann, op. cit. p. 185.
  77. Other observations such as the abundance of candle brackets and the great quantity of rubble in the corridor also back the evidence.
  78. Traces may appear when the vegetation is cleaned.
  79. Erdmann, op. cit. pp. 21 - 22 , Özergin, ibid.
  80. See the following sources : Erdmann, op. cit. Band I. pp. 187 - 188, Band II, Tafel XXXII., Mükrimin Halil Yınanç, “Elbistan” maddesi, İslam Ansiklopedisi, 1957, vol. 4, pp. 223 - 230. Tahsin Özgüç - Mahmut Akok, “Afşin Yakınındaki Eshab-ı Kehf Kulliyesi”, Yıllık Araştırmalar Dergisi, 1, Ankara 1957, pp. 77 - 94. Özergin, op cit., p. 149, Özergin, tez. Kuban, op. cit. p. 149. Rice, op. cit. p. 206.
  81. See the following sources : Özergin, op. cit. p. 157. Özergin, thesis Rice, op. cit. p. 101, fig. 7 Pegolotti, La Practica della Mercatura, A. Evans, Masachusetts 1936, p. 39. Evliya Çelebi, Seyahatname, II, IX, Ist. 1314 - 1935. Abdurrahim Şerif Beygu, Erzurum, 1939, p. 2G1, W. Taeschner, Das Anatolische Wegenetz nach Osmaanischen Quellen, 1 - 2 Leipzig, 1924 - 26.
  82. Suut Kemal Yetkin, “The Mausoleum of Mama Hatun”, Burlington Magazine, XCIX, p. 147. The same article is published in Turkish and English in Yıllık Araştırmalar Dergisi, I, Ankara, 1957, pp. 75 - 79.
  83. Verbal information from architect Yılmaz Önge, from the General Directorate of Vakıflar, Ankara.
  84. See the following publications : Ernst Diez, Die Kunst Der Islamische Völker, Berlin 1915, pp. 99 - 150. Ernst Diez, Die Globus, 1904. Joseph Stryzugowski, Die Baukunst Der Armenien und Europe, Band I, Wien, pp. 650 - 651. Kuban, op. cit p. 159.
  85. Diez. Die Kunst...., p. 99. Diez also thinks that Tajrabat is the origin of Persian and other Islamic medrescs comparing it with “Hasan Madrasa” in Cairo.
  86. See the following publications : / Müller, op. cit. p. 30. / Ernst Diez, Chtirasanisehe Baudenkmäler, Band I, Berlin 1918, p. 85.
  87. Erdmann, op. cit. p. 187.
  88. Özgüç-Akok, op. cit. p. 84.
  89. See footnote 85.
  90. Kuban, ibid.