Umm
el-Jimal
The Umm el-Jimal Project, 1972-1977Annual of the Department of Antiquities of Jordan. 26. Amman. 1982. Pg. 97-116. Same in Bulletin of the American Schools of Oriental Research. 244. 1981. Pg. 53-72.Bert de Vries TABLE OF CONTENTS
Umm el-Jimal, located 20 kilometers east of Mafraq in the semi-arid lava region of the southern Hauran, ranks behind Petra and Jerash in size and state of preservation among Jordan's archaeological monuments. However, it differs dramatically from those two monuments because it is built entirely of somber-colored steel gray basalt, which gives it the appearance of a bombed out and burned modern city from a distance. The great strength of the basalt stone lent itself to an unusual style of architecture: corbeling. The ancient builders were able to construct ceilings and roofs by laying stone beams two to three meters long on cantilevered supports protruding from the walls. The application of cantilevering to the construction of stairways eliminated the need for cumbersome and costly support structures and increased useable living space on the lower floors. The success of this building method impresses itself on the visitor by the fact that numerous buildings are still standing two to three stories high with some of their ceilings intact after 1500 years. The city was founded in the Early Roman period when it enjoyed considerable Nabataean influence, flourished as a frontier city of the Roman and Byzantine Empires, and continued to prosper in the Umayyad period, perhaps because of its proximity to the Desert Castles and the Pilgrim Route. Umm el-Jimal was destroyed by earthquake at the end of this period and not rebuilt because the region of the Hauran lost its preeminence when the seat of government shifted to Baghdad under the Abbasid Caliphs. Most visitors to Umm el-Jimal are surprised to learn that its ancient citizens were not Roman or Greek, but Arab. Numerous gravestones, although written in the Latin and Greek scripts, have given us a catalogue of names of the inhabitants. The names indicate that the residents were local Arabic nomads who settled in this region and built Umm el-Jimal under the security provided by the Roman Empire-a process that has been repeated by their modern descendents under the security of the Jordanian Government. I. Previous ResearchPrior to 1972 a minimal amount of work had been done at Umm el-Jimal. The first major project was the survey by the Princeton University Archaeological Expedition to southern Syria in 1904-05 and 1909 directed by Howard C. Butler (Butler 1913). This expedition produced a map of the city, including plans of 40 major buildings. The resulting publication included detailed architectural and historical analysis of the buildings surveyed and a record of the numerous inscriptions found in and around the city. This publication has served as the basic reference on the site ever since. In the late 1930s and early 40s Nelson Glueck made several visits to Umm el-Jimal in search of evidence for his thesis that the city was a major link on the Nabataean trade route from southern Jordan to Syria via the Wadi Sirhan (Glueck 1939: 140-46; 1942: 3-8; 1944: 7-17; 1951: 1-34). From the absence of Nabataean pottery at Umm el-Jimal and other Syrian sites he concluded that Nabataean influence (attested by numerous inscriptions) there was in the form of commercial and strategic interest rather than in the form of dense settlement (Glueck 1951: 13, 17). In 1937 G. Horsfeld published aerial photographs of Umm el-Jimal that are extremely useful for studying the process of decay and human alteration that has taken place from the time of Butler’s survey to the present (Horsfeld 1937: 456-60). In 1956 G.U.S. Corbett did an architectural study of the "Julianos" Church in order to test Butler’s conclusion that this church had been built in A.D. 344. Corbett concluded that the Julianos Inscription from which the date was derived was actually a funerary inscription reused secondarily in the building and revised the building date of the church to the early 5th century on the basis of architectural parallels (Corbett 1957: 39-66). Back to TopII. Purpose of Present Research
During his first visit in 1971 the author was amazed that a site of this magnitude and in such an excellent state of preservation was not being worked on and he decided it needed to be "rescued" from oblivion. The first step was the completion (and correction, if necessary) of the Princeton Survey, which had dealt with only onefourth of the existing architecture. Next it was necessary to begin stratigraphic excavation in order to test the various chronological, historical, and cultural conclusions that earlier scholars had drawn from their study of remains above the surface. Finally, it was important to study the historical-geographical context of Umm el-Jimal as one of many similar settlements in the southern Hauran. The effort to complete these goals developed into a five-phased project over an 11 year period, from 1972 to 1983.
The results of the three completed phases will be summarized below. Back to TopIII. The Architectural SurveyThe survey was carried out on a part time basis from 1972 to 1974 while the author was the W. F. Albright Fellow of the American Schools of Oriental Research (1972-73) and Research Associate of the American Center of Oriental Research (1973 -74). (see note) The southeast quadrant of the city, including all of the buildings south of the main reservoir and east of the central open space, was mapped wall for wall with theodolite and tape with the help of one assistant. In the spring of 1973 the author and the Aero-Precisa Co. of Beirut completed an aerial survey of the site, under the auspices of the Ministry of Municipalities and Rural Affairs and with funding from the Jordanian Department of Antiquities. After this it was possible to complete the mapping of the remaining two-thirds of the city much more rapidly by combining the photograph and topographical map produced by Aero-Precisa with ground control measurements. The results of this two-fold process are the maps published with this article, one showing the city plan in detail, while the second shows the buildings schematically for identification purposes. For ease of cross reference, the names given to buildings by Butler have been kept. In cases in which these reflect a now questioned interpretation, names have been put in parentheses. For the 20 housing complexes mapped by Butler his Roman numerals have been retained, but for the rest it was necessary to switch to Arabic numerals to preserve space. In comparing the results of the new survey with that of the Princeton Expedition it must be stressed that the author considers his survey completion rather than correction of that expedition's work. The main impression of the earlier survey is that the fieldwork was generally accurate and that the published drawings are of outstanding quality. In this respect the major change in the new map is the addition of 107 housing complexes and two housing insulae (133 and 134) that were too badly ruined to survey in detail. The shape and size of the overall city plan drawn by Butler is generally identical with the new one, with several important exceptions. The location of the city wall between the Southwest and West Gates was misplaced, due perhaps to the incompleteness of the survey at that point. The Princeton map shows the main aqueduct originating from behind the dam to the northwest of the city. Actually this aqueduct was fed by a channel that runs for many kilometers up the sloping terrain to the northeast. A second channel from the northeast supplied reservoirs in the northern part of the city. At the southern end the aqueduct does not end in the cistern of House XX but continues to the cistern in House 132. The aqueduct crosses the city wall at the approximate point of the South Gate on the Princeton map. Because there is no evidence for a gate anywhere in this section of the wall, it may be possible that Butler mistook the entry of the aqueduct for a city gate.
In the plans of individual buildings the Butler map shows a general tendency toward layout on a ninety degree grid. However, with the exception of the "Praetorium" and several churches, almost all buildings are laid out haphazardly. This is sometimes due to the carelessness of the builders (e.g. the Barracks), but most often it is in order to take advantage of available space (many housing complexes). There are a few cases of discrepancy concerning the locations of walls and doorways. It is often impossible to determine whether these are the product of inaccuracy or of remodeling since the Princeton Survey was completed. The most significant differences between the Princeton Survey and the author's are due to reconstruction and quarrying activity in this century. From both the Princeton publication and analysis of surface artifacts, it is clear that between the Umayyad Period and 1904 the city remained almost totally undisturbed. In the past 70 years, on the other hand, a variety of human actions has caused major changes at Umm el-Jimal. The fact that Butler had the opportunity to study the city before these modern changes took place makes his publication an invaluable record that cannot be duplicated today. Major alterations in the structure of many buildings resulted from a Druze settlement of the site that began between 1905 and 1909 (Norris and Stever 1930: 34, 91) and lasted for three decades. This Druze community worked energetically at the reconstruction of the ancient buildings in order to make them habitable. This work is most obvious in the numerous ceilings and roofs that are still intact today. These are often mistaken as ancient because the Druzes were so adept at copying the ancient methods of corbeling and erection of arched partitions. Their work can usually be distinguished from the ancients', however, by the irregular masonry at the tops of walls, the mixed shapes of corbels and ceiling beams, and fresh cut marks on the voussoirs of the arches. In some cases, identification of Druze construction is also possible by comparison with the Princeton publication. Butler's plan and photograph of the room with the screened west wall in House XIII (Butler 1913: 203, fig. 205:183) is without the transverse arch that covers the middle of the screened wall and supports the ceiling today. Butler's plan shows the "Nabataean Temple" as a simple rectangular room with a twocolumned pro-style porch (Butler 1913: fig. 155: 131). Today the room contains two transverse arches that support the ceiling, while the porch is no longer apparent on the surface. Examination of the standing architecture leads one to conclude that everything except the front wall with the triple doorway is contemporary with the Druze reconstruction of the building. The survival in several buildings of temporary walls used to support voussoirs during arch construction is evidence that the remodeling was still in process when the Druzes abandoned the city in the 1930s. In the 1920s, before agreement was reached on the Mandate border, a French army detachment was camped in and near the Barracks, according to the present villagers. It left behind raised tent platforms and a basalt walk in the Barracks courtyard, as well as whitewash on some doorposts and lintels in rooms near the tower. Similar tent bases also remain in House 47. About the same time the tribe of the present villagers began to settle in the area. Most rooms restored by the Druzes were used as stables and barns until the government fenced off the site in 1975. Recent government construction includes the paving of the road through the southeast comer, the restoration of the main reservoir, and the building of a dam in the wadi several meters to the south of the remains of the ancient dam. Although domestic water is now supplied to the villagers from the Azraq line, the reservoir is still in great demand for the watering of flocks of sheep, goats, and camels. Significant changes have also taken place since 1904 as the result of natural collapse and quarrying. The Numerianos Church, today ruined almost to ground level, was preserved to roof height at the time of the Princeton Survey (Butler 1913: figs. 192:171-173). Some accessible wall sections of good quality masonry, such as the south wall of the "Praetorium" courtyard (Butler 1913: fig. 160:140) and the east city wall near the East Gate (Butler 1913: fig. 158:137), have almost entirely disappeared. The Barracks especially have suffered much recent damage. The interior tower floors and stairs were destroyed in 1970; a section of roof adjacent to the tower collapsed in 1973; and the large V-shaped gap in the north enclosure wall was created by a combination of recent collapse and quarrying. Sections of the exterior aqueduct, described as intact by Butler (Butler 1913: fig. 160:139), are now detectable only as a depression in the terrain. Howard Butler's estimate that most of the domestic architecture dates from the Late Byzantine period is correct (Butler 1913: 196; see below for stratigraphic evidence). In fact, the completed map represents the city as it was inhabited in the 6th, 7th and 8th centuries. Because of extensive remodeling of apparently older structures, not much can be said about the nature and extent of the earlier Roman and Nabataean settlements. Butler's interpretation of the function of the domestic architecture and his population estimate based on that can now be revised on the basis of the more complete survey. His conclusion that "only a few houses have stables, and there is no evidence that any considerable number of domestic animals were housed within the city gates" (Butler 1913: 195) is substantially incorrect. It is in fact difficult to find housing complexes that do not include at least one stable with a row of five to ten mangers built in. These long, narrow, low-ceilinged stables are usually adjacent to a large, arched room with which they form an L-shape and share a set of mangers. Corbett's interpretation of these large rooms as barns or stables (Corbett 1957: 43-45, fig. 44:3) is more plausible than Butler's view that they are the core rooms of a family's living quarters (Butler 1913: 195). The fact that many of these large rooms have a latrine built into one corner need not preclude their use as barns; one can imagine the advantage of locating such a facility away from the living quarters. An especially fine example of this stable-barn combination is in House XIII (Butler 1913: fig. 203:182). In this case a row of four well-built mangers is separated from the large room by means of a screen constructed of well-dressed, thin slabs of basalt perforated by cone-shaped openings. Butler's attempt to identify these rooms as a small industrial establishment (1913: 204) ignores the numerous parallels of the stablebarn arrangement, of which this is merely a more elaborate version. The perforations in the partition were necessary simply to supply air to the animals tethered in this otherwise closed section of the stable. The doorway from the large room to the courtyard of House XII (shown much too large in Butler's drawing) is too small for people or larger animals but adequate for sheep and goats. It seems likely, therefore, that the long stable with the mangers was used for larger animals (cattle, horses), while the larger room provided indoor shelter for sheep and goats.
From the above it is clear that in a typical housing complex much of the ground floor space was devoted to animals. The people lived on the second and higher floors and had access to their living quarters by the exterior stairways. One would have to agree with Corbett "that cattledealing was one of the principal occupations of the inhabitants" (Corbett 1957: 40), at least in the Byzantine period. Further substantiation for this lies in the fact that much of the open space in the city was taken up by walled enclosures. Many of theseenclosure walls have not been included on the map because they are barely visible at ground level. However, the walls enclosing the area between the Barracks, the Numerianos Church, and House 119 are typical (fig. 5). It is possible that what appears as open space today was partitioned in the Byzantine period and served as outdoor animal pens. Butler's population estimate of from 7,000 to 10,000 people (Butler 1913: 195) will have to be reduced because it is based on the inclusion of numerous stables and barns as living space for people. Even the conservative estimate that 25% of this space was devoted to animals would lead to a new figure of about 5,000 as the population of Umm el-Jimal within the city walls in the Late Byzantine period. Back to TopIV. Stratigraphic Soundings, 1974In the week of June 16-22, 1974, four small probes were excavated in order to obtain a stratigraphic check on the periods apparently represented by both the architecture and inscriptions on the surface and to test the feasibility of carrying out more extensive excavation in the future. (see note) The four probes were located in various types of open spaces throughout the city. They were deliberately placed away from architecture in order to be able to achieve rapid progress in simple soil layers. Area P.1 was located 80 meters east-southeast of the Barracks Tower in the large open space now intersected by the modern pavement. The 2 X 2 m. square was excavated to bedrock 0.30 in. below the surface. Area P.2 was located in the courtyard of House I in the southwest corner of the city. This I X 2 in. square was excavated to virgin soil 1.60 in. below the surface. Area P.3 was located 80 meters north of the large reservoir (no. 9) on a three-meter high mound in the open space between Houses 76, 78, 80 and 133. This I X 2 in. probe was excavated to virgin soil 2.70 in. below the surface. Area PA was located 50 meters south of the southeast corner of the large reservoir (no. 9) in a raised area contained by a retainer wall and adjacent to an artificial pool on the north of Houses 91 and 92 = XIV. This I X 2 m. probe was partially excavated to a depth of 1. 10 in. below the surface. On the basis of Jim Sauer's pottery analysis the following outline of the occupational history of Umm el-Jimal was determined:
Back to TopV. Excavations in the Southern Half of the City, 1977.The 1977 season of excavation, carried out in July, was part of a six-month project done jointly by the author and the Department of Antiquities of Jordan. In addition to the excavation, the project included the consolidation of the Barracks. (see note) The goal of the excavation was to study a representative sample of buildings in the southern half of the city, with the northern half to be done in a future season. Specific structures were selected according to three functional motifs: public buildings except churches (also left for a future season), the city's defense, and domestic buildings. With this in mind, six squares were opened in the Barracks, A. 1, 3-7, in order to verify its building date, study the history of its remodeling, and, if possible, determine the function of the various components of the structure. Two squares were placed against the city wall, one, A.2, against the south wall and the other, B.2, against the west wall in order to determine the building date and reconstruction phases. One square, A.8, was placed on the location of the porch of a building Butler had called a Nabataean Temple (Butler 1913: 155-56) in order to test that thesis. Three squares were placed in the "Praetorium," B.1, 3, 5, in order to determine the building's date and history of usage and to test Butler's guess that the building was a praetorium (Butler 1913: 164-66). One square, B.4, was placed against the "Praetorium" courtyard wall in order to check its chronological relationship to the building proper. Two squares were located in House XVIII. This partic ular domestic complex was selected not only because it is a typical house, but also because its outstanding quality of construction and decoration make it a primary candidate for consolidation and reconstruction. The first square, C.1, was located in a west room in order to check floor construction while the second, C.2 , was placed in a basement reservoir under the main floor of an east room (P1. 1). All of these squares were treated as soundings for stratigraphic sampling. They were, therefore, kept as small as was practically possible, for in no case was the clearing of an entire building or room an excavation objective.
The brief summary of the results will follow the stratigraphic scheme determined by the 1974 soundings. This summary will follow the interpretations of Area Supervisor Tom Parker for the Barracks, the city walls, and the so-called Nabataean Temple (Parker 1978), and those of Area Supervisor Robin Brown for the "Praetorium" and House XVIII (Brown 1978). It should be noted that the almost total absence of foundation trenches made the dating of walls difficult. In some cases dates were based on material in the soil layer under the foundations, and, where possible, dismantling was resorted to.
In looking back at the above chronological summary of the excavated remains, two concluding observations are worth making. First, the city's long standing reputation as a significant Nabataean site has obscured the fact that it really prospered in the Late Byzantine and Umayyad periods. Second, enough problems remain with the so-called Nabataean Temple, the "Praetorium," and House XVIII to warrant continued investigation in the next season. Refined definition of the various strata of occupation will be the major goal of the 1981 season of excavation. Back to TopNOTESThe work was financed with the stipend of the Albright Fellowship of the American Schools, the author's salary as director of the Christian Reformed World Relief Committee and his wife's salary as teacher at the American Community School, and a grant from the Calvin Foundation. Thanks is in order to these organizations for their support, to the Jordan Department of Antiquities for its cooperation, and to the succession of volunteers on the other end of the measuring tape, especially former student Gary Rozeboom. Back to SectionThe project was carried out under the same sponsorship and financing as the architectural survey. In addition to the author, the Staff consisted of Jim Sauer, Co-director and Ceramicist; Hikmat Ta'ani, Department Representative; Zee Aime' Dick Dorsett, Gordon Malarckey, and Sue Sauer, Square Supervisors. Back to SectionIn exchange for the author's services on the restoration project the Department hired the laborers needed on the excavation. In addition to the Department's generous contribution, the excavation was funded by grants from the KyleKelso Fund and the Calvin Foundation, participants' camp fees and course tuition, Sally De Vries's salary at the American Community School, and the author's sabbatical pay from Calvin College. The project was affiliated with ASOR and received a great deal of assistance from ACOR, ranging from excavation equipment to logistical support and advice on pottery analysis from Director Jim Sauer. The financial support and personal involvement of so many organizations and individuals was overwhelming. I am deeply grateful to all. The core staff was a talented group of excavators and specialists, almost all of whom did double duty by supervising the digging in addition to their specialty. Robin Brown and S. Thomas Parker teamed up as chief stratigraphers, with Robin taking responsibility for the "Praetorium" and House XIII, while Tom supervised the Barracks, the city wall, and the so-called Nabataean Temple. In addition Tom Parker did the ceramic analysis. The rest of the core staff and their specialties were: Sally De Vries, Administrative Director; Jennifer Groot, Object Registrar; Bud Haggard, Pottery Registrar; Frank Koucky, Geologist; Paul McDermott, Epigrapher; Jim Sauer, ACOR Advisor; Hikmat Ta'ani Department Representative and Foreman; Mike Toplyn, Osteologist. These were assisted by a very able group of novices, six of whom participated for course credit at Calvin College. They were: Richard Abma, photographer and square supervisor; Roger Brummel, photographer; Craig Bultsma, square supervisor; Laurette De Veaux, square supervisor; Shelley Dunn, square supervisor; Charles Fondse, surveyor; Barbara Oppewall, square supervisor; Keith Vander Laan, surveyor. Daughters Tara and Tanya De Vries also participated full time as assistant square supervisors and pottery registrars. The marvelous cooking of Mohammed Adawi contributed to both the good health and morale of this group. Back to SectionBibliographyBrown, R. M. (1978)
Butler, H. C. (1913)
Corbett, G. U. S. (1957)
Glueck, N. (1939, 1942, 1944, 1951)
Horsfield, G. (1937)
Littman, E., et al. (1913)
Norris, F. A., and Stever, E. S. (1930)
Parker, S. T. (1978)
Sauer, J. A. (1981)
Back to Top |
 |
|