Roman Baths: July 10-13

For more than a decade, research in the Roman Baths has been made possible thanks to the generosity of Louis and Fanny Lamberts Van Assche and their four children.

With their upper level 6,825 square meters in size, the Roman Baths at Sagalassos formed the largest building complex in the city. It occupied a natural hill to the east of the lower agora, which was leveled at its top around A.D. 120 after all of older structures had been removed. Its load-bearing capacity was extended towards the west and southwest by the construction of at least six brick-faced concrete vaults of about 100 square meters each. Together with the natural hill, this created a surface that could carry the actual bath complex, which seems to have had straight walls on its north and east sides, but "dented" ones towards the west and south. The north wall was aligned with the main east-west street of the city, and its main entrance was probably located there. The east and south facades were mostly built on the natural hill, except for the western half of the south face, where a vaulted substructure with at least two large arched windows extended the hill in that direction.

The facades towered over the immediate vicinity and must have been visible from most places in the lower city. The nearly 16 meter high west facade was separated by a paved street--covering a large gutter and built simultaneously with the baths--from the back wall of the Lower Agora's east portico. The northern section of the west facade contained two doors and its southern "dented" wall segment had a rectangular window, which together with the arched opening in the adjoining south wall illuminated the vaulted room in the southwest corner at ground level. The first door from the north was blocked in late antiquity and the room behind it filled with debris, possibly the result of an earthquake that occurred around A.D. 500. The second door opened to a vaulted vestibule, pierced toward the north and the south by an arched opening leading to the vaulted corner rooms of the west facade. The northwest corner room may have housed the original latrina, or public toilet, of the complex, whereas the room to the south probably became a latrina only after the catastrophe of A.D. 500. The other vaulted spaces at ground level apparently did not fulfill any specific function other than creating a large supporting surface for the bath structure above them. The eastern end of the vestibule contained a staircase leading through a large vaulted opening into a shaft-like, possibly open-air central structure that passed through all levels and was covered with marble. The structure has been only partially excavated, but its principal function seems to have been to connect the various levels of the whole complex.

The rooms of the actual baths were organized around this structure. All of the rooms must have been covered with brick-faced concrete vaults that were supported by massive piers starting inside the vaulted subterranean rooms and piercing vertically throughout the building. The curtain walls in between them where either made of solid brick, brick-faced concrete, or opus vittatum; mortared rubble wall sections alternating with horizontal bands of brick. According to a nearly 6 meter long inscription with huge characters, the central room was opened in the spring of A.D. 165 and dedicated to the co-emperors Marcus Aurelius and his son-in-law Lucius Verus. This inscription was reused as the inner facing of one of the large hot water pools, laid-out in the late fourth century A.D. The central room of the complex was originally a representative space dedicated to the imperial family, known as Kaisersaal (K on the aerial view) or Marmorsaal. It measures 25 meters by 20 meters. After its completion, the complex underwent several repairs or larger transformations until its partial destruction by the mid-seventh century earthquake.

The Kaisersaal belonged to the original building face, which had at least four entrances. One of them led to caldarium 1 (C 1 on the map), a hot water room stretching along most of the northern indentation of the west facade, with a tepidarium (tepid water room: T on the aerial view) or apodyterium (undressing room) located to its north. To its east, there was the baths' central structure. Whereas the rooms on the western side of the baths were used continuously throughout late antiquity, the central section underwent significant changes between the late fourth and early fifth centuries A.D. These transformations may have been due mainly to the introduction of Christianity and the abolishment of the imperial cult, which made the role of the Kaisersaal superfluous. This vast central space was therefore transformed into a second hot water room or caldarium 2 (C 2), provided with larger pools (P) at either end and a smaller tub (t on the aerial picture) in the northeast and southwest corners. This may have allowed for simultaneous bathing of both sexes. A second vast space (34 by 12 meters) located to the north of the Kaisersaal, of which the original function has not yet been established, was transformed into a second frigidarium or cold water room (F 2). Its western end was arranged as an apse framed by two smaller curved recesses in the long walls. Henceforth a large pool neatly followed the contours of the walls and occupied the eastern third of the space; the central third had three rectangular recesses, one of which in each wall contained a doorway to rooms on either side of it. The other recesses were filled with marble bath tubs (t). The western third of the vast space was subdivided into two halves: an undressing room in the northwest corner surrounded by low benches and built in the early fifth century, (A), and a heated corridor (HC) linking frigidarium 2 to caldarium 2 in the southwest corner. During the late fourth and early fifth centuries the floor of frigidarium 2 and parts of its walls were covered with geometrical, vegetal and perhaps even figural patterns made of more than 20 different exotic colored stones. A series of service rooms occupied the corner formed by the eastern end of caldarium 2 (C 2) and of frigidarium 2 (F 2). Adjoining this was a cross-shaped room of 1150 square meters (max. L.: 53, 60 meters; max. W.: 16, 75 meters), apparently completely covered with black and white mosaics. Its central section, framed by four massive ashlar piers of more than 8 meters high and originally covered with marble veneer, measured 12.8 by 12.1 meters. We assume that it contained another cold water pool.

Of the four arms of this cross-shaped suite of rooms, only one small (west) and one long one (north) have been excavated (both F 1): the former contains against its east wall a curved water basin with sitting benches, fed by a cascade falling behind a statue of Apollo. The two long arms measure respectively 20.40 by 16.80 meters (northern arm) and 20.70 meters by 16.70 meters (southern arm), each flanked by two rows of three massive piers each. Only the northern arm was excavated in 2005 (F 1). We believe that this frigidarium 1 (F 1) still reflects the original layout and function of the bath complex. Its massive piers were then covered with marble wall veneer, probably heavily damaged during the A.D. 500 earthquake. After the quake, the veneer and the hooks holding it in place were taken down. Other hooks, which could not be removed, were simply hammered down. After that, a simple white plaster covered the walls and piers. Although they were covered by nearly six meters of mortar, brick, concrete parts and ashlars, the exceptionally well-preserved mosaics in respectively frigidarium 2 and 1 suggest that during the final catastrophe the roof covering both spaces did not fall down at once. The spots where the floor covering was completely smashed or was missing clearly received the full impact of large and compact chunks of ceiling material: mortar, vertically placed brick and a thick layer of concrete. The well-preserved floor segments, however, were covered by a ten centimeter thick mortar layer, a layer of loose and chaotically arranged brick above it, concrete and finally by voussoirs of huge ashlars. All of this seems to indicate that during the mid seventh-century earthquake some parts of the concrete vaults covering frigidarium 1 and 2 as well as caldarium 2 smashed into their floors, destroying the floor coverings. These chunks are still preserved in their original layers. As almost nothing survived of the hypocaust floors in caldarium 1 and the tepidarium/apodyterium to the north of it, one must assume that their vaults had almost immediately collapsed and went right through the heated floor system and its supporting brick piers. Yet large sections of the concrete vault above both frigidaria remained in place. After the first impact they lost only their plaster coverage. Then, over a certain period, first brick from both brick layers and then the concrete that they had covered, came loose and fell upon a thick layer of plaster and brick protecting the floor. The last part of this roof system to come down were the more than 12 meter wide arches made of ashlars, once spanning the four huge piers in the central part of frigidarium 1 that were found on top of the destruction layer.

This evidence that the destruction actually did take place over a larger time span was also shown by the discovery in 2004, in the middle of the thick layer of rubble several meters above the original floor of frigidarium 2, of the remains of 24 quarters of sheep and possibly goats slaughtered for food. They would have fed a group of 50 to 100 people. In 2005, in the plaster layer of frigidarium 1, about 0.20 to 0.30 meters above floor level, several owl pellets (containing the undigested remains of small birds and rodents) were retrieved. This showed that at least one pair of owls had lived inside the still partially covered ruins of frigidarium 2. A coin minted by Constantius II in A.D. 742 discovered on one of the floors provided a terminus post quem for the earthquake. Calibrated radiocarbon dating of both bone collections resulted in a fairly accurate date for the creation of the collections. The contents of the two owl pellets gave a radiocarbon date between A.D. 530 and A.D. 645. Combined with the coin of A.D. 642. this suggests a date for the devastating earthquake of in the 640s, most likely between A.D. 642 and 645. The radiocarbon date of the food refuse found in frigidarium 2 gave a probable period of between A.D. 605 and 670. This suggests that within a generation of the animals being eaten, the lower half of frigidarium 2 had already been filled with falling debris.

In order to clear out the debris filling the rooms to the south of the old Kaisersaal and to the west of frigidarium 1's southern arm with a crane, the southern limit of the bath complex had to be excavated first. These rooms consist of slightly larger space of about 18 by 18 meters in the west (Roman Bath 1) and a somewhat smaller space of about 15 by 15 meters (Roman Bath 2) adjoining it to the east as far as the southern arm of cross-shaped frigidarium 2. The excavation of Roman Bath 1 sector is being supervised by Frank Carpentier of K. U. Leuven and Hasan Uzunoglu of Mimar Sinan University; the Roman Bath 2 section by Tijl Vereenooghe and Annelies Coenen, both of K. U. Leuven, and by Rukiye Sen of the University of Erzurum. During the first days of the project, they struggled with the newly developed recording system (see recording 10-20 July, 2006) based on loci instead of strata (layers).

At the same time, the surface of the excavation sector had to be levelled in order to create an access road for a mobile crane. The Roman Bath 1 team focused on the open space between the two arches in the south wall of the Kaisersaal (K or C 2). Thus far no remains of a bath tub (t) were discovered here. However, behind the ashlar piers the ends of two opposed brick walls were discovered, suggesting the presence of a passage towards the Kaisersaal. The inner face of the eastern brick wall was covered with charred or smoothed mortar, suggesting the heated wall segments in the room excavated by the Roman Bath 1 team. Most of the finds consisted of architectural fragments such marble veneer, and a few vessel parts.

Meanwhile, the Roman Bath 2 team started excavating a mere 20 meters to the east of the Roman Bath 1 team. This area lies immediately south of frigidarium 1. (F 1), which was partially unearthed last year (see Roman Baths: August 14-18, 2005). We started excavating the first layers in the sectors adjacent to the southern wall of this frigidarium. The presence of large ashlars, however, prevented us from continuing the work in these sectors. We decided to continue the excavations further to the south until these large limestone blocks are removed. Here, the uppermost stones of the baths' southern exterior wall are still visible at the surface. We hope to learn more about this wall in the next few weeks.

As is usual with Roman baths, our finds were not spectacular. They consisted mainly of marble crustae, including wall veneer, ceramics, and metal clamps. More remarkable was the discovery of part of a column in verde antico, a serpentinite breccia, popular as a decorative facing stone. This stone, which is only rarely found at Sagalassos, was quarried near Larissa in Thessaly, Greece.