Content uploaded by Peter M.M.G. Akkermans
Author content
All content in this area was uploaded by Peter M.M.G. Akkermans on Nov 16, 2018
Content may be subject to copyright.
Deutsches Archäologisches Institut • Orient-Abteilung
BAND 9
2016
Sonderdruck aus
Zeitschrift für
Orient-Archäologie
© 2016 Deutsches Archäologisches Institut
Der Autor/die Autorin hat das Recht, für den eigenen wissenschaftlichen Gebrauch
unveränderte Kopien dieser PDF-Datei zu erstellen bzw. das unveränderte PDF-File
digital an Dritte weiterzuleiten. Außerdem ist der Autor/die Autorin berechtigt, nach
Ablauf von 24 Monaten und nachdem die PDF-Datei durch das Deutsche Archäolo-
gische Institut der Öentlichkeit kostenfrei zugänglich gemacht wurde, die unverän-
derte PDF-Datei an einen Or t seiner/ihrer Wahl im Internet bereitzustellen.
Deutsches Archäologisches Institut
Orient-Abteilung
Zeitschrift für
Orient-Archäologie
Band 9 • 2016
Ernst Wasmuth Verlag
Druck und buchbinderische Verarbeitung:
Deutsches Archäologisches Institut
Orient-Abteilung
⑧
Khirbat al-ʿUmari. View to area F (photo: DAI Orient Department / K. Bartl).
Khirbat al-ʿUmari
The Rediscovery of an Early Islamic Site South
of Azraq
Karin Bartl – Peter M. M. G. Akkermans
Introduction
The Early Islamic occu-
pation of the central and
northern desert steppes
in Jordan (Badia) and
the adjacent areas is rel-
atively well-known from
numerous archaeologi-
cal studies conducted
since the end of the 19th
century (Fig. 1).
Systematic aerial surveys carried out since 1998
under the APAAME project1 and about 10 years of
simplied satellite image analysis have furthermore
not only led to the discovery of countless prehistoric
sites,2 but also to that of additional sites from this late
period.3
A yet less known and almost forgotten site is
represented by the comparatively drawn-out ruins at
Khirbat al-ʿUmari, which initially were briey de-
scribed by Nelson Glueck in 1944 during his survey
of Wadi Sirhan.4 Recently, the site was sought up
again by the present authors.5
Khirbat ʿAmri6 or Khirbat al-ʿUmari7 is located
about 26 km in a beeline southeast to the Azraq oasis
(southern part of Azraq al-Shishan) and about 17 km
north of today’s Jordanian-Saudi Arabian border post
at al-ʿUmari.
Despite its relatively remote location, the site
nevertheless displays severe marks from looting and
vandalism. Its size, expansion, and layout all the
Abstract/Kurzfassung/
Der südöstlich von Azraq gelegene Fundplatz Khirbat al-ʿUmari ist eine
frühislamische Siedlung, die in den 1940er-Jahren erstmals von N. Glueck
erwähnt wurde. Die günstigen naturräumlichen Bedingungen der Umge-
bung, seine Lage sowie die Struktur der Anlage deuten auf eine mögliche
Nutzung als Karawanenhalt in umayyadischer Zeit.
ZOrA 9, 2016, 200–221
Jordan · Khirbat al-ʿUmari · Caravan Halt · Early Islamic · Umayyad · Report
Jordanien · Khirbat al-ʿUmari · Karawanenhalt · Frühislamisch · Umayyadisch · Bericht
1 Kennedy – Bewley 2004; <http://www.apaame.org>.
2 Müller-Neuhof 2006; Müller-Neuhof 2013; Müller-Neuhof
2014 a; Rollefson – Rowan – Wasse 2014.
3 Kennedy 2011; Kennedy 2014.
4 Glueck 1944, 14–15.
5 This paper is the outcome of observations made by the au-
thors during short ‘tourist’ visits to the site during their re-
search projects at Qasr Mushash and Jebel Qurma in 2015
and 2016.
6 Transliteration according to Glueck 1944, 14–15.
7 Transliteration according to map sheet Jordan 1:50.000,
3353.I, series 737, Wadi el-Ghadaf, prepared for the Minis-
try of Economy and the United States Agency for Develop-
ment to Jordan, 1961. This way of spelling was adopted in a
simplied form for the present purposes.
The site Khirbat al-ʿUmari to the southeast of Azraq is an Early Islamic
settlement which was rst mentioned by N. Glueck in the 1940s. Its
favourable natural environment, its geographic location, as well as the
site’s structure indicate a possible function as a caravan stop during the
Umayyad Period.
202
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
cultivation, which by now also has a tangible effect
regarding the preservation of archaeological sites.
Exploration
Archaeological exploration of the Eastern Badia began
towards the end of the 19th century,11 and it initially fo-
cused of the investigation of settlements and buildings
of the Roman and Early Islamic Periods. The study of
the prehistoric settlement begins in the rst half of the
20th century. To name but one project, was the survey
conducted on a large-scale by H. Field in the years be-
tween 1925 and 1960.12 However, it was not before
the 1970s and 1980s that similar survey and sounding
work resumed on an equivalent scale. Among others, it
included that carried out in the Azraq Basin as well as
that at Jawa and its surroundings.13
An important impulse to archaeological research
occurred around a decade ago with the use of satellite
imagery, thus offering new prospects for the discov-
ery of yet unknown settlement sites.
same suggest an important settlement in the Early
Islamic (Umayyad) Period. As in many other cases of
this period, a Late Roman / Early Byzantine precursor
occupation cannot be ruled out.
Region
The eastern desert steppe located within Jordan’s so-
called ‘panhandle’8 counts among the country’s most
outstanding landscapes. In addition to the basalt de-
sert with its extinct volcanic cones extending from
Southern Syria to Northern Saudi Arabia, limestone
formations, sand deserts, and mudats otherwise
known as sabkhas form other geomorphological fea-
tures of the desert that make up approximately 80 %
of the country’s entire surface.
Annual precipitation varies between 250/200 mm
in the north and less than 50 mm in the southeast,
therefore sanctioning a worthwhile rainfed agricul-
ture only along the northern fringes.9
Today, over wide areas the landscape’s vegetation
cover is characterised by a sparse scatter of scrubs
and herbaceous plants with occasional occurrences of
denser vegetation, which consists of terebinth, tama-
risk, and broom. Some areas, however, are entirely
bare. The present landscape therefore hardly attracts
permanent settlement, both for reasons of insufcient
precipitation as well as for ones of poor soil qual-
ity.10 Despite these shortcomings, the desert regions
are exposed to an ever increasing development, in
particular in terms of road construction and wadi
8 As referring to the country‘s drawn-out, eastern extension.
9 Ababsa 2013; Amr – Modry – Shuedat 2011.
10 Today, even pastoral nomads herding small livestock (sheep
and goat) are dependent on supportive measures through the
supply of water and fodder from the outside.
11 Brünnow – von Domaszewski 1905; Musil 1907; Moritz
1908; Gregory-Kennedy 1985.
12 Field 1960.
13 Helms 1981; Rollefson 1983; Betts 1998; Betts 2013;
Richter 2009; Richter et al. 2012; Garrard 2013; Müller-
Neuhof 2014a; Müller-Neuhof 2014 b; Richter et al. 2014;
Maher et al. 2014.
Fig. 1 Sites mentioned in
the text (map: DAI Orient
Department / Th. Urban using
USGS / NASA 3-arc second
SRTM data).
203
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
The site of Khirbat al-ʿUmari
Among the regions which were neglected by older
research and are not being included to newer study
programs, is the region stretching between the Azraq
Oasis and the border to Saudi Arabia. Its north-south
extension covers a distance of about 50 km, while its
width is about 25 km from east to west. It reaches out
on both sides of the Azraq–al-ʿUmari highway, be-
tween the western limestone region and the eastern
basalt area. This shallow depression is considered
as the northern end of Wadi Sirhan, and it forms the
main route between Bilad al-Sham and the region of
Jawf in the western part of Saudi Arabia as well as
al-Jawf and Central Iraq.
To the east is the area around Jebel Qurma that
for long has been known for its numerous rock carv-
ings and for several years has been under intensive
study in surveys and soundings.14
In the area between Azraq and the border, the Qaʿa
al-ʿUmari, located about 18 km southeast of Azraq al-
Shishan (Azraq South) and 21 km southeast of Azraq
al-Duruz (Azraq North), forms a striking landscape
marker. It is a mudat formed by the alluvial fan from
Wadi Jashsha al-Adla and Dashat al-ʿAmari/ʿUmari
and covers an area of approximately 6 km length and
1.7 km width. It belongs to the numerous sabkhas of
the eastern desert, whose virtually impermeable soil
gives rise to the formation of seasonal lakes for sev-
eral months in winter. Next to the groundwater, such
‘desert lakes’ belong to the most signicant water
resources of the eastern desert steppes. They still are
occasionally exploited by the nomadic population, at
least for purposes of animal watering.15 Due to the
durable hydrology, the vegetation cover in this area is
considerably denser, especially with regard to plants
with a higher tolerance towards increased soil salinity
after the drying-up of the Qaʿa in summer (Fig. 2).
Various small wadi courses ow from the south
into Qaʿa al-ʿUmari. Among them is (Wadi) Shuʿeib-
al-Jashsha, which today is partly blocked by various
barrages and embankments along its middle course,
mainly for agricultural reasons.
Khirbat al-ʿUmari is located about 3 km east of
(Wadi) Shuʿ eib-al Jashsha along a smaller (unnamed)
wadi, about 4 km south of Qaʿa al-ʿUmari (Fig. 3).
The settlement was visited by Nelson Glueck in April
1944 during his large-scale survey work concerning
the distribution of Nabataean sites.
He describes the area as follows:
“ …About 25 kilometers south-southeast of Az-
raq, one comes to ʿAmrī in the Wâdī Sirḥân.16 It is
already well inside the Wadī Sirḥân proper, and
14 Akkermans – Huigens – Brüning 2014; Huigens 2015;
Akkermans – Huigens in press.
15 Considerable precipitation in winter 2016 led to the com-
plete ooding of the Qaʿa al-ʿUmari, thus covering a surface
of c. 10.44 km2 (5,8 × 1,8 km) and exemplifying the signi-
cance of the area as a temporary water reservoir (see Image
© 2016 CNES / Astrium; © 2016 ORION-ME from March
4th, 2016).
16 Transcription of the Arabic topographical names and terms
according to Glueck 1944, 14–15.
Fig. 2 View to Qaʿa al-
ʿUmari (photo: DAI Orient-
Abteilung / K. Bartl).
204
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 3 Location of Khirbat al-ʿUmari (map: DAI Orient
Department / Th. Urban using map collection Jordan
1:50,000; 3353.I, series 737, Wadi el-Ghadaf, prepared for
the Ministry of Economy and the United States Agency for
Development to Jordan, 1961).
205
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
them. And again, among all the thousands of sherds
which I examined there, not one could be claimed as
Nabataean. …”
(Glueck 1944, 14–15)
Glueck arrived at the ruins probably coming from the
northwest as is suggested by his description of the
palm trees. Today, the palm trees concentrate in only
one area near Qaʿa al-ʿUmari, and to judge by their
size, they seem relatively old.
It is therefore possible that the place Glueck refers
to corresponds with the now abandoned police fort
which was built only after his visit (Fig. 4). Although
possible, it yet remains to be answered whether or not
this location coincides with the “sand-covered hill”
where he recovered the Roman and Byzantine sherds.
A small elevation with a modern cemetery rises
about 950 m further southeast to the police fort.
However, it doesn’t nd any mention in Glueck’s re-
port. The archaeological site of Khirbat al-ʿUmari is
about 1.9 km away from the police station and 1.2 km
southwest from the cemetery.
Khirbat al-ʿUmari is an elongated site along the
western and eastern banks of a small wadi in a north-
south ow at the southern foothills of the Qaʿa. The
settlement stretches out over about 850 m in a north-
east south-southwest direction, while its east-west
extension measures no more than 250 m (Fig. 5). The
main settlement area is in the east and extends over
650 m along the wadi. A smaller complex of approxi-
mately 150 m length and 50 m width follows on the
western side of the wadi. Both in the east and in the
west the arrangement of the building units is natu-
rally structured into smaller complexes by east-west
fortunately is still within the connes of Transjor-
dan. It is about 85 kilometers due east of the Hejâz
railway. Next to Azraq, the water obtainable at ʿAmrī
forms the best source of water supply in the entire
Wâdī Sirhân.
The site of ʿAmrī consists of a sabkhah area,
about 5 sq. km. in extent. In numerous places the un-
derground water wells up to the surface, and else-
where it is possible to nd it by digging down less
than half a meter-it being sometimes necessary to
break through a limestone layer about half a meter
thick more or less. When the desert sands or debris
ll up one water-hole, or when the water becomes
too foul to drink, it is a simple task to dig out another
water-hole.
From ancient times on, caravans have made
ʿAmrī a place to rell water-skins before proceed-
ing on to Azraq or turning off westward towards
ʿAmmân or Mâdebā by the route that passes by Qaṣr
Kharâneh and Muwaqqer. When we arrived at ʿAmrī,
we found herds of camels grazing and drinking there.
A couple of scrubby palm trees can be seen near one
of the water-holes. Close to them is a sand-covered
hillock, which looks as though it might conceal the
remains of a small ruin.
On and around it were fairly numerous Roman
and Byzantine sherds. Indeed, scattered throughout
the entire area one nds Roman and Byzantine sherds,
although not in any quantity. Among the sherds were
several pieces of sigillata. My disappointment with
the sherds of ʿAmrī was increased by the fact that,
as I had feared, there was not among them a single
indubitably Nabataean sherd.
We looked in vain, furthermore, for the remains
of some sort of a qaṣr or khân, but aside from the
above-mentioned hillock, there seemed to be no an-
cient ruins whatsoever.
Our quest, however, was not to prove completely
in vain. About a kilometer south of this sabkhah area
of ʿAmrī, we came across the ruin of a very large
caravanserai, which stretched over an area almost a
kilometer long.
It was at this site, which for want of a better des-
ignation I am calling Khirbet ʿAmrī, that the cara-
vans halted, while their beasts were watered at the
wells of ʿAmrī proper and then grazed nearby. The
buildings of Khirbet ʿAmrī, preserved in a sadly ru-
ined state, seemed to consist of single rooms or series
of chambers, constructed in one story of soft sand-
stone, which crumbles practically at a touch.
There are long hillocks formed by such struc-
tures, whose ruins are almost completely covered
over with debris. Very large quantities of Roman and
particularly of Byzantine sherds were found around
Fig. 4 Deserted police station and palm trees to the south
of Qaʿa al-ʿUmari (photo: DAI Orient-Abteilung / K. Bartl).
206
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 5 Schematic map of Khirbat al-ʿUmari (map: DAI
Orient Department / Th. Urban using Image © 2016
CNES / Astrium; © 2016 ORION-ME).
207
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
owing wadis, and referred to in the following as
areas A–J.
To the east of the ruins is a modern trapezoidal
plot of 1.3 km length and 1.05 km width bounded
from the surrounding terrain by an earthen embank-
ment. Within the walled area there are several build-
ings and planted trees. According to the satellite
images the plot seems to have been established after
2004 only.
Almost all building complexes at Khirbat al-
ʿUmari reveal recent looting pits, apparently hand-
dug by aid of picks and shovels. Furthermore, some
structures have been partially or completely destroyed
by the employment of heavy earth-moving equip-
ment. Tire tracks are partly still well visible, hence
indicating that this had occurred in the recent past.
In reporting on the buildings’ even then “a sadly
ruined state” Glueck (see above) certainly only meant
to describe the absence of upright walls or recognisa-
ble structures. The recent damages on the other hand,
can no doubt be attributed to widespread vandalism
throughout many of Jordan’s archaeological sites as
a result of a late upsurge in “treasure hunting”.17 In
the present case destruction has reached a level which
seriously narrows down the possibilities of recogniz-
ing the settlement’s original structure.
In the following, an attempt is nevertheless made
to give an accurate as possible description of the
settlement by aid of the observations on the visible
ndings at the surface as well as an evaluation based
on available satellite images. However, conclusive
and more detailed assessments all the same remain
dependent on surveying and exploratory work in the
eld.
Structural units
For the preliminary record we decided to subdivide
the entire site into individual units bordered by the
smaller east-west oriented wadis according to the
above description.18 The following list is arranged in
a south to north order. The still undamaged structures
are usually shaped as long rectangular elevations
from which variously wall remains emerge.
Areas A–B (Figs. 6–8, Tab. 1)
Area A is located in the settlement’s southwest and
comprises ve recognisable building units respec-
tively referred to as ‘structure’. Area B is located
north of area A, and is separated from it by a small
east-west oriented wadi. Here, the built-up part con-
sists of two complexes.
Area C (Fig. 6)
Area C is located on the eastern bank of the wadi and
is parallel to area A. The elevations visible on the sat-
ellite image are without exception natural features.
Some looting pits are recognisable at the western
edge near the wadi.
Area D (Figs. 9–11, Tab. 2)
Area D is located north of area C and separated from
it by a small east-west oriented wadi. Its terrain is
raised and at, and the southern and western peri-
meters reveal at, broken limestone slabs. In all,
there are nine structures, of which all are severely
disrupted or otherwise fully devastated.
Areas E–F (Figs. 12–14, Tab. 3)
Area E follows directly to the north of area D and is
separated from it by a small, SE-NW oriented side
wadi. The area covers a surface of about 100 m SE-
NW and 65 m SW-NE and comprises four structures.
Area F is adjacent to the north of area E and
like E is delimited by two SE-NW orienting tribu-
tary wadis. It covers an area of approximately 150 m
E-W × 85 m N-S and comprises ve, originally per-
haps six structures.
Areas G–J (Figs. 15–17, Tab. 4)
Area G is located 135 m northeast of area F and
reveals only two structures, both respectively in an
isolated position. About 150 m further to the north of
area G is area H, which has only one structure. Area
J apparently forms the northern end of the settlement
area of Khirbat al-ʿUmari, just north of a small east-
west running tributary wadi where it borders onto
the main north-south oriented wadi in the west. The
settlement area is at the southern end of a limestone
plateau that breaks off abruptly 30 m further north to
form small cliff towards the wadi. The satellite im-
age from November 2013 reveals three structures, of
which two have entirely disappeared by now through
the use of heavy machinery.
17 The quest for the so-called treasure of the Ottoman Empire,
which had allegedly been left behind somewhere in the
Levant after the Turkish withdrawal in WW 1, is viewed by
many in rural Jordan as the motivation behind the increased
activities in illicit digging.
18 Some of the spatially adjacent areas are presented here toge-
ther, even if they are relatively far apart from each other.
208
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 6 Khirbat al-ʿUmari, areas A–C, schematic map
(DAI Orient Department / Th. Urban using Image © 2016
CNES / Astrium; © 2016 ORION-ME).
209
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Area Structure Type Orientation Size (ca.) Preservation
A 1 house comprising three rooms, of which two
are interconnected
SW-NE 12 × 15 m (?) few small looting
pits
A 2 square house (?) SSW-NNE 10 × 10 m surface intact
A 3 5 pits in a row containing stones (long rectan-
gular house with room suite?)
SSW-NNE 14 m long fully destroyed
A 4 long rectangular elevation (house?) E-W ori-
ented walls visible at the northern and southern
perimeters
SSW-NNE 23 m long surface intact
A 5 house comprising 1 or 2 rooms, access area
with preserved defensive wall, walls visible in
looting pit
SSW-NNE 7.50 m long major robbery pit
inside
A 6 visible traces from walls EW-SN 10 m long looting pits
B 1 building with several rooms NNE-SSW 15.5 × 13 m minor looting pits
B 2 large complex with several rooms covered by
soil, unclear whether one or several houses,
exterior walls visible in south and north
N-S 23 × 20 m surface mostly intact,
minor looting pits in
the south
Tab. 1 Areas A–B.
Fig. 7 Khirbat al-ʿUmari. Area A,
structure 5, from North (photo: DAI Orient-
Abteilung / K. Bartl).
Fig. 8 Khirbat al-ʿUmari. Area B,
structure 2, from Northeast (photo:
P. M. M. G. Akkermans).
210
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 9 Khirbat al-ʿUmari, area D, schematic map (DAI
Orient Department / Th. Urban using Image © 2016
CNES / Astrium; © 2016 ORION-ME).
211
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Area Structure Type Orientation Size (ca.) Preservation
D 1 square building, interior layout
unclear, exterior walls from disinte-
grated mudbricks
SW-NE 10.5 m side
length
northern part disrupted, minor
looting pits in the south
D 2 small square complex SW-NE 4 m side
length
interior fully disturbed
D 3 long rectangular building SW-NE 21 × 9 m minor looting pits inside and along
the western exterior wall
D 4 devastated terrain, no visible walls,
uncertain if at all presence of
former architecture
SW-NE 4 × 13 m totally attened
D 5 agglomeration of possibly natural
stone slabs (broken away from the
terrace) no structures
SW-NE minor looting pits
D 6 trapezoid elevation SW-NE 15 × 5–9 m minor looting pits in the centre,
major disruption at the southern end
D 7 large area with several small loot-
ing pits surrounded by stones (each
pit possibly representing a room)
SW-NE 25 × 30 m fully destroyed, original structure
unclear
D 8 long rectangular elevation SW-NE 19 × 7 m prior to 2013 the destructions con-
cerned the southern end only, today it
is fully obliterated
D 9 large area with several small loot-
ing pits surrounded by stones (each
pit possibly representing a room)
NW-SE 21 × 13 m fully destroyed, original structure
unclear
Tab. 2 Area D.
Fig. 10 Khirbat al-ʿUmari. Area D,
structure 1, from North (photo: DAI Orient-
Abteilung / K. Bartl).
Fig. 11 Khirbat al-ʿUmari. Area D,
structure 6, from North (photo: DAI Orient-
Abteilung / K. Bartl).
212
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 12 Khirbat al-ʿUmari, areas E–F, schematic map
(DAI Orient Department / Th. Urban using
Image © 2016 CNES / Astrium; © 2016 ORION-ME).
213
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Area Structure Type Orientation Size (ca.) Preservation
E 1 two or three rooms from upright standing stone
slabs inside and next to a small wadi oriented E-W,
possibly a hydraulic installation
SW-NE 12 x 6 m collapsed stone
slabs, looting pits
inside rooms
E 2 building comprising several rooms, visible walls
remains from stone slabs and mud
SW-NE 10 x 8 m looting pits inside
rooms
E 3 directly adjacent to structure 2, but uncertain
whether independent compound
SW-NE 15 x 9 m minor looting pits in
the north and south
E 4 long rectangular building, visible walls from stone
and mud lumps
SW-NE 24 x 15 m minor looting pit in
the south
E 5 long rectangular elevation, two long walls recog-
nisable at the surface
SW-NE 24 minor looting pit in
the NW, otherwise
intact
F 1 long rectangular platform with delimiting mud-
brick wall in the north and south, three singular
rooms/houses to the south, visible stone founda-
tions
NW-SE 35 x 13 m severely disrupted
in the east from earth
moving equipment
all rooms disturbed
F 2 trapezoid elevation SW-NE 30 x 16 m several looting pits
in the north
F 3 long rectangular building, wall remains visible in
the west and north
SW-NE 15 x 6 (?) m ve looting pits in
the south and west
F 4 square building (?), visible walls SW-NE 4.50 m side
length
interior destroyed
F 5 long rectangular elevation, walls visible at the
surface
SW-NE 22 x 12 m major looting pit in
the north
F 6 amorphous elevation, uncertain whether containing
architecture
SW-NE looting pit in the
north
Fig. 13 Khirbat al-ʿUmari. Area E,
structure 1, from West (photo: DAI Orient-
Abteilung / K. Bartl).
Fig. 14 Khirbat al-ʿUmari. Area F,
structure 1, from East (photo: DAI Orient-
Abteilung / K. Bartl).
Tab. 3 Areas E–F.
214
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Fig. 15 Khirbat al-ʿUmari, areas G–J, schematic map
(DAI Orient Department / Th. Urban using Image © 2016
CNES / Astrium; © 2016 ORION-ME).
215
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Area Structure Type Orientation Size (ca.) Preservation
G 1 long rectangular structure consisting probably
of two or three rooms of which the one to the
west is still well recognisable, visible walls
NW-SE 12 x 4.50 m looting pit in the
western room, the
eastern one has been
completely bulldozed
away
G 2 building comprising one or two rooms with
visible wall remains, poor state of preservation
SW-NE 5 m side
length
surface intact
H square structure, visible remains of stone walls SW-NE 4.5 m side
length
surface intact
J 1 square (?) structure though shape hardly
recognisable, covered by soil
SW-NE 5–6 m side
length (?)
surface intact
J 2 small square structure SW-NE 5 m side
length
in the meantime
completely destroyed
by bulldozer
J 3 small square structure SW-NE 5.50 m side
length
completely destroyed
by bulldozer
Fig. 16 Khirbat al-ʿUmari. Area H,
structure 1, from Northeast (photo:
P. M. M. G. Akkermans).
Fig. 17 Khirbat al-ʿUmari. Area J,
structure 2 (destroyed), from Northwest
(photo: P. M. M. G. Akkermans).
Tab. 4 Areas G–J.
216
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Summary
The archaeological site at Khirbat al-ʿUmari, which
was discovered by Nelson Glueck in the 1940s and
later largely forgotten, consists according to the vis-
ible structures at the surface of a ‘central area’ of
about 280 × 150 m (areas D, E and F) as well as three
geographically separate units. The central area is
characterised by a number of buildings of different
structure and size, of which most are aligned SW-NE
like the course of the wadi. Structure F1 forms an ex-
ception, both in terms of orientation and layout.
The buildings are preserved only in their foun-
dations or base by roughly hewn stones. The rising
walls seem to have consisted of clay lumps or mud-
brick, whose raw material was mixed with ne sand.
This material is now preserved as amorphous chunks
only which disintegrate when touched.
The settlement part (areas A–B) located south-
west of the ‘central area’ reveals other buildings of
varying orientations and sizes. The complexes in
areas G, H, and J north of the central area are small,
isolated structures that may be taken as single or
two-room houses or towers. However, further details
concerning the settlement structure are only obtain-
able through exhaustive investigations in the eld
involving soundings and excavations in remaining
undamaged sectors and possibly also geophysical
surveys.
Dating
The surface nds from Khirbat al-ʿUmari, which
primarily compose of ceramic and some glass sherds
indicate an occupation during Late Antiquity and the
Early Islamic Period. Noteworthy is the relatively
high density of surface pottery, especially in the cen-
tral area D, as compared to other Early Islamic sites
in the region.19
As perceived from the preliminary examination,
different types of cooking pot ware with ribbed sur-
faces display the highest frequency of occurrence in
all areas. This pottery type had generally been in use
between the Hellenistic-Roman Period and the Early
Islamic Period. A more accurate chronological clas-
sication would require an analysis of the shapes.
On the other hand, a light coloured ware with
red paint applied in waves, circles, and bands may
be considered diagnostic for the Early Islamic, i.e.
Umayyad Period. Such evidence known is for in-
stance from Pella20, Heshbon21, and Qasr Mushash22
(Figs. 18–19).
This material is a reliable indicator for an occu-
pation in the 7th and 8th centuries. For the time being,
an earlier occupation cannot entirely be excluded,
19 Glueck’s account gives enormous amounts, indeed thou-
sands of sherds lying at the surface (see above). But con-
sidering the thin scatter of the ceramic deposit observable
today, this hardly seems likely.
20 Smith – Day 1989, 113.
21 Walker 2012, 529.
22 Bloch in Bartl et al. 2013, 185–186.
Fig. 18 Khirbat al-ʿUmari, pottery
from area D (photo: DAI Orient-
Abteilung / K. Bartl).
217
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
ognisable structures at the surface reveal little as to
their former functions. The site is located in an iso-
lated position in a landscape largely devoid of set-
tlements. The closest ones are Azraq (South) to the
north at 28 km distance and Haditha to the south at
26 km, in present-day Saudi Arabia. Both sites benet
from favourable conditions with regard to water sup-
ply and hence were perfectly suited as halting places
along the caravan routes throughout Antiquity and
the Islamic Period.37
How water was accessed at Khirbat al-ʿUmari
is basically unknown. A temporary water supply
23 Apparently Glueck had found a couple of terra sigillata
sherds near the Qaʿa al-ʿUmari (see above), but not in
Khirbat al-ʿUmari.
24 Genequand 2008.
25 Urice 1987.
26 Gaube 1977; Carlier 1989.
27 Gaube 1979; Helms 1991.
28 Musil 1907.
29 Kennedy 1982; Ghraiyib – Ronza 2007; Arce 2008.
30 Brünnow – von Domaszewski 1905; Musil 1907.
31 King – Lenzen – Rollefson 1983; Bisheh 1989; Bartl et al.
2014; Bartl 2016.
32 Vibert-Guigue – Bisheh 2007.
33 Kennedy 1982; Gilbertson – Kennedy 1984; Abu-Azizeh –
Vibert-Guigue in Corbett et al. 2014, 632.
34 Helms, 1990.
35 Kennedy 2011; Kennedy 2014.
36 Kennedy 2014, 98–99. Among them also Qasr as-Swab in
Western Iraq and Jabal Says in Southern Syria.
37 Azraq is an oasis irrigated by the waters from aquifers and
springs (Nelson 1973, 37–51; UN-ESCWA – BGR 2013,
514), Haditha is within the catchment area of the Wadi
Sirhan Basin of the Tawil-Quaternary Aquifer Systems
(UN-ESCWA – BGR 2013, 514 g. 421). For the ancient wa-
ter management installations in Haditha see Rees 1929, 92.
although during our short visit, no red slip wares
from the Roman – Late Roman / Early Byzantine
Period were observed.23
Discussion
Next to the so-called desert castles, such as Qasr al-
Mshatta24, Qasr Kharana25, Qastal26, Qasr Burqu27,
or Qasr al-Tuba28, Jordan’s Eastern Badia also counts
a number of Late antique / Early Islamic settlement
sites that are characterised by a relatively large num-
ber of buildings belonging to different types and partly
also by water managing installations. These include
Qasr al-Hallabat29, al-Muwaqqar30, Qasr Mushash31,
Qusayr ʿAmra32, Qasr al-Azraq33, al-Risha34, as well
as only recently discovered or published sites such
as Hibabiya and a building agglomeration north of
Azraq35. The latter as well as some other structures,
mostly identied through aerial prospecting in the
APAAME project, have lately been dened as “no-
madic settlements”.36 A characteristic feature of this
type of settlement is the scattered arrangement of a
relatively small number of buildings often including
the so-called ‘qasr’ type, a square structure with rows
of rooms, arranged around a central courtyard.
With regard to size and building distribution,
Khirbat al-ʿUmari corresponds to this settlement
type. It nevertheless yet remains to be established
whether it also includes ‘qasr’ type buildings. The
settlement’s occupation period seems to have been
relatively short-lived counting a limited number of
buildings with different shapes and sizes, whose rec-
Fig. 19 Khirbat al-ʿUmari, pottery
from area F (photo: DAI Orient-
Abteilung / K. Bartl).
218
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
38 Dams, open reservoirs, subterranean cisterns, and wells
reaching groundwater levels were commonplace facilities
serving the water supply in numerous settlements in the
Badia between the Roman and the Early Islamic Periods,
as for example at Qasr al-Hallabat, al-Muwaqqar, Qasr
Mushash, Qusayr ʿAmra, and Azraq. It is however generally
difcult to determine, whether these installations actually
allowed for a perennial supply at these sites.
39 Glueck himself actually afrmed that in his days camel
herds were brought to Qaʿa al-ʿUmari for grazing and wa-
tering (see above). Huge herds of camels were also noticed
by the authors in spring/summer 2016.
40 Tschanz 2004.
ensured by the wadi, at least during the winter months,
is never theless conceivable. The protected location of
the settlement on the elevated banks on the western
and eastern sides of the wadi may be an indication for
high water levels in winter. It is also imaginable that
the inhabitants of al-ʿUmari exploited the water from
the mudat of the Qaʿa, which is only 4 km away and
also easy to reach.
Barrage systems and storage facilities would
however have been required for exploiting the wadi
waters on a longer term, but no such hints were dis-
cerned during our visits in the area.38
Nelson Glueck believed that Khirbat al-ʿUmari
was a former caravanserai. Yet, if the question as to
the settlement’s water supply still awaits a satisfac-
tory answer, its use as a halt seems evident on
grounds of its proximity to the Qaʿa al-ʿUmari.39
This is also supported by its distances to the two near-
est sites Azraq and Haditha, which can be considered
as stations on the caravan route to the Wadi Sirhan.
Even in taking into account the topographic hurdles,
the respective distances between Haditha and Khirbat
al-ʿUmari, and Khirbat al-ʿUmari and Azraq do not
exceed 35 km, which is equal to the average distance
a camel caravan travels a day40 (see Fig. 1).
Khirbat al-ʿUmari may thus be considered as a possi-
ble caravan stop along the route at the northern outlet
of Wadi Sirhan, which indisputably was in operation
at least in the Early Islamic Period. Whether Khirbat
al-ʿUmari, like other settlements of the Early Islamic
Period, founds on an older occupation, for example
from the Roman or Late Roman / Early Byzantine
Period, remains for the time being unanswered.
It is therefore hoped that this comparatively at-
tractive, but nevertheless highly vulnerable settle-
ment area will come under closer examination in the
foreseeable future.
Addresses
PD Dr. Karin Bartl
German Archaeological Institute
Orient Department
c/o German Protestant Institute of
Archaeology
Shari’a Al-Habbab Bin Al-Munthir Nr. 32
P.O. Box 183
11118 Amman
Hashemite Kingdom of Jordan
karin.bartl@dainst.de
Prof. Dr. Peter M. M. G. Akkermans
Leiden University
Faculty of Archaeology
Van Steenis Building
Einsteinweg 2
2333 CC Leiden
The Netherlands
p.m.m.g.akkermans@arch.leidenuniv.nl
219
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Betts, A. V. G.
1998 The Harra and the Hamad. Excavations and
Explorations in Eastern Jordan, I (Shef-
eld).
2013 The Later Prehistory of the Badia. Excava-
tions and Survey in Eastern Jordan, II (Ox-
ford).
Bisheh, G.
1989 Qasr Mshash and Qasr Ain al-Sil: Two
Umayyad sites in Jordan, in: M. A. al-
Bakhit – R. Schick (eds.), The Fourth Inter-
national Conference on the History of Bilād
al-Shām during the Umayyad Period: Pro-
ceedings of the Third Symposium, English
Section, vol. II (Amman) 81–103.
Brünnow, R. E. – von Domaszewski, A.
1905 Die Provincia Arabia, Zweiter Band. Der
Äussere Limes und die Römerstrassen von
El-Ma‘an bis Bosra (Strassburg).
Carlier, P.
1989 Qastal al-Balqā: An Umayyad Site in Jor-
dan, in: M. A. Bakhit – R. Schick (eds.),
The Fourth International Conference on
the History of Bilād al-Shām during the
Umayyad Period: Proceedings of the Third
Symposium, English Section, vol. II (Am-
man)104–139.
Corbett, G. J. – Keller, D. R. – Porter, B. A – Tuttle,
Ch. A. (eds.)
2014 Archaeology in Jordan, 2012 and 2013
Seasons, American Journal of Archaeology
118/4, 627–681.
Field, H.
1960 North Arabian Desert Archaeological Sur-
vey, 1925–1950. Papers of the Peabody
Museum of Archaeology and Ethnology
45/2 (Cambridge).
Garrard, A. N. – Byrd, B. F.
2013 Beyond the Fertile Crescent. Late Palaeo-
lithic and Neolithic Communities of the
Jordanian Steppe. The Azraq Basin Project,
I (Oxford).
Ababsa, M.
2013 Atlas of Jordan. History, Territories and So-
ciety (Beyrouth).
Akkermans, P. M. M. G. – Huigens, H. O.
in press Long-term Settlement Trends in Jordan’s
North-Eastern Badia. The Jebel Qurma
Archaeological Landscape Project, Annual
of the Department of Antiquities of Jordan.
Akkermans, P. M. M. G. – Huigens, H. O. – Brüning,
M. L.
2014 A Landscape of Preservation. Late Prehis-
toric Settlement and Sequence in the Je-
bel Qurma Region, North-Eastern Jordan,
Levant 46/2, 186–205.
Amr, Z. S. – Modry, D. – Shuedat, M. F.
2011 Badia. The Living Desert (Amman).
Arce, I.
2008 Hallabat: Castellum, Coenobium, Praeto-
rium, Qasr. The Construction of a Palatine
Architecture under the Umayyads (I), in:
K. Bartl – A. Moaz (eds.), Residences,
Castles, Settlements. Transformation Pro-
cesses between Late Antiquity and Early
Islam in Bilad al-Sham, Orient-Archäologie
24 (Rahden) 149–178.
Bartl, K.
2016 Water Management in Desert Regions. Qasr
Mushash as an Example in Early Islamic
Times, in: S. McPhillips – P. D. Words-
worth (eds.), Landscapes of the Islamic
World. Archaeology, History, and Ethno-
graphy (Philadelphia) 50–68.
Bartl, K. – Bisheh, G. – Bloch, F. – Bührig, C. –
Saleh, H. – Urban. Th.
2014 Qasr Mushash: „Wüstenschloss“ oder Ka-
rawanenhalt? Zeitschrift für Orient-Archäo-
logie 7, 222–245.
Bartl, K. – Bisheh, G. – Bloch, F. – Richter, T.
2013 Qasr Mushash Survey: First Results of Ar-
chaeological Fieldwork in 2011 and 2012,
Annual of the Department of Antiquities of
Jordan 57, 179–193.
Bibliography
220
ZOrA 9, 2016, 200–221
Karin Bartl – Peter M. M. G. Akkermans
Gaube, H.
1977 ‘Amman, Harane und Qastal. Vier frühisla-
mische Bauwerke in Mitteljordanien, Zeit-
schrift des Deutschen Palästina-Vereins 93,
52–86.
1979 Die syrischen Wüstenschlösser. Einige
wirtschaftliche und politische Gesichts-
punkte zu ihrer Entstehung, Zeitschrift des
Deutschen Palästina-Vereins 95, 180–209.
Genequand, D.
2008 Trois sites omeyyades de Jordanie centrale:
Umm al-Walid, Khan al-Zabib et Qasr al-
Mshatta (travaux de la Fondation Max van
Berchem 1988–2000), in: K. Bartl – A. al-
Razzaq Moaz (eds.), Residences, Castles,
Settlements. Transformation Processes
from Late Antiquity to Early Islam in Bilad
al-Sham, Orient Archäologie 24 (Rahden)
125–151.
Ghrayib, R. – Ronza, M. E.
2007 Archaeological Evidence of Water Dis-
tribution and Quarrying Activity at Qasr
al-Hallabat, Studies on the History and Ar-
chaeology of Jordan 9, 423–429.
Gilbertson, D. D. – Kennedy, D. L.
1984 An Archaeological Reconnaissance of Wa-
terharvesting Structures and Wadi Walls in
the Jordanian Desert North of Azraq Oasis,
Annual of the Department of Antiquities of
Jordan 28, 151–162.
Glueck, N.
1944 Wâdī Sirḥân in North Arabia, Bulletin of
the American Schools of Oriental Research,
No. 96, 7–17.
Gregory, S. – Kennedy, D.
1985 Sir Aurel Stein’s Limes Report, British Ar-
chaeological Reports International Series
272 (i) (Oxford).
Helms, S. W.
1981 Jawa. Lost City in the Black Desert (Lon-
don).
1990 Early Islamic Architecture of the Desert.
A Beduin Station in Eastern Jordan (Edin-
burgh).
1991 A New Architectural Survey of Qasr Burqu,
Eastern Jordan, Antiquaries Journal 71,
191–215.
Huigens, H. O.
2015 Preliminary Report on a Survey in the
Hazimah Plains. A Hamad Landscape in
North-Eastern Jordan. Palestine Explora-
tion Quarterly 174, 180–194.
Kennedy, D. L.
1982 Archaeological Explorations on the Roman
Frontier in North East Jordan. The Roman
and Byzantine Military Installations and
Road Network on the Ground and from the
Air, British Archaeological Reports Interna-
tional Series 132 (Oxford).
2011 Recovering the Past from Above. Hibabiya
– an Early Islamic Village in the Jordanian
Desert?, Arabian Archaeology and Epigra-
phy 22, 253–260.
2014 ‘Nomad Villages’ in North-Eastern Jordan:
from Roman Arabia to Umayyad Urdunn,
Arabian Archaeology and Epigraphy 25,
96–109.
Kennedy, D. L. – Bewley, R.
2004 Ancient Jordan from the Air. Council for
British Research in the Levant (London).
King, G. – Lenzen, C. – Rollefson, G. O.
1983 Survey of Byzantine and Islamic Sites in
Jordan, Second Season Report, 1981, An-
nual of the Department of Antiquities of
Jordan 27, 387–436.
Maher, L. – Richter, T. – Stock, J. – Jones, M.
2014 Preliminary Results from Recent Excava-
tions at the Epipalaeolithic Site of Kharaneh
IV, in: M. Jamhawi (ed.), Jordan’s Prehis-
tory: Past and Future Research (Amman)
81–92.
Moritz, B.
1908 Ausüge in der Arabia Petraea, Univer-
sité Saint-Joseph, Mélanges de la Faculté
Orientale III, fasc. I, 387–436.
Müller-Neuhof, B.
2006 Tabular Scraper Quarry Sites in the Wadi
ar-Ruwayshid Region (N-E Jordan), Annual
of the Department of Antiquities of Jordan
50, 373–383.
2013 Nomadische Ressourcennutzung in den ari-
den Regionen Jordaniens und der südlichen
Levante im 5. bis frühen 3. Jahrtausend
v. Chr., Zeitschrift für Orient-Archäologie
6, 64–80.
221
Khirbat al-ʿUmari. The Rediscovery of an Early Islamic Site South of Azraq
ZOrA 9, 2016, 200–221
Rollefson, G. O. – Rowan, Y. – Wasse, A.
2014 The Late Neolithic Colonization of the East-
ern Badia of Jordan, Levant 46/2, 285–301.
Smith, R. H. – Day, L. P.
1989 Pella of the Decapolis II. Final Report
on the College of Wooster Excavation in
Area IX (Ohio).
Tschanz, D. W.
2004 Journey of Faith, Roads of Civiliza-
tion, Aramco World 55/1, 2–11; <http://
archive.aramcoworld.com/issue/200401/
journeys.of.faith.roads.of.civilization.htm>
(21.7.2016).
UN-ESCWA and BGR (United Nations Economic and
Social Commission for Western Asia; Bundesanstalt
für Geowissenschaften und Rohstoffe).
2013 Inventory of Shared Water Resources in
Western Asia (Beirut).
Urice, S. K.
1987 Qasr Kharana in the Transjordan (Durham).
Vibert-Guigue, C. – Bisheh, G.
2007 Les peintures de Qusayr ‘Amra. Un bain
omeyyade dans le bâdiya jordanienne
(Beyrouth).
Walker, B.
2012 The Islamic Period, in: J. A. Sauer – L. G.
Herr (eds.), Ceramic Finds: Typological
and Technological Studies of the Pottery
Remains from Tell Hesban and Vicinity,
Hesban 11 (Berrien Springs) 507–593.
Müller-Neuhof, B.
2014 a Introduction. Recent Research on the Late
Prehistory of the Arid Regions in Jordan,
Levant 46, 151–160.
2014 b A ‘Marginal’ Region with Many Options:
The Diversity of Chalcolithic/Early Bronze
Socio-economic Activities in the Hinterland
of Jawa, Levant 46, 230–248.
Musil, A.
1907 Kusejr ‘Amra, I. Text, Kaiserliche Akade-
mie der Wissenschaften (Wien).
Nelson, B.
1973 Azraq: Desert Oasis (Ohio).
Rees, L. W. B.
1929 The Transjordan Desert, Antiquity 3/12,
389–407.
Richter, T.
2009 Marginal Landscapes? The Azraq Oasis
and the Cultural Landscapes of the Final
Pleistocene Southern Levant (PhD thesis,
University College London).
Richter, T. – Bode, L. – House, M. – Iversen, R. –
Otaegui, A. A. – Saehle, I. – Thaarup, G. – Tvede, M.
– Yeomans, L. M.
2012 Excavations at the Late Epipalaeolithic Site
of Shubayqa 1. Preliminary Report on the
First Season, Neo-Lithics 2/2012, 3–14.
Richter, T. – Otaegui, A. A. – House, M. – Rafaiah, A.
M. – Yeomans, L. M.
2014 Preliminary Report on the Second Season
of Excavation at Shubayqa 1. Neo-Lithics
1/2014, 3–10.
Rollefson, G. O.
1983 Two Seasons of Excavation at Ain el-Assad,
Eastern Jordan, 1980–1981, Bulletin of the
American Schools of Oriental Research
252, 25–34.
Inhaltsverzeichnis
Mesopotamien und regional übergreifende Themen
PETER PFÄLZNER – PAOLA SCONZO with contributions by RALF BEUTELSCHIESS,
ALEXANDER J. EDMONDS, BENJAMIN GLISSMANN, SIMON HERDT,
JASON T. HERRMANN, SAMAN HEYDARI-GURAN, JOHANNES KÖHLER,
MARTINA MÜLLER-WIENER, IVANA PULJIZ AND MELISSA SHARP, The Eastern
Ḫabur Archaeological Survey in Iraqi Kurdistan. A Preliminary Report on the 2014 Season ...... 10
EMMANUELE PETITI – ARNULF HAUSLEITER – MARGARETE VAN ESS
in collaboration with DAVID CARAMELLI, Bioarchaeology and Neo-Assyrian Burial
Customs: Case Study on a Tomb Excavated in the City of Arbil ............................. 70
Levante
FRANCES PINNOCK, Royal Images and Kingship Rituals in Early Syrian Ebla:
A Multi-Faceted Strategy of Territorial Control in EB IVA North Inner Syria ................... 98
MARIUSZ GWIAZDA, A Hybrid Style Terracotta Protoma from Porphyreon
(Central Phoenicia) ................................................................ 118
SIGNE KRAG – RUBINA RAJA, Representations of Women and Children in Palmyrene
Funerary Loculus Reliefs, Loculus Stelae and Wall Paintings ............................... 134
EMANUELE E. INTAGLIATA, The Post-Roman Occupation of the Northern Courtyard
of the Sanctuary of Baalshamin in Palmyra. A Reassessment of the Evidence Based on the
Documents in the Fonds d’Archives Paul Collart, Université de Lausanne..................... 180
KARIN BARTL – PETER M. M. G. AKKERMANS, Khirbat al-ʿUmari. The Rediscovery
of an Early Islamic Site South of Azraq ................................................ 200
Arabische Halbinsel und der Region verwandte Themen
ARNULF HAUSLEITER – HANSPETER SCHAUDIG, Rock Relief and Cuneiform
Inscription of King Nabonidus at al-Ḥāʾiṭ (Province of Ḥāʾil, Saudi Arabia), Ancient Padakku .... 224
BARBARA FINSTER mit einem Beitrag von AHMED WAHBY, Qubbat ʿArrāf in der
Provinz Wuṣāb al-ʿĀlī/Jemen ........................................................ 242
HINWEISE FÜR AUTOREN ....................................................... 265
GUIDELINES FOR AUTHORS ..................................................... 267
ZOrA 9, 2016, 5–8
ZOrA 9, 2016, 5–8
8
ZOrA 9, 2016, 5–8