Umm el-Jimal

THE 1994 UMM EL-JIMAL CEMETERY EXCAVATIONS: AREAS AA AND Z

Annual of the Department of Antiquities of Jordan. 39. Amman. 1995. Pg. 457-468.

Janet Brashler

TABLE OF CONTENTS

• Introduction
• Excavation Strategy
• Progress of Excavations
• Summary of Results
• Area AA
• Area Z
• Comparison of Areas AA and Z and Future Research
• Bibliography

Location of Umm el-Jimal Area AA (Drawing by Herve' Irion and John Soh).

Introduction

During the 1994 field season at Umm el-Jimal, excavation of burials was undertaken with the objective of providing a systematically collected sample that could be used to address a number of questions regarding the inhabitants of the site. Of particular interest was the role that plague may have played in the depopulation of the site during the late antique period. De Vries (1981, 1982, 1993, 1995) and others (Cheyney 1995) have suggested that endemic, cyclical plague contributed to the depopulation and instability of the community during the centuries prior to the great earthquake of A.D. 747. While the community continued into the 9th c., buildings that had collapsed during the earthquake were not restored (de Vries 1995) possibly because there simply were not sufficient people to undertake the job. Therefore, a primary objective of the 1994 excavations in areas AA and Z, was to collect information from tomb excavations that could address the question of site abandonment. This information is also useful for questions about the general health, diseases, burial customs, physical characteristics, demography and socio-economic status of the residents of Umm el-Jimal.

Areas selected for excavation were chosen based on prior knowledge of tomb locations. Excavations during 1993 (Cheyney 1995) revealed tombs in two areas, though other areas of the site also produced tombs (Butler 1913, de Vries 1981, 1982, 1993, 1995). Area AA, located near the Umm el-Jimal girls' school approximately 200 m W of the main area of ruins is a level space with no evidence of tombs on the surface, though several small depressions from an old olive grove are visible. It includes a roughly semi-circular stone wall remnant, several modern fence rows and late 20th-c. construction including the girls' school of Umm el-Jimal, a house built in 1993 and roadways. Area Z, situated in the olive grove of Sheikh Hail es-Serour on the E side of a wadi, approximately 75 m W of the S limits of the main part of the ruins today, is covered with olive, grape and pomegranate, a small shed and is bounded by the Sheikh's house and driveway to the W and two roads to the E and N.

Back to Top

Excavation strategy

Because tomb location was unknown and unpredictable from surface examination, initial excavation in area AA was undertaken relying on three different strategies. Six 1 by 6 m trenches (AA.2, 3, 4, 12, 13, 14) were excavated at an orientation 45 degrees E of N. Each trench was separated from the next by a 1 x 1 m balk that was not excavated. This series of trenches roughly spans the length of the open area behind the girl's school from one end of the rock wall remnant on the SW to the NE. Tombs were located in 4 of the 6 trenches (AA.2, 3, 4 and 14). In addition to these trenches, squares were excavated relying on the local tomb robbing method of discovery. A thin steel rod pounded into the hard soil of this area produced differential resistance based on the presence of tomb cover stones in several cases. Squares AA.5 through AA.11 were situated using either this method, or were located based on surface evidence of concentrated rock. Of these, 3 squares contained tombs: AA.8, 9 and 11. AA.10, the only square of these not located in the immediate vicinity of the trenches was placed E and N of the remnant rock (cemetery?) wall in an area that was slightly higher and contained abundant surface rock. Finally, two other squares, AA.15 and 16 were located in areas where judgment suggested that there should be tombs based on architecture extending into balks. Square AA.15 connects 2 squares (AA.8 and AA.9) and AA.16 is a 1 by 5 m trench that proceeds on a N-S axis from AA.8 toward AA.13 and AA.14.

Area AA showing locations of trenches and tombs (Drawing by Herve' Irion and John Soh).

Back to Area AA Overview Map

In Area Z two squares were located on prior knowledge of where tombs were located in the cemetery. Previous excavation of Z.1 by Cheyney (1995) and the landowners revealed a pattern of tombs. Squares were located between already excavated tombs by probing for cover stones with a steel rod. The two squares laid out (Z.2 and Z.3) were located adjacent to each other and had beginning dimensions of 2.2 x 3.0 m based on the results of the probing.

Back to Top

Progress of excavations

Excavations on both areas AA and Z produced a consistent stratigraphic sequence of soils deposited through natural processes, interrupted by tombs excavated in antiquity to eroded bedrock. Excavation proceeded by removing 1 to 4 loci (0.4-1.0 m) of geologically deposited red-brown aeolian soil until possible burial pits were identified, or until tomb architecture (cover stones) was encountered. In most cases, tombs extended beyond the limits of trenches so that probes had to be excavated to expose their full extent. Articulated tomb cover stones were drawn and removed, revealing in most cases a pit defined by less compact soil, or tomb side architecture. Fill that had seeped in above burials was removed following the pit outline or side architecture. Burials were articulated in situ, photographed, drawn and field determinations of age, sex, obvious pathologies, and burial position vis-a-vis the tomb and any grave objects were recorded. In the case of tomb Z.3, repeated mapping, photographs and careful labeling were required to document multiple individuals present. Fill immediately surrounding most burials was screened with 3 mm mesh screen, especially in areas where beads or other artifacts were found. Soil samples were taken from the region of the os coxae when possible. As bones were removed, they were wrapped in newspaper and labeled.

In-field laboratory work included repacking and registering burials from the area. Registration included determining which bones were present, more refined aging and sexing and additional examination of skeletons for pathologies. No wet cleaning was done, though dirt accumulated in the calivaria was removed, and some bone was dry brushed. All skeletal material was packed for shipment. Due to time constraints several burials from Area AA and all of the Z burials were not registered in the field. Analysis of these materials is currently underway by M. Cheyney at Western Michigan University.

Back to Top

Summary of results

Area AA

Table 1 presents descriptive data on burials from Area AA. Excavation produced evidence of 11 tombs containing the remains of 13 individuals. One small tomb had no preserved skeletal remains. All tombs were identified by the presence of cover stone architecture located beneath between 1 and 4 loci of naturally deposited wind blown silt. The upper layers of deposits contained modern debris (in the top 0.1 m) but only ceramics from LRom through Byz were found below in both the naturally formed soils and infrequently in the tombs. Sherds all but disappear more than 0.4 m below the surface in squares excavated to deeper levels in search of tombs. It seems likely that tombs in Area AA originated in deposits that are currently 0.2 to 0.4 m below the surface. In several cases (AA.3A and B and AA.16) a pit containing soil much softer than adjacent soil was identified lying above the cover stone architecture, suggesting that pits were dug, burials put in pits or cists and covered with stones and then covered with loose fill. It is difficult to ascribe any precise dates to the tombs based on the few sherds that were found in the fill that had dribbled in. With a few exceptions (sherds on the surface primarily), these sherds appear to come from mixed LRom through Byz, but most likely LRom/EByz use of the area for purposes other than burial was very minor considering the density of materials recovered from excavations in the main portion of Umm el-Jimal.

[above] Ceramic juglet with copper wand

[below] "Heirloom" chalice, plaster moulded around core of Byzantine cooking pot

Grave objects found within only three of the tombs, (AA.3B, 4, and 15A) are not diagnostic, or temporally sensitive including a copper or brass ring, an array of glass, limestone and frit core beads, and a pair of gold boat-shaped earrings, which are of a style apparently found in Nabataean through Byzantine contexts elsewhere including the Amman Airport site (Ibrahim and Gordon 1987). None of these objects is likely to afford more accurate dating for the three tombs.

Though there are only 11 tombs and 13 individuals, considerable diversity exists in the burial sample. Four different patterns of burial present include the following: 1) burial of individuals in unlined pits with heads to the W, with no grave objects, as single interments; 2) single individuals buried in unlined pits, heads W, with associated grave objects; 3) individuals buried in well-constructed stone-lined cists, with no objects; and 4) individuals buried in or on coffins without objects. Most burials in simple pits were likely wrapped in shrouds. Two burials suggest post-burial tomb disturbance, possibly robbery. In one case (AA.9), an 8-year-old was found at the east end of a well-built coffin, more than adequate for the body that was dismembered at the 4th thoracic vertebrae, roughly in the middle of the back. Other body parts were also disarticulated, though some portions were intact, suggesting that disarticulation occurred before soft tissue decomposed. In a nearby tomb, one individual found inside a coffin was only partially articulated, and a second individual was casually placed on the lid of the coffin. No objects were found associated with either tomb.

Burial of single individuals in one tomb is modal, but two cases occur of similar ages, viz., sub-adult individuals interred in the same tomb. In one tomb, a simple stone-capped pit, two infants approximately one year in age were buried together, probably at the same time. The second example is the burial of two 5-year-old children in an unusually well-prepared stone-capped cist tomb. The cover stones and side architecture of the tomb were made of carefully dressed stone, and the inside walls of the cist were coated with mud plaster.

None of the patterns described above predominate nor do they present the image of affluence typical of LRom, EByz loculus burials at sites such as Jerash, Pella and Umm el-Jimal itself. Less than 250 m NE of the area excavated is the large chamber tomb of Sareidos. Other chamber tombs and well-constructed cist tombs have been identified or excavated in other areas of the community (Butler 1913, de Vries 1993: 444-5, Cheyney 1995). Compared to preliminary analysis of other excavated tombs at Umm el-Jimal (see below and Cheyney 1995), tombs excavated in area AA in 1994 seem to contain the earthly remains of people who had access to the fewest resources in the Umm el-Jimal community. If access to resources is a measure of status, then the burials excavated in Area AA represent low or poor status. The possibility that the variation observed in Area AA is time dependent rather than class dependent does not seem likely, given the apparently planned nature of the cemetery area, and proximity to the elaborate chamber tomb. Within the 1994 excavated tombs, some attributes do segregate by area (all of the coffin burials and stone-lined cist burials occur in the NE portion of the 1994 excavated area) but grave goods, skeletal orientation, and other characteristics vary across the area. In sum, Area AA presents one of the first systematic examinations of a low-status LRom-EByz cemetery.

Back to Top

Area Z

Tombs in Z.2 and 3 were located 0.7 and 1.0 m below the present surface, covered by naturally deposited fine red-brown silt with little pedogenesis of caliche layers above the tombs. Though predominantly dry, some soil was moist due to irrigation of the surface. Tomb Z.2 contained the remains of four individuals located in a wood coffin lacking a lid covered with field-dressed basalt cover stones. The individuals in the coffin included two well-preserved, articulated infants found in mostly extended position at the E end of the coffin. Both infants had heads at the E and occurred stratigraphically above the other two burials, those of adults who occupied a position more central in the coffin. With the two adult burials, orientation of the heads would have been to the E, but the heads were removed and placed on the mid-section of the individuals. Excavators suggested that the tomb was robbed, based on the lack of grave goods (only a single large bead was found near the wrist of one adult) and the disturbance done to the skulls of the two adults. An alternative explanation is that the two adult burials were buried first, the coffin was then re-opened for the two infants, at which time the heads of the adults were moved. The latter interpretation seems plausible, since the first individuals were not disturbed with the exception of the heads, and other examples of tomb robbing such as Z.1 (Cheyney 1995) show much disturbance to the skeletal material. As in the case of AA.9, the post-mortem alteration to the adults seems to have happened while the bodies were still articulated.

"Heirloom" chalice with glass bottleneck in situ in Area Z.3	Closeup of glass bottleneck
Earrings, gold with stone inset

Z.3 is the most complex tomb encountered during the 1994 excavations. Tomb construction itself is well executed, with carefully dressed and chinked basalt cover stones overlying a neatly laid series of cobbles above large well-dressed side stone architecture. At the W end of the tomb, a series of dressed and fitted stones present the image of a "monument" or complex headstone that may have marked the cist location prior to placement of the tomb cover stones. Deteriorated fragments of plaster from the surface of side stone architecture occur in the fill soil within the tomb. Within the cist is a wooden coffin consisting of a thin boarded construction, most of which had deteriorated (especially at the W end) leaving wood traces and impressions in the soil, especially at the E end of the cist. The burials occur stratigraphically within this structure, which is covered by a large wooden cover constructed of beams 0.08B0.12 m thick on four sides and thin slabs lapped over the tops of the end beams. The coffin was held together with metal brackets found at each of the four corners near the top of the cover section and the top and bottom of the thin box section. Stout metal dowels served to fix the coffin parts together. Preliminary analysis of one beam by P. Bikai indicates it may be wild pistachio. The wood is relatively well preserved, and it will be possible to date the age of the beam(s) using dendrochronology or radiocarbon.

Remains of at least 14 individuals based on a field count of intact skulls ranging in age from infant to adult occur in the tomb. This number does not include remains of possible fragile neonates or others lacking skulls. Burials all appeared to be articulated and oriented in two directions predominantly with heads at either the E or W end of the coffin. Only a few sub-adult individuals had heads which occupied a more central position in the coffin. Bodies appear to have been placed so as not to disturb the heads of individuals already in the tomb.

Little or no post-mortem disturbance of any of the burials (with the exception of settling and in filling of the coffin by fine silt soil) is indicated by the presence of a variety of grave objects including an Early Byzantine cook pot remodeled into a chalice with plaster, a mirror and make-up palette made of glass and plaster, a tear drop glass bottle with snake motif on the neck; a head ornament consisting of woven or braided copper and bone hairpins; a bone "make-up" bottle with copper spatula; a juglet with copper spatula, several bracelets and rings of copper and iron; and numerous glass beads of various sizes, colors and construction. Beads were most frequently associated with sub-adult individuals in the tomb. Some beads were extraordinarily small, and may represent items attached to clothing or shrouds of the individuals with whom they are associated (de Vries, personal communication). The depth of 0.75 m from the top of the side architecture to the bottom of the coffin, and 0.49 m within the coffin itself, suggests that Tomb Z.3 was used over a period of time long enough to allow settling of remains. Several lines of evidence have yet to be explored including genetic studies which may indicate whether this was a family tomb, or a tomb of strangers.

Z.2 and Z.3 differ from each other markedly in construction, number of burials, burial orientation, associated grave objects and perhaps length of use. In all cases, Z.3 presents evidence of a more carefully executed type of burial than Z.2. It is possible that Z.3 and Z.1 (Cheyney 1995) were more similar to each other, but the robbing of Z.1 in antiquity prevents close comparison of Z.1 with many attributes of Z.2 and 3. Given the apparent presence of loculi tombs, cist tombs and coffin-lined tombs in area Z, a range of socio-economic level may be represented within this cemetery area. The difference between Z.2 and Z.3 is sharp, but perhaps not as sharp as that between a loculus tomb and Z.2 would be. Without final analysis, Z.2 appears to be a tomb of relatively poor individuals while Z.3 appears to contain remains of people with access to more resources than Z.2. Chamber/loculi burials could be indicative of the most prosperous residents of Umm el-Jimal. Specific comparisons remain to be completed for each of these tombs and an over arching interpretation needs to be undertaken of all known tombs at Umm el-Jimal though some brief comparisons are presented below.

Back to Top

Comparison of areas AA and Z and future research

Excavation in Areas AA and Z during the summer of 1994 produced a variety of different tomb construction styles, burial orientations, associated grave objects and other information which from this preliminary analysis present a picture of considerable diversity. Area Z contains remains of perhaps average to more prosperous members of the community, though tomb Z.2 is more modest than others. Rather, Z.2 is much more like several tombs from Area AA in its construction and contents. But do the elaborate construction and the presence of more numerous objects outweigh the presence of gold earrings in AA.15A? Care should be exercised not to impose Euro-American value systems (gold, more elaborate construction = "high class") on burial programs from 1500 years ago. Using an ethnographic interpretation, the gold earrings may be indicative of poorer more mobile nomadic women who carry the extent of their wealth on their persons. Thus, AA.15A could be the burial of an infant of a poorer woman, consistent with the general trend of area AA as a cemetery of somewhat poorer people. The fact that similar style earrings occur in the Amman Airport Cemetery, interpreted as a Bedouin/nomads' burial ground (Ibrahim and Gordon 1987), leads to the hypothesis that Area AA too may have been the cemetery for nomadic "dwellers-around" Umm el-Jimal, who lived in a symbiotic relationship with more affluent settled agriculturalists and trades' people of the community. Another provocative line of inquiry will be to explore the apparent relationship between poorer burials and more elaborate loculi/chamber tombs and what that might say about social organization and status in the community.

Future research will include statistical analysis of the cultural dimensions of the tombs such as different types of tomb construction, grave objects, orientation, and cemetery planning in conjunction with analysis of the age, sex, demography, stature, paleo-pathology and health status of burials from the two areas excavated this summer, as well as other burials excavated at Umm el-Jimal in the past. A comprehensive map of tombs and cemeteries associated with Umm el-Jimal is planned 1996. Finally, anatomical (facial and body) reconstruction is clearly possible with several individuals excavated during the 1994 season so that the people who once thrived an died at the site may live again through the evidence of their remains.

Back to Top

Bibliography

Back to Top

Publications Home Page