Umm
el-Jimal
THE UMM EL-JIMAL PROJECT, 1981-1992Annual of the Department of Antiquities of Jordan. 37. Amman. 1993. Pg. 433-460.Bert de Vries
TABLE OF CONTENTS
Bottom of Page
|
||||||||||||||||||||||
 |
Map of Umm el-Jimal showing town and village (Drawing by B. de Vries and Randa Sayegh). |
This was the fourth season of the Umm el-Jimal Project. In the previous seasons we accomplished the following: the mapping of the town (1972-3); preliminary soundings (1974); the excavation of the Barracks (1977), the Praetorium (1977), the town wall and Southeast Gate (1977); House XVIII (1977); excavation of the so-called Nabatæan Temple (1977); partial excavation of House XIV (1977).
1. Excavation.
Teams dug forty soundings in the various structures listed below. These soundings were positioned with two objectives in mind: 1) studying various types of structures and installations (i.e. houses, churches, water channels, dumps, town gates and walls), and 2) dating buildings in order to sort out the various periods of inhabitation. Because extensive Byzantine and Umayyad remains were found in 1977, our particular goal in the 1981 season was the discovery of the Early and Late Roman cities.
Excavation included further work on the Praetorium, the East Gate, the so-called "Nabatæan Temple" and House XIV; new work on the Northeast Church, the Numerianos Church, the newly identified Roman Castellum and various water channels, the Via Nova at Qasr el-Ba'ij.
2. Surveys.
Ethnographic. Throughout the season, supervisor Robin Brown (very ably assisted by representative Hafzi Haddad) conducted taped interviews with Sheikh Khail es-Serour. From this work a very interesting oral history of Umm el-Jimal in the twentieth century emerges.
Epigraphic. Throughout the season Vincent Clark and Paul McDermott followed up on reports of new inscriptional materials, and several Nabatæan and numerous Greek funerary inscriptions were recovered.
Perimeter. The work of Vincent Clark and Scott Rolston in a 5 to 15 km sweep around the town resulted in the recording of the following phenomena:
In addition to the main Roman road passing west of Umm el-Jimal, there is a Roman road leading from the North East Gate in the direction of Sabha and Umm el-Quttein. A mile stone, possibly erected to commemorate a rebuilding in the reign of Diocletian (Parker 1986:256-8), located one mile east of the town was removed and stored in the excavation tool room at Umm el-Jimal to save it from ongoing agricultural expansion.
Both east and west of the town there are numerous garden plots, dams and reservoirs indicative of extensive Roman agricultural activity. The entire perimeter region around the town is dotted with dozens of tombs and mausolea. While some have been disturbed, others may still be intact.
3. Consolidation
 |
Gate of House XVIII; sketched from the courtyard by Randa A. Sayegh. |
Clearing and cement force-filling of the walls of House XVIII was begun. This work was continued for three weeks in January, 1983, with funding from the Department of Antiquities and under the foremanship of Mifleh al-Gharaibeh.
Back to Top
B. The 1984 Field Season: Goals and Procedures
The work of previous seasons gave a fairly good cross-section of the extant buildings and a much better understanding of the history of the town. Hence the goals of the 1984 season were the completion of excavation already begun inside the town and investigation of the surrounding area with excavation as well as further survey. The work has to include: completion of the Numerianos Church and long hoped for excavation of the Julianos Church; extensive excavation of the Roman Fort (castellum) with one supervisor working in the Principia area and another excavating a barracks block; excavation of various tombs and cemeteries around the town; study and excavation of the agricultural system; continuation of the regional survey. The project was therefore split, with half the work inside the town and half in the region around it.
1. THE FORT (Castellum)
 |
Plan of the castellum (Drawing by B. de Vries). |
In 1981 the chronology of the 100 x 100 m fort located northeast of the main reservoir was already clear from excavation of the barracks block excavated in the NW quadrant. The fort was founded in the late Roman period (ca. AD 300), used for military purposes in the fourth century, then converted to domestic use in the Early Byzantine period when the Barracks were built to accommodate the much smaller military presence. In the Late Byzantine period (sixth century) a church and extensive domestic complex was built in the SE quadrant in which much of the fort masonry was reused. The robbed out northern half of the fort lay in ruins during this period. In 1984 the purpose of excavation was further refinement of this story and better delineation of the architecture. To achieve this end Jennifer Groot continued work on the West Gate and excavated a set of barracks rooms immediately adjacent to the Late Byzantine complex in the SW quadrant (Area J). Anne Haeckl continued work on the NE corner tower and opened up soundings in rooms thought to be part of the Principia complex; later she opened a sounding on the east wall to investigate masonry, discovered by the architect (Area F). Meanwhile architect Bill De Kraker spent much of the season refining the plan of the Fort.
2. THE CHURCHES
The 1981 excavation had concentrated on the narthex, nave and side aisle; a probe in the apse had not been completed. This season excavation of Area E was continued in the apse and in a new probe in the chancel area.
The North half of the Double Church
Because of unavailability of the crane for the Julianos Church, Robert Schick's team opened up probes in this church, famous for its pre-Islamic Arabic inscription. This excavation was Area X. The probes consisted of an L-shaped section across the apse, chancel area, nave and side aisle.
The Julianos Church
After a department crane became available, removal of collapse debris from the apse and nave of this church took
 |
Thumiasterion found in Julianos Church (Drawing by B. de Vries). |
place during the last eight days of the season. This project yielded several inscriptions, including some Greek tomb stones (one with the interesting name ABD al-GO ....) and a Nabatæan one badly interrupted by the incising of an altar screen base across the middle of the stone. A very interesting piece is a small square pillar-altar that Michel Piccirillo thinks is a thumiasterion (an incense burner) which found its way into the church after first having seen use in a pagan temple. There was no time to begin the actual excavation of the church.
3. THE TOMBS
Investigation of the cemeteries of Umm el-Jimal was a major effort of the season. The existence of monumental tombs below ground of superb basalt masonry had long been known. These tombs, with doors opening onto central chambers from which multi-tiered burial vaults radiate on three sides and with various ceilings (corbels, arches, vaults), are equivalent in quality to those at Umm Qeis. In addition, numerous tomb stones simply inscribed with the name of the deceased person and her/his age (written in Greek but Arabic in culture) have been known from their reuse as corbels and stairway treads in the Byzantine town. Hence, supervisor Bruce Dahlberg and osteologist Stephanie Damadio were assigned a team to find an undisturbed tomb in an obvious cemetery east of the town (Area O) while Laurette De Veaux probed possible tomb locations around the West Church. As it turned out, tombs were discovered elsewhere by accident rather than by design.
4. THE AREA T WADI SYSTEM
Under direction of Ghazi Bisheh the wadi system east of Umm el-Jimal was extensively surveyed and its two main reservoirs were excavated. This wadi is at the center of a gently sloping valley in which numerous terrace walls, dams and channels indicate intense agricultural activity. The ground survey was supplemented by the use of aerial photographs to achieve the detailed mapping of the area. Fortunately, this entire system, although reused for some run off farming, was not yet disturbed by the extensive irrigation farming based on the drilled deep wells around the north side of the town. Because of the rapid development of the area, including extensive leveling by bulldozers, such a study has become very difficult since 1990.
5. THE AREA R PROBES
During the previous seasons of excavation a great puzzle emerged. Umm el-Jimal had a reputation as a Nabatæan - Early Roman town. However, while major structures existed in the Late Roman to Umayyad periods, not one Early Roman wall was found. One began to wonder if the early town was merely a myth of modern fabrication.
In order to go from the town to the wadi system, excavated by Ghazi Bisheh, one had to cross two to three hundred meters of moonscape: rubble of undressed, lichen covered, basalt that made walking next to impossible. Scattered throughout this three hundred by three hundred meter wilderness were a half dozen small black mounds that were extremely rich in surface pottery. We assigned Laurette De Veaux and her team to do a systematic surface sherding of these hills.
With only three weeks of the season left it was not possible to do detailed excavation there. The strategy adopted was to open four or five small probes, Area R, distributed throughout the "black hills" to test the distribution of surface sherds and enable planning for a major effort in the coming season.
Back to Top
C. The 1992 Field Season: Goals and Procedures
The season's goals included precise mapping, architectural drawing of selected buildings, extensive photography on the ground, and low level aerial photography with cameras suspended from a balloon. The purposes of these recording activities were to provide illustrations and photos for excavation publication currently under preparation and to provide detailed plans and maps for a major excavation season to take place in 1993.
The surveyors made precise control measurements on the Late Roman fort using an electronic distance meter. They also laid out a triangulation grid for mapping controls in the Early Roman-Late Roman village discovered.
The team of architects did detailed study of four building complexes. These included House 119, a simple domestic structure with two well preserved stables with interesting rows of mangers. This house is located at the entry to Umm el-Jimal, and is an ideal candidate for conversion into a museum/rest house.
 |
[above] Plan of House 119 (Drawing by Edwin Orogo). |
[below] Plan of House 49 (Drawing by Jason Yeats). |
 |
House 49, which includes the so-called Nabatæan Temple, the object of clearing by the Department of Antiquities, was also carefully measured and drawn. It is important not only because it incorporated an earlier public building into this domestic environment, but also because it contains an L-shaped stable arrangement with a plan that provided the standard structural core for several other Byzantine houses on the site. The larger room was used to house untethered animals; it has a cubicle-like bathroom stall built into its northeast corner. The other room is long and narrow with a row of mangers running down the side connecting it to the first room. Here, animals could be tethered one per stall. The doors of the rooms opened onto a central courtyard, where exterior stairs give access to upper rooms where animal fodder was stored and the human residents slept.
House 35, located in the same quarter, was studied to compare the building methods of the fifth/sixth century Byzantine inhabitants to those of their Druze successors ca. 1925. The great room of this house is preserved to ceiling level. Its corbeled roof was supported on a wonderfully finished central arch that allowed the ceiling to soar over five meters above the floor. A second room had an intact ceiling supported on the walls and an arch built by the Druze. This arch was relatively low slung, and constructed of variously sized reused building blocks, aligned on the entry-side of the room. Although the makeup of this arch and the associated corbel-beam roofing system was much more irregular than its Byzantine predecessor, it is clear that the Druze of the twentieth century had adopted the Roman-Byzantine building system with great expertise. Thanks to them, many houses at Umm el-Jimal have been preserved remarkably in their ancient style.
The highest wall of the Praetorium with the bit of remaining gable roof at the southwest corner of the atrium has long been in extremely precarious condition. While the Department of Antiquities cleared previously excavated rooms, the project team systematically numbered all the stones in the walls that are threatened by collapse. All threatened walls were then drawn and photographed. As soon as a crane and other equipment become available, the affected masonry can be dismantled and reconstructed under the supervision of a restoration architect by master stone masons.
A highlight of the season was the aerial photography done by balloon. Two specialists, Dr. Wilson and Mrs. Ellie Myers were brought with their equipment from Greece in a joint effort in which three other projects participated. The Myers do low altitude aerial photography by suspending cameras from a balloon, with the shutter trigger and film advance operated by remote radio control.
 |
Tombstone from House 97. "Masikh, also called Suma, [son of] Rawah, all [his] years [were] 40." (Drawing by Randa Sayegh). |
In the site of the earlier village photographs were taken from heights of 100 and 225 meters. From these heights the photographs show every stone of this badly destroyed area. Because the surveyor's triangulation grid can be located on the photographs, it will be possible to produce scale maps of distinguishable features, such as exposed wall lines, in preparation for the coming excavation. The Late Roman Castellum was photographed at 300 m. in order to have the entire structure appear in a single photo from which a scale plan may be produced using the ground control measurements.
The tall buildings of Byzantine Umm el-Jimal were photographed in five sections from a height of three hundred meters. Then a Hasselblad camera with a wide angle lens was raised to 600 and 700 m. to take the whole site in a single photograph.
This balloon photography is a first for Jordan. It is expected that this method of remote sensing will add significant data to both that gathered on the ground and that available from overflights by aeroplanes and satellites.
Both the clearing by the Department of Antiquities and the close examination of specific buildings resulted in the discovery of a number of inscriptions and decorative stones. Recording these was a major component of the season. Over fifty inscriptions and five decorative fragments were drawn and photographed. Several inscriptions already published by E. Littmann, including the important historical memorial of a soldier in the Third Cyrenaica (fig. 12; Littmann 1913a: 177, inscr. 349) were redrawn. Most of these inscriptions were tombstones in Greek using the formula: X (son or daughter) of Y of the age Z. Two were Nabatæan.
All the buildings studied were photographed in detail, and illustrative sketches were made to supplement precise plans and elevations.
Back to Top
II. STRATIGRAPHIC RESULTS
A. Stratum VII, Early Roman (63 BC - AD 135)
1. Inside the Town
An extensive ash deposit was located under the foundations (LR) of the East Gate. This ash contained much Early Roman Pottery, unmixed with earlier or later pottery. In the north, dumping of possible kiln cleanout was begun at area K. The extensive ash deposits do nevertheless indicated some major activity.
2. Via Nova Traiana
In connection with the regional survey Vincent Clark excavated a section across the Via Nova just east of where it crosses the road northwest of Qasr al-Ba'ij. No artifacts were found in the soil around or under the surviving cobble underlayment.
3. Area T Tomb
While investigating the agricultural installations alongside the wadi east of the town, Ghazi Bisheh probed one of a number of stone circles laid on apparently fertile ground. Instead of a tree plantation he found a deeply dug, very well constructed, simple cist tomb, with one of its capstones erect. Puzzle of puzzles: the tomb was completely empty except for the clean silt that had filled it through the open cap stone. Construction is dated to the Early Roman period from potsherds in the backfill. Whether it was ever used, or why it was robbed so cleanly cannot be known.
4. Area T Reservoirs
The excavation focused on two huge reservoirs (each ca. 40 x 20 m) several hundred meters apart along the wadi. The one farthest west was constructed by laying a massive clay oval dike right on the bed of the wadi. The interior was then lined with basalt masonry which leans against the inside of the clay bank. The masonry provided the strength while the clay provided the waterproofing and the bullwork against the wadi's stream which was diverted around the north of the reservoir. By building right in the wadi bed it was possible to create a very deep pool: the height of the still intact retainer wall at the low end is an astounding five meters! The reservoir was fed from a take off channel that parallels the wadi for some distance and spilled its water over top of the reservoir wall. Water was drawn by hand via sloping ramps on the north wall. Today the wadi had punched through the reservoir walls at both ends to resume its pre-Roman path. This structure is indicative of the simple but extensive efforts to develop the agricultural water supply, and the drinking needs of the small village that was discovered alongside this wadi.
The eastern reservoir is a more traditional rectangular structure built on the northern slope of the wadi. Its inlet channel had a concrete-hard, thick sediment floor produced by centuries of use. Both reservoirs were built in the Early Roman Period and continued in use for centuries to the end of the Late Byzantine Period.
5. The ER/LR Village
The result of the Area R pottery survey was as follows: The sherds were from 80 to 99% Roman, with Early Byzantine sherds typically less than 5% and Late Byzantine sherds from 0 to 1%. What's more, well over half of the Roman sherds were Early Roman rather than Late Roman! Meanwhile, it became apparent that the moonscape rubble had walls crisscrossing through it.
The five probes done during the last two weeks of the season exposed three domestic structures and a roadway or street surfaces. The dates of structures and occupation layers indicate construction and use during the Early and Late Roman periods, with destruction in the late third century. The relatively simple quality of the architecture and the small overall size of the settlement is indicative of a rural village. This village is to be associated with the Area T reservoirs, agricultural terraces and cemeteries in the immediate vicinity.
Meanwhile, geologist Frank Koucky was tracing roads in the region of Umm el-Jimal. Every time he came close to the Byzantine town he would lose the traces of roads that could be followed clearly farther out. However, once the notion of the second, older settlement sunk in, the road patterns began to make sense. They radiated out from the earlier, not the later site! In addition Frank found the traces of a major structure just north of the older settlement, perhaps a fort or possibly a caravanserai.
These results were most gratifying. It appears clear that the Early Roman site of Umm el-Jimal has been found and it is next to, rather than under, the Byzantine town!
Back to Top
B. Stratum VI, Late Roman (AD 135 - 324).
While in previous seasons only the town wall probes showed LR foundations, much new evidence came to light, to the point that major features of the Late Roman settlements can now be described.
1. The Praetorium.
Dating of foundational pottery points to a construction of the building in this period. However, the early use of the structure has been obscured by major rebuilding of floors in the Byzantine and Umayyad periods. In addition to the stratigraphic work, Jennifer Groot did a detailed survey of the plaster fragments remaining on the walls of the building.
2. The ER/LR Village.
While the village appears to have been founded late in the Early Roman period, it enjoyed buildup in the second century AD, and continued without interruption in occupation until late in the third century.
3. Cist Burials.
One morning at 4:30 am on our arrival from Mafraq we discovered
a shaft tomb which had been dug overnight in a
 |
 |
Limestone altar screen from Northeast Church (Drawing by Randa Sayegh) |
|
4. Area O.
Tomb 0.6 was an already open cist tomb built into the bedrock of the area first investigated by Bruce Dahlberg. Its deep shaft below the capstones contained much soil in which the intermixed remains of numerous skeletons were found. Based on the artifacts in the soil contents the construction and first use is late Roman; the robbery and reuse is twentieth century.
5. Area V Mausoleum.
In a field northeast of the town bulldozing to level the soil for orchard use had knocked the roof off a structure and filled the underground cavity thus exposed with masonry and surface soil. Its excavation by Bruce Dahlberg's team was a season-long task that revealed a monumental tomb with intact stone door and two tiers of four burial vaults (eight total) across a small chamber. The vaults were filled with soil and a few bones and artifacts. The presence of glass and pottery inside the chamber and in the passage outside the door indicates an extensive robbing, dated to some time in the Ottoman period, based on pottery in the soil fill. An interesting feature of the tomb is its ossuary, constructed on the side of the tomb and connected to it by a small opening in the east side wall of the chamber. The ossuary was a small (1.5 m) oval chamber filled from top to bottom with the bones of numerous disarticulated skeletons. This ossuary represents the practice of discarding the skeletal remains when the tomb was cleaned out for reuse in antiquity.
 |
Plan of Area V Mausoleum (Drawing by B. de Vries). |
The artifacts in the disturbed debris indicate that the tomb was constructed in the Late Roman period and used from three to four centuries in the EB and LB periods. Although it was well constructed, the size and quality was inferior to the numerous robbed out tombs already seen by H.C. Butler in 1905. Its excavation, however, serves as an ideal occasion for an architectural study of this type of tomb.
6. Empty Mausoleums.
Many of the known mausoleums are currently being robbed of their very fine masonry. The five famous Nabatæan inscriptions published by H.C. Butler with the so-called Nabatæan Tomb were rescued by our crane from a local landowner's courtyard. However, many truckloads of monumental masonry have been spoiled.
7. The Roman Fort (castellum).
Constructed on the east side of the town, this structure, with 1.90 m walls from 95 to 112 m in length, had east, south and west gates with offset towers as well as four offset corner towers. It is located in the very badly ruined area between the large reservoir and the east side of the town. Because of its badly ruined condition it had not been identified in the course of both Butler's 1905 and de Vries' 1972 surveys. While in 1981 excavation was begun in the barracks rooms, the east and west gates and the north east corner tower, major stratigraphic conclusions followed the 1984 season.
The West Gate. The West Gate and its south flanking tower were clearly built in the Late Roman Period. However, most of the masonry, including the massive doorstep is the product of a Late Byzantine remodeling.
The Barracks. The rooms investigated were a suite on the West side of a spine wall and opening up onto a secondary north-south street. Although both the street and the walls were built in the Late Roman period, Byzantine reuse was so intense that these founding levels were difficult to reach.
The Principia. The open flat area with adjoining rooms in the north center of the fort was thought to be the Principia. A series of probes were opened up from this area north across rooms on both sides of an east-west street and up against the north wall of the fort. It was found that the room most likely to be the aedes opened onto the east-west street rather than on to the Principia. It, along with the room built against the north wall of the fort was interpreted to be a store room. Then, a probe laid out on the apparently vacant south side of the large Principia wall met with success: Under a Late Byzantine robbing platform was a finely finished room complete with excellent plaster wall facing and an excellent concrete floor into which was built a base for the military standards of the fort -- right on the line of the north-south axis! The aedes was found, but only in the last days of the excavation.
The Northeast Corner Tower. The work begun in 1981 was completed in 1984. The tower was found to be 4 x 6 m, with a doorway from the fort at ground level on the six meter side. There is no evidence of a staircase to provide access to the ramparts either inside (as at el-Lejjun) or outside (as at Khirbet es-Samra) the tower.
The Staircase-platform. Roughly halfway between the Northeast Corner tower and the East Gate a massive platform was found with a monumental staircase leading up to it. The architects have discovered similar installations on the north and west walls of the Fort. Perhaps these staircase-platforms provided instant access to the ramparts for soldiers living in barrack blocks close to them; or perhaps these platforms served as bases for defensive ballistas. A rather elaborate Late Byzantine domestic structure that took advantage of the staircase (as a piece of house furniture) delayed complete excavation of the staircase-platform. These staircase-platforms are unusual in Castellum design.
8. The East Town Gate.
Located in the southern part of the east town wall, this gate was founded on the ER ash layers.
Various installations indicate a support system for a major population buildup during this period. The way in which builders contorted the Fort into the available space may indicate that the large open reservoir had already been built when the Fort was laid out. Both major aqueducts, running for many kilometers from the northeast toward the town, were built to substantiate this buildup. The Area K pottery kiln dump was added to extensively, indicative of industrial and commercial activity.
Back to Top
C. Stratum V, Early Byzantine (AD 324 - 491).
Work in the 1977 season determined the building date of the Barracks as AD 412/3. To that, the following information can now be added.
 |
South arch in Druze reconstruction of the so-called "Nabataean Temple" (Drawing by Edwin Orogo and Mohammad Ballout). |
The building of the Barracks indicates a military restructuring that involves the ruin of the Fort and its remodeling for civilian use, perhaps as a market. The "Nabatæan Temple" is actually a building of unknown use constructed in the fourth century. House number 92 and its adjacent cistern may also have been built in this period. More certain dating calls for further excavation of this domestic complex (farm house). Earlier structures like the Praetorium continued to be used throughout this period.
Occupation debris in the Area J.2 street between the barracks west of the principia in the Roman castellum indicate that the fort was being used for domestic and commercial rather than military purposes.
Tombs, like the Area V Mausoleum, originally associated with the now ruined village, continued to be used for burials.
The Early Byzantine period witnesses a shift from the Late Roman military and possibly commercial function of the site, to a greater emphasis on local domestic activities. This shift involves a reduction of military presence and adaptation of less defensible structures.
Back to Top
D. Stratum IV, Late Byzantine (AD 491 - 636).
The trend toward domestic architecture geared for an agricultural economy continued. The 150 private housing complexes now still standing were either built or extensively remodeled in this period. In addition, the 15 churches of the town were constructed at various times, with the Barracks chapel accompanying the LB construction of the two towers within the Barracks. Area T reservoirs and tombs like the Area V Mausoleum continued in use during the Late Byzantine period.
The EB "Nabatæan Temple" was incorporated into a typical domestic complex and used as one of its rooms. The North East Church, built in the late fifth century, suffered two earthquake collapses in the sixth. After the second midcentury reconstruction the apse was abandoned, while the nave continued in use.
1. The Area of the Fort.
The area of the Roman castellum was extensively reused. A Church and adjacent domestic complex were constructed along the east wall south of the Northeast Gate, and the barracks between those buildings and the great reservoir were intensively used for domestic purposes. A street led from the rest of the town through the west gate to the church and to the open country through the Northeast Gate. The northern half of the fort, especially the principia area, was extensively quarried for these construction purposes.
West Gate. Street surfaces were frequently reapplied through the gate. All this and a number of street surfaces in and outside the gate represent the use of the east-west street of the Fort to provide access to the East Church and the rebuilt Northeast Gate adjoining that church.
The barracks in the south half of the fort were extensively reused in the Late Byzantine period. It took the entire season to excavate through the heavy collapse and the Late Byzantine occupation below it. The presence of a number of contemporary (LB) taboons in the outer room indicates that it was used as a commercial cooking installation, perhaps a bakery for the large domestic complex adjacent to the room. The inner barracks room was used as an ash depository for the taboon cleanout.
Pottery. The complex stratigraphic deposits from the Late Byzantine period in both the Area J barracks and the Area F staircase-platform have yielded a much better corpus of Late Byzantine pottery than the excavations of LB structures (like House XVIII) in previous seasons.
2. The Numerianos Church
This ecclesiastical complex was built up in three successive stages, all in the Late Byzantine Period. First came the church itself, then the enjoining cloister, and finally a flagstone court to the west of both structures. The church itself was remodeled in the successive construction phases.
The synthronon was founded directly on a fine mosaic floor. This floor was replaced everywhere else by a well constructed plaster floor, perhaps because the mosaics had deteriorated too greatly to be repaired (or perhaps to suit a more austere and simple decorative taste). The section through the series of underlayments of the mosaic provided an unusually fine study of mosaic floor construction methods.
In addition to very fine stratified flooring, a probe in the southeast corner room (a presbuterion or marterion) produced an unusual architectural fragment, a substantial stone with two circular cups (ca. 15 cm diameter x 10 cm deep), that may have something to do with the storing of relics.
Back to Top
E. Stratum III, Umayyad (AD 636 - 750).
The transition from the Late Byzantine period showed some collapse, but mostly continued use of the buildings of the town. The later Umayyad decades involved extensive remodeling and continued use of many structures. Umm el-Jimal has to be pictured as a thriving community at the time that the desert castles were being built.
For instance, the abandoned "Nabatæan Temple" and the partially collapsed Barracks enjoyed squatter use. The Praetorium was extensively refloored and replastered and may have been as lavish in decor as the desert castles. (The same reflooring took place in House XVIII excavated in 1977.) The East Gate was rebuilt with a much flimsier door construction, in keeping with the strictly localized security needs of the period.
1. The North East Church
The North East Church nave was refloored and an altar screen was installed, evidence of continued Christian liturgical use. The continued abandonment of the apse may be the result of a squaring off of the sanctuary in parallel to the three churches excavated at Khirbet es-Samra.
2. The Numerianos Church.
The Numerianos Church was also still in use. Whether the blocking of the apse and southwest door means a conversion into a mosque is uncertain. There exist neither artifactual nor architectural evidence of Islamic religious use.
Unlike in the nave area, the collapse debris in the apse contained Umayyad pottery. Under that it was found that the synthronon (deacon's bench) was added in a possibly Umayyad remodeling of the church. If this is true then the church was used as a church in the first part of the Umayyad period. However, its function as church discontinued later in the Umayyad period when the blocking wall separated the apse from the nave completely. In the chancel area these Umayyad phases may be represented by a repair of the plaster floor that was necessitated by the removal of the chancel screen foundation stones.
3. The North Half of the Double Church.
The church proved to be constructed, used and ruined entirely in the Late Byzantine Period, with no evidence of Umayyad use and reuse as in the Northeast and Numerianos Churches. The church had a series of three well constructed plaster floors, the last of which was well preserved and painted red. The chancel screen wall stretched straight across nave and side-aisles and contained a reused possibly Safaitic (poorly legible) inscription. The famous Arabic inscription (Littmann 1913b: 1-3, inscr. 1) was rediscovered still in situ under some collapse debris in the northwest corner of the church.
Area T reservoirs and tombs like the Area V Mausoleum continued in use during the Late Byzantine period.
Back to Top
F. Post Stratum III Gap (AD 750 - 1900)
Umm el-Jimal was nearly totally abandoned during this long period. Perhaps the earthquake of AD 747 destroyed the town. Certainly the shift of political center from Syria to Iraq brought on by the Abbasid revolution caused the Hauran area to lose its strategic significance. The survivors moved to a less strenuous environment in Syria and Palestine, or reverted to a primarily nomadic mode of living. Even in the Ayyubid-Mamluk period, when so much of Jordan was densely populated, there was only evidence of sporadic occupation (perhaps pilgrim stopovers) in the "Praetorium" and in the so-called "Nabatæan Temple."
Back to Top
G. Stratum II, Late Ottoman/Mandate (AD 1900 - 1946)
Umm el-Jimal was reoccupied in two ways during this period: 1) The Druze who moved into Jordan did extensive remodeling of still standing buildings before moving on to Azraq and Amman in the late 1930s; and 2) various troop movements took place during the post World War I settlement, including encampment by Turkish, British and French army units.
The entire superstructure of the "Nabatæan Temple" including the two arches (except the facade around the triple doorway) is Druze construction. Numerous domestic complexes like House 35 were extensively remodeled and reroofed. Rebuilt in this period were the southwest corner of the North East Church, including the room with the still intact roof, and the aqueduct in the southeast area, just outside the town wall. This aqueduct construction possibly coincided with the re-plastering of the cisterns of Houses XX and 132.
Back to Top
H. Stratum I, Modern (AD 1946 - present)
The tribe of the current village of Umm el-Jimal made extensive use of the site before the erection of the fence by the Department of Antiquities. The court area east of the "Praetorium" was occupied by a family which apparently repaired the water channel into the Praetorium reservoir and built the concrete troughs on its edge. Also, the platform area east of the town wall and north of the East Gate was used as an orchard a generation ago, according to local tradition. The ethnographic survey and on-going contact and working relations with the living inhabitants make this phase of the history of the site the most clearly documented (though not necessarily most clearly understood).
Back to Top
III. Towards a history of Umm el-Jimal.
Introduction: The meaning of the discovery of the ER/LR Village.
Some time in the Early Roman period the first settlement was constructed, perhaps by Nabatæans but more likely by local Arab settlers. Whether to call it town or village is still a problem: the architecture is less elaborate than in the Byzantine town, but the place is extremely rich in high quality pottery, including much imported wares such as terra siggilata. The new balloon photographs reveal the intriguing possibility that the settlement may have been planned with a rectangular layout. Nevertheless the term "village" has been adopted to distinguish the settlement from the later town.
The settlement continued into the Late Roman Period. At this point, perhaps in the third century, a major disturbance resulted in the destruction of the village. (What it was we do not know. The fact that the tombs and tomb stones were being robbed may indicate the arrival of a new people.)
Meanwhile the growing presence of the Roman government had focussed on what was to become the new town 300 m west. The Commodus Gate was built in the late second century. The Praetorium and Castellum followed in the wake of the military reorganization of the late third century. The new civilian arrivals found it more attractive to settle there. What to use for building blocks? The village was a convenient source. The systematic quarrying and stone chipping throughout the village over a long period left the pottery strewn "Black Hills," mixtures of basalt stone chips and cooking fire ashes. By the end of the Late Roman period all that was left of the village was a useless rubble moonscape. Very few Byzantine people bothered coming there at all. However, the memory of its people lived on in the inscriptions on the tomb stones, hauled over, uncut, to make corbels and stairway steps in Late Roman and Byzantine houses. Even the Late Roman fort, so badly ruined at the time of the Late Byzantine town, was built of reused masonry borrowed from elsewhere. Where else but from the village?
Back to Top
A Tentative Reconstruction of Events
Umm el-Jimal was occupied for 700 years from the first century AD to the eighth century, and again early in the twentieth century. In its first 700 years Umm el-Jimal had three quite distinct personae. What follows is a revision of two earlier publications (de Vries 1986 and de Vries 1990.)
In the second to third centuries it was a rural village that received its impetus from late Nabatæan sedentarization, but its prosperity from the incorporation of the region into the Roman Empire after the peaceful transformation of Nabatæan heartland in Jordan into the provincia Arabia (the Roman Province of Arabia). Thus from Trajan (AD 106) to the end of the Severan Dynasty (AD 235) the village appears to have had an undisturbed and relatively prosperous rural life. We glean that from the hundreds of tombstone inscriptions in Greek and Nabatæan, most of which were reused in the later, still standing town. These inscriptions give us the Arabic and Hellenized names of several generations of villagers: "Asad, (son of) 'Akrab, age 30" (in Greek); "Masik, son of Zabud" (in Nabatæan) (Sartre 1985: 194-200). All this came to an end in the middle of the third century, when the village became a victim of the wave of turmoil that ravaged the Roman Mediterranean; a likely hypothesis is that it was ruined during the civil war triggered by Queen Zenobia of Palmyra's rebellion against Rome.
The second Umm el-Jimal was a military station on the limes Arabicus, the fourth to fifth century fortified frontier defensive "system" created and constructed by the emperors Diocletian and Constantine. Already in the second century, the Roman imperial authorities had begun construction of a gate and a wall next to the village, on the site of the later town. This is known from a dedicatory inscription (recorded by H. C. Butler from the ruins of the "Commodus Gate"; Littmann 1913a: 131-2, inscr. 232) commemorating the construction of a defensive wall during the co-regency of Marcus Aurelius and his son Commodus. That was followed by the construction of the great reservoir and the Praetorium. But after the destruction of the village, Diocletian's imperial reorganization caused the construction of a major fortification, the castellum on the east side of what later became the town. Now Umm el-Jimal functioned as a stitch in the blanket of total defensive security in which Diocletian had attempted to swaddle his empire. One has to imagine the fourth through fifth century site with the Commodus Gate, the Praetorium, Reservoir no. 9 and the fort in place, but without the Barracks, the now standing houses and the churches.
This military security may have enabled or forced the civilian resettlement of the site, to set the stage for Umm el-Jimal's third persona, that of a prosperous rural farming and trading town of the fifth to eighth centuries. The transformation from military station to civilian town was gradual, and is typical of the general transformation from imperial to late antique culture that took place in the East Mediterranean in the fourth century. This resulted from the failure of Diocletian's system of massive defenses along the eastern frontier and the reaction to the debilitating economic oppression such a system required. Ironically, as imperial military security weakened and decentralized, the prosperity of frontier towns like Umm el-Jimal increased to reach a peak in the sixth century.
At Umm el-Jimal the Tetrarchic castellum appears to have lost its military function in the early fifth century, and may have been converted into a market place, to judge from the nature of the artifacts found in the streets between its barracks. At the same time the Barracks was constructed as a bivouac for the diminished Early Byzantine garrison inside the now constructed town walls. The last vestige of paganism, a small Roman temple, dated to the earlier Nabatæan era by H. C. Butler, was constructed at this time.
This little temple was soon engulfed in a domestic complex constructed by Christians (House no. 49), one of 128 such complexes that were in use in the sixth century. The conversion to Christianity also brought churches; 15 were constructed in the late fifth and sixth centuries. The Barracks building went through a major remodeling that included the construction of the southeast corner tower, a symbol of the localized defense prevalent during the period of the Ghassanid phylarchy. Thus the town enjoyed its moment of greatest prosperity at the time that centralized imperial controls were at their weakest.
Undoubtedly, the fortunes of Umm el-Jimal were diminishing during the rather catastrophic decades of the late sixth century, when the east suffered the twin ravages of plague and Persian wars. Centralized controls returned in somewhat different form after the Muslim conquest under the Umayyad caliphs ruling from Damascus (AD 661-750). Occupation at the site appears to have been uninterrupted to the end of the Umayyad period. The population continued to occupy at least the central structures (the Praetorium and House XVIII), and at least one church was remodeled in the Umayyad period. The Praetorium was extensively refloored and decorated with frescoed plaster, and appeared to be a prosperous equivalent of contemporary desert castles like Qusair 'Amra. However, Umayyad controls could not stem the tide of depopulation that had begun in the previous century. When the earthquake of AD 747 hit, neither manpower nor economic energy remained for the sort of rebuilding that had been readily tackled in more prosperous times.
For more than 1100 hundred years the town lay virtually abandoned and untouched. Though buildings experienced the continuous slow crumbling that non-maintenance causes, as well as occasional earthquake jolts, the durable basalt masonry and the high quality construction enabled a remarkable state of preservation. Thus when the Druze were expanding out of their mountain perch to the north, a number of them found Umm el-Jimal an attractive place to remodel. Before 1935 a number of buildings had their walls reconstructed, and corbeled roofs newly laid on arches set in the Roman style. When the Druze left to move south to Azraq and Amman, they abandoned a place that appears even more remarkably preserved than when travelers and explorers like Howard Butler and Gertrude Bell visited it earlier in this century.
The role that Umm el-Jimal played in these historical developments is difficult to determine, because its ancient name was unknown. H. Butler and others have associated it with the Thantia of the Roman-Medieval map called the Peutinger Table (1913: 151). This, however, is a tenuous hypothesis at best. Henry MacAdam has recently suggested the name Surattha, an unidentified place name of a north Arabian town from Ptolemy's Geography, as a possibility (1986: 17). The meaning of the modern name, too, is enigmatic. Some insist it means "Mother of Beauty", but most translate it as "Mother of Camels". Anyone who has seen the herd of beautiful, thoroughbred camels wintering among the ruins, will agree that the two translations are in fact compatible.
Without the help of literary sources, the role of the settlement has to be understood from the ruins themselves. Umm el-Jimal is no Jerash. No emperor commissioned monumental buildings here. The place is plain, without frills. But this has great significance. Whereas Jerash is monumental, a symbol of the glory of Rome, Umm el-Jimal is ordinary, a symbol of the real life of Rome's subjects. Umm el-Jimal gives us a glimpse of local people, Arabs, Nabatæans, Syrians, living ordinary lives.
Whereas Jerash has monuments that awe us with their majestic beauty, Umm el-Jimal has ordinary houses that give clues to the habits of their residents. The architecture of these plain houses makes it clear that the residents were chiefly farmers. In most the bottom floors were barns and stables; the enclosed courtyards doubled as living rooms for humans and pens for animals. Outside, the elaborate water collection systems of both the second century village and the sixth century town, combined with the terraced fields along the wadis, indicate extensive cultivation. Both the ruins of the village and the standing buildings of the town indicate enough economic surplus from the export of animals and produce to nearby cities to enable the construction of the no-frills, but sturdy and comfortable houses that are so characteristic of the entire region.
This income may have been bolstered by caravaneering as well. Places like Umm el-Jimal and Umm el-Quttein are convenient stopovers for goods in transit from Arabia and the Gulf via the Wadi Sirhan and Azraq to Bostra and Damascus. To explain the very existence of these places as "Caravan Cities", however, is an exaggeration. The backbone of local economy and sustenance was the agriculture that flourished from the twin benefits of superb water engineering and excellent soil.
Who were these residents? It is tempting to call them Nabatæans, because the site has one of the largest numbers of Nabatæan inscriptions anywhere. However, indications are that Nabatæan was an adopted culture. The villagers who wrote the Nabatæan used much more Greek, another second language, if poor grammar and spelling are any indication; and practically none of the typically Nabatæan Pottery was used. The post-third century town people used no Nabatæan at all. The best interpretation is that the earlier village was settled by local Arab tribes under Nabatæanizing influence from the nearby capital at Bostra. These villagers probably spoke an Arabic or Aramaic dialect, and super-imposed both Nabatæan and Roman cultures on their own desert ways as they became sedentary. After the destruction of the Nabatæanized village in the third century, the builders of the new town were other regional Arabs who adapted and refined the earlier hydraulic and architectural technology, but had no interest in the Nabatæan language and religion.
It is also a misnomer to call the people of this later town Roman or Byzantine. True, the pottery they used was mostly typical of the Roman and Byzantine pottery of the Mediterranean. But the architecture and the way of life is indigenous, with only some Greek and Roman influences, and the names of people on the Greek inscriptions are predominantly Arabic (Sartre 1985). The answer to the question, therefore, is that the people of Umm el-Jimal belonged to local Arabic tribes who settled into rural communities in the context of Nabatæan, Roman, Byzantine and Umayyad political and cultural expansion and control.
Back to Top
Acknowledgements
Primary recognition must go to the Department of Antiquities for its generous logistical support and the energetic labors of its representations on the project. These included Ghazi Bisheh, Hafzi Haddad, Hussein Tayyib, Mifleh Gharaibeh and Amjad Bataineh. ACOR also provided unstinting logistical help and advice. Calvin College has acted as the home base for advanced preparation and field school accreditation. Ambassador College fit the field school into its archaeology curriculum in 1992. The project is affiliated with the American Schools of Oriental Research.
The following organizations are warmly thanked for their grants and other financial and in-kind support: The Department of Antiquities (1981), the Royal Jordanian Airline (1984), The Ministries of Social Affairs and Labor (1981, 1984), the Ministry of Education (1992), Calvin College (1981, 1984, 1992), The Calvin Alumni Association (1984, 1992), the Kyle-Kelso Archaeological Foundation (1981, 1984), National Endowment for the Humanities (1984), Baker Book House (1984) and the Ambassador Foundation (1992).
Senior staff over the three seasons included (in alphabetical order): John Betlyon, Ghazi Bisheh, Joanne Bloom, Robin Brown, Vincent Clark, Robert Coughenour, Bruce Dahlberg, William De Kraker, Laurette De Veaux, Sally de Vries, Bert de Vries, Jennifer Groot, Anne Haeckl, Frank Koucky, S. Thomas Parker, Paul McDermott, Robert Schick, Richard Sherrod, Michael Toplyn, and David Vis. We still mourn the passing of Jennifer Groot and miss both her great stratigraphic skills and her warm friendship. These excellent team members were supported by a total of 49 graduate and undergraduate students, 21 of whom were enrolled in Calvin College's Field School and 8 in Ambassador College's Field School. A total of 73 Umm el-Jimal village people were employed over the three seasons.
Back to Top
Bibliography
Butler, H.C.
1913 Architecture. Syria. Publications of the Princeton University Archaeological Expedition To Syria (PPUAES) Div. II, Sect. A, part 3. Umm idj-Djimal. Leiden: E.J. Brill.
de Vries, Bert
1979 Research at Umm el-Jimal, Jordan, 1972-1977. Biblical Archaeologist, Winter: 49-55.
1981 The Umm el-Jimal Project 1972-77. Bulletin of the American Schools of Oriental Research 244: 53-72.
1982 The Umm el-Jimal Project 1972-77. Annual of the Department of Antiquities of Jordan XXVI: 97-116.
1985 Urbanization in the Basalt Region of North Jordan in Late Antiquity: The Case of Umm el-Jimal. Studies in the History and Archaeology of Jordan II. Ed. Adnan Hadidi. Amman: Dept. of Antiquities of Jordan: 249-56.
1986 Umm el-Jimal in the First Three Centuries A.D. Pp. 227-41 in The Defence of the Roman and Byzantine East. Eds. Philip Freeman and David Kennedy. Oxford: BAR International Series 297.
1988 Jordan's Churches: Their Urban Context in Late Antiquity. Biblical Archaeologist, December: 222-6.
1989 Umm el-Jimal. Pp. 590-7 in Archaeology of Jordan II 2. Field Reports. Sites L-Z. Eds. D. Homes-Fredericq and J.B. Hennessy. Leuven: Peters.
1990 Umm el-Jimal: "Gem of the Black Desert". Amman: Al-Kutba Publishers.
Littman, E.
1913a Greek and Latin Inscriptions. Pp. 131-223 in Syria. PPUAES, Div. III, Sect. A, part 3. Umm Idj-Djimal. Leiden: Brill.
1913b Arabic Inscriptions. Pp. 1-3 in Syria. PPUAES, Div. IV, Sect. D, Umm idj-Djimal. Leiden: Brill.
MacAdam, H.
1986 Studies in the History of the Roman Province of Arabia. Oxford: BAR International Series 295.
Parker, S.T.
1986 A Tetrarchic Milestone from Roman Arabia. Zeitschrift für Papyrologie und Epigraphik 62: 256-8.
Sartre, M.
1985 Le peuplement et le développement du Hauran antique à la lumière des inscriptions grecques et latina. Pp. 189-202 in J.-M. Detzer, ed. Hauran I. Paris: Libraire Orientaliste Paul Geuthner.
Back to Top
 |
|