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 29, 2019
Content may be subject to copyright.
Living on the Edge or Forced into the Margins? Hunter-Herders in Jordan's
Northeastern Badlands in the Hellenistic and Roman Periods
Author(s): Peter M. M. G. Akkermans
Source:
Journal of Eastern Mediterranean Archaeology & Heritage Studies
, Vol. 7, No. 4
(2019), pp. 412-431
Published by: Penn State University Press
Stable URL: https://www.jstor.org/stable/10.5325/jeasmedarcherstu.7.4.0412
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/10.5325/jeasmedarcherstu.7.4.0412?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Penn State University Press
is collaborating with JSTOR to digitize, preserve and extend access
to
Journal of Eastern Mediterranean Archaeology & Heritage Studies
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
, . , . ,
Copyright © e Pennsylvania State University, University Park, PA
: ./jeasmedarcherstu...
LIVING ON THE EDGE OR
FORCED INTO THE MARGINS?
Hunter-Herders in Jordan’s Northeastern Badlands
in the Hellenistic and Roman Periods
Peter M. M. G.
Akkermans
e arid and desolate, basalt-strewn uplands of north-
eastern Jordan have been perceived as the natural home
of pastoralist communities, which lie on the very fringes
of the early urban polities of the eastern Mediterranean.
However, current eldwork in the area has revealed the
presence of many and diverse sites from the late prehis-
toric to the early historic periods that were nely tuned
to their harsh environment. Some of these sites include
rich assemblages, including rock art and inscriptions on
stone. is paper investigates the hunter-herder com-
munities that successfully exploited the “margins” in
the Hellenistic and Roman periods. It is argued that
these desert populations were forced into obscurity by
the Roman military intrusion in the region in the third
century CE.
: hunter-herders, basalt desert, harra,
Hellenistic, Roman, nomadic threat, enforced marginality
Just south of Damascus begin some , km² of black
basalt wasteland or harra, which stretch across north-
eastern Jordan into Saudi Arabia up to the sands of
the great Nafud. e broken basalt cover so character-
istic of the lava range has resulted in a rough and rocky,
dissected terrain of forbidding appearance (Fig. ). e
extensive boulder elds are cut by numerous erosion
gullies and wadis. Some of these form small depressions
that can hold moisture, while others debouche out of
the basalt into stretches of at and silted-up mud ats.
e wadis provide the only eective means of movement
in this barren area, which is otherwise very dicult to
access and travel through (Huigens a). East of the
harra are the gently rolling limestone hillocks and wide
gravel plains (or hamad). ese gradually decrease in
elevation from west to east and extend as far as the lower
Euphrates valley in Iraq. To the south, both the hamad
gravel and harra basalt are overlaid by stretches of wind-
blown Nafud sands.
Summers in the harra are dry and hot, with mean
temperature maxima of –C and outliers reaching as
high as C. Winters may be severe, with cold winds and
an average temperature of –C, with minima as low as
-C. Average annual precipitation ranges from up to
– mm in the northwest of the basalt expanse to
less than mm in the south. Most rainfall occurs inter-
mittently in the form of thunderstorms from November
until March, resulting in considerable surface runo into
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
FIG. 1
Harsh modern desert landscapes in the Jebel Qurma region in northeast Jordan. Basalt-covered table-mounds alternate
with extensive gravel plains, cut by erosion gullies and wadis. (Courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
FIG. 2
Map of Jordan showing the Jebel Qurma research area and the principal sites referred to in the text.
(Courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 415
the wadi systems and stretches of level mud ats with
very limited permeability. However, these mud ats
may hold sub-surface moisture to some degree even in
arid years. In times of sucient rainfall, they also may
hold shallow but extensive pools for several weeks or
months, attracting birds and other game. Permanent
natural water sources are absent in the harra, except
for the spring-fed seepages and marshes at Azraq and,
much further to the north and east, at Wisad, Burqu’, and
Nemara, all located at the very fringes of the basalt (but
see Lancaster and Lancaster for the many, diverse,
man-made waterholes).
Nowadays, the extensive basaltic wasteland is
exploited by small and dispersed Bedouin groups, who
build their tents around wadis and mud ats for only a
couple of weeks or months because grazing and water are
otherwise in short supply. Extensive parts of the region
are desolate, due in particular to modern overgrazing
by large herds of sheep and goats. e use of trucks to
transport water and, to a lesser extent, animal fodder has
allowed herds to remain longer in previously inaccessible
and ecologically fragile areas. e number of grazing live-
stock may easily disrupt the subtle reproductive cycle of
local ora and substantially exceed the sustainable car-
rying capacity of the area (Barham and Mensching ;
Roe ). However, the remarkable resilience of the
local vegetation should not be underestimated and sev-
eral species of plants reappear when there are sucient
rains at the right times (Lancaster and Lancaster :
–; see also Müller-Neuhof a: ).
Early Western visitors to Jordan’s northeastern badia
and its basalt-strewn uplands were intimidated by the
harsh living conditions and diculties of access, and
unsurprisingly were rather unattering about the area in
their travel accounts. T. E. Lawrence (of Arabia), who was
in the region in the icy winter of , noted: “e land-
scape was of a hopelessness and sadness deeper than all the
open deserts we had crossed . . . there was something sin-
ister, something actively evil in this snake-devoted Sirhan
waste” (Lawrence []: –). A similar account is
given by RAF pilot Roderic Hill in the s: “I once landed
in this country for the night on a grey winter evening . . .
at place was the epitome of loneliness. All around the
hills rose like odious at-topped slag-heaps, and lled me
with a sinister foreboding.” (Hill : ). Another British
colonial ocer, Captain Lionel Rees, points out: “this coun-
try looks like a dead re—nothing but cold ashes” (Rees
: ). Agnes Horseld traveled through the desert in
, stating: “On the way there we met neither man nor
beast for the country was undulating harra, a black waste
. . . with no sign of life.” (Horseld : ).
ese few examples emphasize notions of remote-
ness and solitude, and, by implication, marginality. Such
impressions tended to guide the common but unproven
assumption that these modern conditions were similar
in the past. However, this conclusion is dicult to rec-
oncile with the astonishingly rich archaeological and epi-
graphic record of the basalt range, which testies to the
abundant presence of indigenous peoples in the area in
antiquity. ere are many thousands of habitation sites
and stone-built installations of dierent types and sizes
across the basalt uplands at large, in addition to the innu-
merable pieces of rock art and North-Arabian inscrip-
tions on stone (see for example Kennedy ). While
conditions in the arid, rocky lava elds certainly do pres-
ent challenges largely exclusive to the region, the abun-
dance of nd spots indicates that there were also many
opportunities for small-scale communities. ese relied
primarily on hunting and herding, thus challenging any
modern ideas of marginality or cultural insignicance.
e remainder of this paper seeks to briey review the
issue of marginality vis-à-vis new data on Hellenistic and
Roman-era nomadic activity in the Jebel Qurma region
in Jordan’s northeastern harra (Fig. ).
Paleoclimate and Environment
e abundance of ancient sites in an area that appears
poorly suited for settlement raises an obvious question:
Were environmental conditions in the basalt area the
same as they are today? Ecological data from the desert
for the period under consideration (the Hellenistic to
Roman era) are still extremely scarce (see below), but
there is evidence that conditions were not always as
desolate as today. Sediments from the Azraq oasis on
the southwestern fringe of the basalt expanse suggest
that there were no signicant changes in the current
Saharo-Arabian vegetation during, at the least, the past
years. However, there were episodic variations in the
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
416 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
composition of the plant assemblages, indicative of both
dryer and wetter conditions (Woolfenden and Ababneh
). Fluctuations in the Dead Sea level also indicate
several major wet phases, interrupted by multiple arid
events, between roughly and BCE (Migowski
et al. ). Similar evidence for a series of short-term dry
and wet events in the fourth and early third millennium
BCE comes from analyses of speleothems from the Soreq
Cave near Jerusalem (Bar-Matthews and Ayalon ).
Several excavated sites in the basalt landscape pro-
vide clues for vegetation reconstruction. Charcoal from
replaces at the Neolithic site of Mesa- in Wadi Qatta
were dated to roughly – BCE; they contain evi-
dence of drought-resistant desert shrubs like Anabasis sp.
and Zygophyllum sp., but also g (Ficus carica), a tree with
relatively high water requirements (Rollefson et al. ).
Hearths at the site of Wisad, which is located some km
to the east of Mesa-, were dated between and
BCE. ese yielded charcoal evidence of deciduous oak
(Quercus ithaburensis) and tamarisk (Tamarix sp.), both
constituents of a forest-steppe vegetation with a required
minimum annual precipitation of mm. So, it appears
that rainfall in the area in the late seventh to early sixth
millennium BCE was signicantly higher than the hyper-
arid conditions (less than mm) of today (Rowan et al.
, ). is conclusion is supported by pollen evi-
dence from a nearby mud at, which, in addition to arbo-
real species like pine (Pinus sp.), oak (Quercus sp.), elm
(Ulmus sp.) and hophornbeam (Ostrya sp.), yielded wet-
land species like bulrush (Typhus lattifolia) and duckweed
(Lemna sp.). Together, this combination of plant species
implies marshy conditions in places (Rowan et al. ).
e site of QUR- lies in the Jebel Qurma region,
west of both Mesa- and Wisad. Recent analysis of char-
coal from replaces at the site yielded dates between
and BCE, originating from juniper (Juniperus sp.),
which is well adapted to rocky soils and arid climates,
and from the plane tree (Platanus orientalis). e latter
is found most often in riverine settings but is capable of
surviving in dry environments once it is established. A
replace at the nearby excavated site of QUR-, dated
to – BCE, provided proof for pistachio (Pistacia
atlantica), typical for a forest-steppe environment. e
same site also had several small, circular structures of
a much later date, which contained large amounts of
charred wood. ese related to local iron working and
were radiocarbon-dated to the third and early fourth
century CE. Wood analysis revealed the presence of pine
(Pinus sp.), with rather moderate water requirements, and
hawthorn (Crategus sp.), a component of forest-steppe
vegetation. Perhaps more stunning is the presence of
several hydrophilous tree taxa indicative of relatively wet
habitats, including plane (Platanus orientalis), g (Ficus
carica) and ash (Fraxinus sp.). ese species are sugges-
tive of a wetter and more tree-covered local environment,
very dierent from the arid conditions of today.
Relevant information also comes from the Bronze
Age settlements at Khirbet al-Umbashi and Jawa,
both located on the western boundary of the basalt
belt (close to Jebel Arab), in areas where the present-
day annual precipitation rate does not exceed –
mm. However, these sites contained evidence of a
forest-steppe vegetation that depends upon a minimum
annual rainfall in the order of mm per year. Such
forest-steppe is characterized by taxa like deciduous oak
(Quercus sp.), pistachio (Pistachio atlantica), tamarisk
(Tamarix sp.) and Amygdalus. e forest-steppe species
decline in frequency in stratigraphic layers at Umbashi
dated to – BCE. Instead, there is an increase
in more drought-resistant chenopods and olive (Olea
europaea), with a minimum requirement of – mm
rainfall per annum (Willcox ). Signicantly, there
is good evidence at both Umbashi and Jawa for water
management, in the form of dams, canals and cisterns,
indicating that people at these places were actively alter-
ing local conditions to sustain their communities in
the long run. is modication also includes the culti-
vation of domestic crop plants like barley and emmer
wheat, and the herding of caprines and cattle (see Helms
; Willcox ; Köhler ; Braemer, Échallier, and
Taraqji ).
Admittedly, the current evidence remains fragmen-
tary and inconclusive. Yet, it does suggest that local cir-
cumstances in the harra may have been more favorable
(i.e., wetter, greener) during at least some periods in the
past than at present. In this respect, Y. Rowan and oth-
ers recently concluded: “We are quickly realizing that
rather than a marginal environment of little utility, this
landscape was once rich in animals, plants, and people”
(Rowan et al. : ).
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 417
Jebel Qurma in Jordan
e archaeology of the basalt desert is still inadequately
understood. Archaeological interest in the region began
already in the s, but its history of research is
episodic; periods of advances alternate with long inter-
vals of little or no new results (for research histories,
see Betts , ; Müller-Neuhof b). In order
to address the many deciencies still existing at nearly
every level of investigation, a series of research projects
has been initiated in recent years, one of which is the
Jebel Qurma Archaeological Landscape Project (under
the auspices of Leiden University, the Netherlands). e
research project seeks to investigate local settlement,
life ways, and the treatment of the dead from a multi-
period perspective, and how these practices related
to the area’s diverse landscape and environments. To
contend with these questions, the project carries out
archaeological surveys and excavations in the Jebel
Qurma region in northeastern Jordan, some kilome-
ters east of the small oasis town of Azraq, close to the
Saudi border (see Fig. ). e area comprises about
km² of basaltic high grounds together with stretches
of gravel and limestone. is diverse landscape lies
between two major drainage systems, Wadi Rajil in the
west and Wadi Qatta in the east.
Recent research has elaborated on the abundant
proof for the prehistoric occupation of the Jebel Qurma
region until the late third millennium BCE (Akkermans,
Huigens, and Brüning ; Akkermans and Brüning
; Huigens ; Akkermans and Huigens ).
Signicantly, there is still hardly any evidence for the
use of the area, and the basalt desert at large, in the sub-
sequent second and early rst millennia BCE. Notable
exceptions are the early second millennium habita-
tions at Jawa and Khirbet al-Umbashi, on the western
fringes of the basalt (Helms ; Braemer, Échallier,
and Taraqji ). Umbashi also provided evidence for
human activities in the Iron Age, in the form of charcoal
that was radiocarbon-dated to the late ninth to early
fth century BCE (apparently, no structures were identi-
ed in association with this charcoal; Braemer, Échallier,
and Taraqji : ). Iron Age settlements and burial
elds, provisionally dated to the seventh century BCE
on the basis of ceramic nds, are also reported from the
vicinity of Qaf and Ithra, located at the onset of the Wadi
Sirhan near the Saudi-Jordanian border (Adams et al.
: ). e research project in the Jebel Qurma area
used intensive survey strategies that aimed to identify
highly visible sites with relatively dense artefact scat-
ters, as well as small, ephemeral locales with a hand-
ful of nds and low visibility. Despite the use of these
complementary approaches, the conspicuous absence
of nd spots from the second to the early rst millen-
nium BCE in the harra is remarkable and remains unex-
plained. Detrimental climatic uctuations have been
cited as one possible cause (Akkermans, Huigens, and
Brüning : ; Müller-Neuhof : ). Although
cyclical shifts and hiatuses in the local settlement record
are to be expected (Akkermans, Huigens, and Brüning
), there is another, more mundane explanation for
the presumed absence of sites. e eldwork shows that
populations in the region at this time did hardly or not
at all use pottery (or any other durable mass artifact, for
that matter), making them extremely dicult to trace
in the current archaeological record. Pottery, it appears,
did not develop into a mass commodity in the harra until
the third century CE, together with the onset of Roman
dominance in the area (see below). It is crucial to realize
that the lack of pottery should not be taken as evidence
for the absence of people in the basaltic desert region.
ere is good proof, in the form of habitation sites and
burials, for the presence of hunter-herder groups in the
basalt area from, at the least, the second half of the rst
millennium BCE onwards. e remainder of this article
oers an evaluation of wholly new data on Hellenistic
to Roman-period occupation in the Jebel Qurma range,
specically in light of the concept of marginality. It must
however remain an interim review since the ongoing eld
research will undoubtedly improve the available data sets
and associated viewpoints.
Nomadic Remains in the Jebel Qurma Area in
the Hellenistic-Roman Periods
Several dozen dwelling sites dated to the Hellenistic and/
or Roman period have been identied in surveys and
excavations in the Jebel Qurma area (Fig. ). eir precise
chronology is still incompletely understood. Excavated
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
418 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
sites yielded Optically Stimulated Luminescence (OSL)
dates and radiocarbon dates, conrming their attribu-
tion to the Hellenistic and/or Roman period. Other sites
had surface nds of Roman or Roman/Byzantine pottery
(no Hellenistic ceramics were found), or were associated
with pre-Islamic (Safaitic) inscriptions on stone, refer-
ring to the construction or the ownership of enclosures
and other installations at these places. ese inscrip-
tions are conventionally dated between the rst century
BCE and the fourth century CE (see, e.g., Al-Jallad :
–), although brand-new data from Jebel Qurma
suggest that it is more likely that their beginning is in
the second, if not the third, century BCE. Until now,
residential sites have been ascribed to the Hellenistic-
Roman period, although this gure should undoubtedly
be expanded: many sites in the harra, suspected to t in
the period under consideration, cannot be dated prop-
erly, due to the lack of dating tools.
FIG. 3
Distribution of Hellenistic to Roman and Roman/Byzantine-period dwelling sites in the Jebel Qurma region. (Base map:
Landsat 7—United States Geological Survey; courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 419
e dwelling sites comprise stone-walled enclosures and
simple open clearings in the basalt for camping (Huigens
b). In addition, some places featured a single round
building about – m across, with a narrow doorway and
one or more pre-Islamic inscriptions on stone in front of
the entrance (the texts claim the construction or ownership
of these structures). e enclosures, also regularly accompa-
nied with inscriptions, occurred either as single structures
or in small groups, the walls of which often intertwined to
form large but irregular structures several dozen meters
across (Fig. ). On the other hand, the clearings were mostly
– m in diameter, and were devoid of architectural
installations. ey are located in relatively easily accessible,
open areas in low-lying valley oors, gravel plains, and near
mud ats. Such a site preference and the absence of struc-
tural features may reect limited labor investment and
temporary use only. In contrast, the sites with enclosures
imply a more intensive and durable commitment, although
it is not always clear what they were used for. It is often
assumed that the enclosures served as corrals for conn-
ing animals, with the stone walls protecting the livestock
against the cold winds in winter. Yet, the common occur-
rence of shallow replaces in them points to other activi-
ties in the domestic and residential sphere being present as
well. e latter may also be inferred from the (restricted)
pottery distributions in many enclosures. Interestingly,
the enclosures tend to be situated predominantly in rather
secluded areas with low visual prominence in the basalt
area, perhaps indicating security concerns. Another option
is that the enclosures were the preferential winter camps of
the nomadic communities, which provided shelter to both
humans and animals alike. Potential water sources usually
occur within a range of m from the sites, although most
of these sources may have carried water only seasonally.
FIG. 4
The site of QUR-595 in the Jebel Qurma area: an enclosure from the fourth century BCE under excavation. (Courtesy
of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
420 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
e restricted site sizes at both the enclosures and
the clearings suggest that the domestic installations
were used by small groups, in the order of perhaps sev-
eral dozen people at the most. Moreover, the sites were
short-lived and probably occupied on a seasonal basis
only, given the shallow sediments and their limited arti-
fact assemblages. Several enclosures provide evidence
of repeated though intermittent use over a prolonged
period of time. For example, the small site of QUR-
had several small hearths that were radiocarbon-dated
between the fourth to third century BCE (the Hellenistic
period), the late rst century BCE to early rst century
CE (the Roman period), and the fteenth century CE
(the Ottoman period), respectively. Another nearby site
(QUR-) yielded a sequence of replaces and, by impli-
cation, living oors, dated between the third and eighth
centuries CE (late Roman to Ummayad periods; Huigens
b). e sites served the recurrent residential needs
of what may have been highly mobile, nomadic, groups,
which exploited the desert landscape in a variety of ways
(see Huigens b for a detailed account on the nomadic
nature of the communities).
e ancient desert populations are usually consid-
ered to have been migratory pastoralists, who subsisted
predominantly on raising camels and, to a lesser extent,
sheep, goats, and horses. Information in this respect
is derived not from faunal assemblages (which are still
absent from both the surveys and excavations in the
area, probably due to matters of preservation), but from
inscriptions in the area (see, e.g., Graf ; Macdonald
, ). However, the local texts and, in particular,
petroglyphs from the late rst millennium BCE make
clear that hunting wild animals must have been of equal,
if not more, importance to these people. Hunting imag-
ery depicting humans with bows and riders on horseback
is very common (Fig. ), as are the ubiquitous wild ani-
mals shown on the basalt panels, including lion, hyena,
ostrich, gazelle, ibex, oryx, and onager (Brusgaard, forth-
coming). e enormous hunting potential of the harra
also may be inferred from the many hundreds of so-called
“desert kites,” which are large traps for the mass captur-
ing of game that date from dierent periods (Kennedy
; Crassard et al. ). Signicantly, kites commonly
are depicted in the local rock art, suggesting that people
made regular use of them in the period under consider-
ation (see Betts and Helms ; Macdonald ). It is
highly likely that the basalt uplands fully met the year-
round subsistence needs of the people indigenous to the
region. ey constantly moved through the area in small,
probably family-based, groups, in search of both hunting
and grazing grounds. Regular aggregations of people may
have taken place whenever the need arose. In this respect
it is worth mentioning that several so-called “pendant
burials” in the Jebel Qurma region (see below) were asso-
ciated with low, stone-built platforms up to . m in
diameter. ese may have served for rituals and/or gath-
ering. e small nomadic groups must have participated
in more extensive networks of approximately –
people, in order to remain reproductively and socially
viable (Hassan ; Dunbar and Sosis ).
In addition to their places for camping, the native
hunter-herders left numerous burial cairns of dierent
shapes and sizes in the harra of Jebel Qurma (Fig. ).
Little is known regarding the precise date of many cairns,
owing to factors such as their often generic morphol-
ogy and the palimpsest of contents resulting from their
continual reuse. Despite these lacunae, there is good
evidence to suggest that some cairns (out of several
hundreds of cairns in the Jebel Qurma area in total) date
between the fourth century BCE and the fourth century
CE (based on C dates, OSL dates, rock carvings, and
material-cultural items from the tombs; see Akkermans
and Brüning ; Huigens b). Although the total
number of Hellenistic to Roman-period tombs is consid-
erable, their distribution over a period of seven or eight
hundred years at the least is important to keep in mind.
e distribution of currently known tombs is in agree-
ment with a low population density in the area at any
given moment.
While the foci of daily living and domestic activity
were in low-lying areas at the foot of the basalt ranges
or in the deep wadi valleys, the preferential areas for the
disposal of the dead were on the surrounding high pla-
teaus and the summits of the basalt hills. On the basis
of our current excavations, three types of Hellenistic to
Roman-period burials can be distinguished: ring cairns,
tower tombs, and cist graves. It is often dicult to dif-
ferentiate between these types of burials during the
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
FIG. 5
Hunting scenes on basalt rocks. Above: an archer equipped with a composite bow and a quiver is
aiming at two lions. The associated inscription in Safaitic reads: “By Rgl son of Zmhr son of s¹ are
the animals.” Below: the rather roughly executed scene shows an archer hunting an equid.
The inscription in Safaitic reads: “By Gml {son of} {Gs¹l} is the she-ass.” (Translation by
C. Della Puppa; photos courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
422 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
eld survey due to factors such as the continuous reuse,
reconstruction, and modication of the graves over the
course of time. Earlier types of tombs may have been
completely buried (and hence are hidden from view)
underneath later forms of graves. For example, in several
cases tower tombs were set on top of ring cairns, and cist
graves were buried underneath much younger Islamic
tombs. Hence, any counts per type of tomb must remain
highly tentative.
While multiple burials are not uncommon in each struc-
ture, each type of cairn was primarily intended for a single
interment, with the dead resting on the side in a contracted
position. e cairns were also frequently reused, which
could often only be accomplished through ravaging older
burials in the mounds (Akkermans and Brüning ).
Ring cairns seem to constitute the majority of the
Hellenistic-Roman cairns in the Jebel Qurma area,
comprising at least tombs (but probably many
more). ey have an oval- or square-shaped burial
chamber with a corbelled roof in their center, which
is encircled by an outer ring of large stones. e area
between the central burial chamber and the outer ring
is entirely lled in with stones, giving these cairns a
conical appearance (Fig. ). ey measure up to m in
diameter and up to m in height. Signicantly, these
ring cairns often have a chain of to small, circular
or rectangular cairns attached to them in a straight or
curving line: it is easy to see why these installations are
sometimes called “pendants” (Fig. ) (see Kennedy :
–; Rees : ; Parr et al. : ). Excavation
of these smaller cairns at several sites revealed that they
did not contain or cover burials. Most likely, they were
commemorative structures rather than actual tombs
(Kennedy : ).
FIG. 6
A typical burial cairn in the Jebel Qurma area, dated to the late first millennium BCE. Inside the low mound of basalt blocks is a
small corbelled chamber where the deceased was placed to rest. (Courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 423
e second main type is the tower tomb, represented
by cairns so far. ese are round structures up to m
in diameter and . m high, and similar to the previous
type of tombs they also have a corbelled burial chamber
at their center. ey dier from the ring cairns with their
distinct tower-like shape and their clear, straight façade
made of large, attened basalt slabs (Fig. ). Each tower
tomb was solidly lled in with basalt boulders, and usu-
ally they do not have a pendant tail. In several cases,
tower tombs were set on top of (partially leveled) lower
ring cairns, indicating a relatively late date for these
installations. Although some tower tombs in the region
may date to the Bronze Age (on the basis of OSL dates
and/or pottery nds), our recent excavations demon-
strate that these graves were still a common feature in
the basalt wasteland in much later periods. Many tower
tombs present evidence for their construction, recon-
struction, and use from the third or (at the latest) second
FIG. 7
Aerial photo of a ring cairn, about 8.4 m in diameter, at the site of QUR-147 in the Jebel Qurma area. (A) outer ring of large stones;
(B) central burial chamber; (C) stone infill between the burial chamber and the outer ring; (D) remains of tower tomb, built on top
of the ring cairn at a later stage. (Courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
FIG. 8
A characteristic “pendant burial” at the site of QUR-32 in the Jebel Qurma range. Above: Aerial photo of the chain of 48
small, individual cairns, which leads to a large burial cairn on the left and on the right to an extensive platform. Below: Detail of
the “pendant tail.” The small cairns measure each about 1.9 x 0.75 m and stand to a height of about 60 cm. (Photo above by
D.Kennedy; courtesy of the Aerial Photographic Archive for Archaeology in the Middle East, APAAME_20081102_DLK-0146.
Photo below courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 425
century BCE until the third century CE (Akkermans and
Brüning ; Huigens b).
e third type of burial consists of rectangular
cist graves, up to . m long, . m wide, and m high.
Characteristic here are the carefully made dry-stone walls
with smooth outer facades. e cist graves were usually
attached to the walls of tower tombs, a practice that seems
to have started in the rst century BCE. ey remained in
use until the second to third century CE (according to C
dates, OSL dates, and grave inventories; Akkermans and
Brüning ; Huigens b). Cist graves have appeared
at only three, possibly four, sites until now, compris-
ing some seven or eight tombs altogether. ey occur in
other, nearby regions as well, always in association with
tower tombs, such as at Wisad Pools east of Jebel Qurma
(see Rollefson : , who interpreted the cists, incor-
rectly, as “entrance chambers” to the tower tombs).
Each and every type of tomb contained grave goods,
which were consistently in the form of personal jewelry
and belongings such as necklaces, rings, earrings, and
bracelets made of dierent materials (e.g., stone, faience,
glass, shell, mother-of-pearl, coral, bronze, iron, gold;
a detailed study is underway). Occasionally, there was
also iron weaponry (e.g., arrowheads, javelins) present
in the graves. Pottery was entirely absent, but several
tombs contained a single small bronze bowl. An extraor-
dinary, rare nd was the remains of vulture legs in two
tower tombs, which may have had an amuletic meaning.
Another rare nd were the bronze and silver Seleucid
coins found at two sites, which dated to the reigns of
Antiochus VII Euergetes (– BCE) and Antiochus IX
Cyzicenus (– BCE), respectively.
Signicantly, the various types of burials and habita-
tion sites do not display any explicit signs of elites, social
hierarchy, dierentiated wealth distribution, or other
forms of structured ranking. is observation is consis-
tent with the small-scale, family-based organization of
the desert groupings. An overarching heterarchical social
arrangement was in place, instead of explicitly hierarchi-
cal schemes.
FIG. 9
A tower tomb at the site of QUR-118 in the Jebel Qurma region, with a diameter of 4.2 m and a straight façade made of large
basalt slabs. (Courtesy of the Jebel Qurma Project Archive.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
426 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
Exchange, Local Development, and
Marginality
e desert communities themselves may have produced
some of the beads and other jewelry found in the burials,
by working local stone, ostrich eggshell, and land-snail
shells. However, objects made of bronze, iron, silver,
gold, faience, glass, coral, and seashell were external
products. ey must have arrived at the desert sites
through exchange with the urban environments beyond
the fringes of the basalt landscape where such materials
were typically manufactured. Seleucid silver and bronze
coins came from Syria or the northwestern Levant, as
probably did the metal and glass pieces of jewelry. Coral,
cowrie, and other seashells used to manufacture decora-
tive beads and pendants must have been brought from
the Red Sea, in the form of either raw materials or as
nal products. A faience Pataikos amulet undoubtedly
has an Egyptian origin (Fig. ). is exchange must have
existed for many centuries, although its scale and inten-
sity remains elusive for the moment.
While connections to the outside world are certain, the
archaeology of the desert communities displays an over-
whelmingly and deeply entrenched local character, which
is very dierent from the contemporary urban sites of
the western Levant and Syria. e various types of tombs
distinguished in the Jebel Qurma area are, to a very large
extent, typical of the barren basalt expanse of the Harrat
ash-Sham at large. Although it goes without saying that
cairns occur in masses elsewhere (such as in the Sinai, the
Negev, the Syrian desert, and the basalt area near Homs;
see, e.g., Haiman ; Meyer ; Philip and Bradbury
), it is important to realize that these cairns share sim-
ilarities with, but are not identical to, the ones from Jebel
Qurma and its surroundings. e same can be said of the
many enclosures and other ephemeral architecture built
for shelter and domestic activity (evidently, we must not
simply lump evidence from dierent contexts together).
Although a few inscriptions have been found in set-
tled areas, the main distribution of Safaitic petroglyphs
and texts on stone is spatially restricted to the harra
landscape (Macdonald ). e content of these texts
is highly introverted, and deals mainly with life in the
desert; few texts refer to the outside world (Macdonald
; Al-Jallad ). e communities that produced
FIG. 10
A small faience amulet of Egyptian origin, found in a ring cairn at the site
of QUR-147 in the Jebel Qurma area and dated between the seventh and
fourth century BCE. The apotropaic amulet, 1.2 cm in length, shows the
dwarf-god Pataikos. (Courtesy of the Jebel Qurma Project Archive.)
the burials and petroglyphs were not migrants from else-
where who only made use of the desert in the course of
seasonal treks. Rather, they were indigenous peoples who
lived year-round in dierent places in the wasteland (see
Macdonald : ; Sartre : ). ey subsisted
on hunting and pastoral activities, and part of the rev-
enue from these practices may have been used to acquire
exotic commodities. Interestingly, pottery was not among
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 427
the objects imported, despite the fact that it was used in
enormous abundance and in many dierent forms by the
contemporary sedentary communities less than km
to the west. It appears as though local development and
chronological sequence had little or nothing in common
with the highly complex material-culture sequences in
the western Levant and Syria at the time. In short, an
overall picture begins to emerge of the desert sites and
peoples that were highly autonomous and independent
in social and economic terms. ey were only to a (very)
limited extent connected to, and integrated in, the urban
polities and empires of their time.
Although it is tempting, it would be all too easy to
conclude that the desert communities were marginal
and peripheral. Such a binary and one-dimensional
core-periphery interpretation would put the urban societies
at the fore and reduce anything else to a lower level of rel-
evance. As a social construct, marginality is always relative,
and it is used to label deviations from what is considered to
represent the geographical, economic, or social “norm” (see,
e.g., Turner and Young ; Walsh, Richer, and Beaulieu
; Holly ; Crompton ). e highly autonomous
desert groupings lived in a constraining but still highly pro-
ductive environment. From their perspective, the notion
of marginality may have been of no meaning whatsoever:
marginal to whom or what? Why would they consider
themselves to be living on the margins, as they were able
to successfully sustain a long-lived desert culture that was
fully adapted to a dicult terrain and climate? Even if the
rough basalt uplands were ecologically and economically
peripheral in comparison to the fertile Levantine regions
and trading cities to the west, they were still culturally
central to the communities that continually used them for
many centuries. e evidence of this ourishing culture can
be seen in the rich archaeological record we still perceive
(and investigate) today. Marginality, however, became an
issue for the local desert groups in the third century CE,
with the appearance of the Romans in the area.
Forced into the Margins
During the reign of Septimius Severus, Rome’s interest
in Jordan’s northeastern desert was clearly and tangibly
expressed by the construction of a chain of fortresses and
barracks for garrisons in and around the Azraq oasis, stra-
tegically located at the outlet of the Wadi Sirhan, a major
natural trade and migration route between the Levant
and Arabia (Fig. ). Related to these fortresses were
several routes, comprising cleared tracks up to – m
wide and not surfaced roads. With these routes, the mili-
tary sites were connected to the urbanized sites to the
north, including Umm al-Jemal, Deir al-Kahf and Umm
al-Qutayn in Jordan, and Bostra and Damascus in Syria.
A number of building inscriptions and milestones from
the Azraq area date the military works to the very begin-
ning of the third century, – CE. Other inscriptions
belong to the late third and early fourth century, although
Azraq may have been abandoned soon afterwards, around
CE. Dates in the third century also come from the
fortress at Hallabat, some km to the northwest of
Azraq (Kennedy , , ). Several places in
the Jebel Qurma region, characterized by stone-walled
enclosures and simple clearings in the basalt, yielded
also Roman-style ceramics dated to the rst or second
century, although these should not be necessarily inter-
preted in a “pots-and-people” manner to directly indicate
a Roman presence. More likely, the pots were imports by
Nabataean groups in the area. In several instances, Roman
military installations appear to reoccupy locations that
were previously used as strongholds by the Nabataeans,
such as at Aseikhin and Azraq itself (Parker ).
It appears that the Roman impact on the local des-
ert populations was substantial, or in the words of
D. Kennedy (: –): “e Severan road slicing
through the Basalt Desert to the Azraq Oasis was a
massive intrusion into the world of nomad and trader.”
e opening-up of the desert in the third century entailed
profound and hitherto unseen changes to the very foun-
dation of desert lifeways. is included the disintegra-
tion of the local and centuries-old custom of interring
the dead in cairns, in favor of new types of graves such
as pit burials in small cemeteries, identied at three
places on the fringes of the basalt expanse until now.
No new burial cairns were constructed, although a few
tombs appear to have been reused in the fth to sixth
century CE. Changes also took place in the sorts of grave
goods accompanying the dead: pottery vessels occurred
for the rst time, in addition to the ubiquitous jewelry
(Akkermans and Brüning ; Huigens b).
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
428 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
e occurrence of ceramics in third- to fourth-
century burial contexts is not unexpected, considering
that Roman-style pottery also made its debut at many
domestic camp sites in the desert around that time. e
mass occurrence of ceramics was a major novelty in the
material culture of the basalt area, which until then had
rarely or not used pottery for at least some , years. I
suggest that the new pottery wares were not used by the
local, indigenous populations but belonged to newcom-
ers to the region: pastoralists who, in a sense, invaded
the desert and exploited its grazing grounds in the wake
of (and under protection of) the Roman army at Azraq
and elsewhere. e question of ceramic ownership is rel-
evant, as it advances dierent perspectives on the rela-
tionship between the desert groupings and Rome: one
emphasizes assimilation and integration, and the other
(favored here) indicates social stress and conict.
Another major change was the dissolution of the
local rock-art production and the Safaitic writing tradi-
tion, occurring probably in the third, but certainly in the
fourth century CE. Kennedy already linked their demise
to the Roman penetration of the steppe belt, the transi-
tional area stretching between the Ajlun highlands and
the Belqa in the west and the basalt wasteland in the east.
Military installations and farming villages abounded
throughout this steppe. He suggests that these villages in
the steppe were inhabited by sedentarized nomads, who,
probably voluntarily, abandoned their previous mobile
lifestyle in favor of the dominant Greco-Roman culture
(Kennedy : –). eir willingness to do so may
FIG. 11
The Roman fortress at Azraq. Although medieval and later rebuilding conceal the original layout, the lower courses
of large basalt blocks probably preserve the form of the Roman castellum. (Photo by I. Ruben; courtesy of the
Aerial Photographic Archive for Archaeology in the Middle East, APAAME_20080909_IAR-0210.)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 429
have derived from their life on the edge: “ese were
people living in the margins with almost nothing to lose
but a great deal, relatively, to gain through interaction
with the richer people to the west—merchants on desert
routes, farmers and townspeople” (Kennedy : ).
Here I take a dierent stance: the desert populations did
not have much to gain from the contact with Rome, but
rather they risked losing everything.
ere is a widespread opinion not only that there was a
signicant Roman military build-up on the Arabian fron-
tier in the late third and early fourth centuries, but also
that the fortication served primarily to confront the
“nomadic threat” of the time (see for example Graf ;
Parker , ; Kennedy ). e “Saracen” tribes
of the eastern deserts, it is claimed, had to be reined
in to prevent local raiding and pillaging or deep incur-
sions into the urbanized world. In this perspective, it was
crucial to control the nomads’ main natural migration
route through the Wadi Sirhan and its principal watering
places—hence the fortresses in the Azraq basin.
e notion of endemic conict presumably initiated
by local nomads along the eastern frontier has been
rightly criticized for decades due to its lack of solid evi-
dence and its oversimplied desert-versus-the-sown
dichotomy (Banning , ; Graf ; Macdonald
). It is also dicult to reconcile current archaeologi-
cal data from the Azraq oasis with those of the basalt
desert directly east of it. e meagre numbers of desert
populations, their wide distribution, and the limited
threat they may have represented, do not justify fully
the massive labor and resource investments in military
installations in the area. It is more likely that the military
works were part of a much wider initiative of coloniza-
tion of the Transjordan interior for settlement, as well
as for the control of its resources (see Kennedy ).
Azraq and its substantial water reserves had, among
other things, enormous potential for raising and grazing
livestock. Moreover, it may have been the high-quality
salt from Azraq and, further to the south, from Kaf and
‘Ithra that attracted the Romans to the area before all
else (Berghuijs and Akkermans, forthcoming). e neces-
sity for military buildings and infrastructure implies that
the Roman takeover of the Azraq region and beyond was
not appreciated by the local populations: collision and
resistance to the Roman imperialist policy may have been
rampant. It is not surprising that literary sources of the
fourth century, such as Ammianus Marcellinus, tend to
propagandistically portray the desert groups as plunder-
ers and marauders (see Graf : –). In fact, it may
have been the Roman occupation itself that triggered the
hostilities at the base of the “nomadic threat.” However,
the inherently uneven balance of power made clear that
the battle could not be won by the local groups: their
desert-oriented culture was about to disintegrate. I sug-
gest that the nomadic hunter-herder way of life and death
so characteristic for the harra for many centuries was
socially and culturally marginalized by external force. By
the fourth century at the latest, it was replaced by what
may have been a seminomadic pastoral mode of existence
closely attached to the new, settled communities nearby.
References
Adams, R. M., P. J. Parr, M. Ibrahim, and A. S. Al-Mughannum.
. Saudi Arabian Archaeological Reconnaissance .
Atlal :–.
Akkermans, P. M. M. G., and M. L. Brüning. . Nothing but
Cold Ashes? e Cairn Burials of Jebel Qurma, Northeastern
Jordan. Near Eastern Archaeology :–.
, and H. O. Huigens. . Long-Term Settlement
Trends in Jordan’s Northeastern Badia: e Jabal Qurma
Archaeological Landscape Project. Annual of the Department
of Antiquities of Jordan :–.
, H. O. Huigens, and M. L. Brüning. . A Landscape
of Preservation: Late Prehistoric Settlement and Sequence
in the Jebel Qurma Region, North-Eastern Jordan. Levant
:–.
. . . is full professor of Near Eastern Archaeology in the Faculty of Archaeology of Leiden University, the
Netherlands. His research interests include small-scale societies, arid environments and cultural developments therein, as well as
landscape archaeology. He has over years of experience in directing archaeological research projects in Syria and Jordan, and is
currently head of the Jebel Qurma Archaeological Landscape Project in Jordan’s northeastern basalt desert. (Faculty of Archaeology,
Leiden University, PO Box , RA Leiden, e Netherlands; p.m.m.g.akkermans@arch.leidenuniv.nl)
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
430 | LIVING ON THE EDGE OR FORCED INTO THE MARGINS?
Al-Jallad, A. . An Outline of the Grammar of the Safaitic
Inscriptions. Leiden: Brill.
Banning, E. B. . Peasants, Pastoralists and “Pax Romana”:
Mutualism in the Southern Highlands of Jordan. Bulletin of
the American Schools of Oriental Research :–.
. . De Bello Paceque : A Reply to Parker. Bulletin of the
American Schools of Oriental Research :–.
Barham, N., and H. Mensching . Pressure upon the
Pastureland Environment in Jordan. Applied Geography and
Development :–.
Bar-Matthews, M., and A. Ayalon . Mid-Holocene Climate
Variations Revealed by High-Resolution Speleothem Records
from Soreq Cave, Israel, and eir Correlation with Cultural
Changes. e Holocene :–.
Berghuijs, K., and P. M. M. G. Akkermans. Forthcoming. e
Politics of Salt: Azraq and the Wadi Sirhan.
Betts, A. V. G. . e Later Prehistory of the Badia: Excavations and
Surveys in Eastern Jordan, Volume 2. Oxford: Oxbow Books.
. . Exploring the Badia. Near Eastern Archaeology :–.
, and S. Helms. . Rock Art in Eastern Jordan: “Kite”
Carvings? Paléorient :–.
Braemer, F., J.-C. Échallier, and A. Taraqji. . Khirbet Al
Umbashi: villages et campements de pasteurs dans le “désert
noir” (Syrie) à l’âge du Bronze. Beirut: Institut français du
Proche-Orient.
Brusgaard, N. Forthcoming. Images and Interactions: Nomadic Rock
Art from the Jebel Qurma Region of the Black Desert, Northern
Jordan. PhD diss, Leiden University.
Crassard, R., O. Barge, C.-E. Bichot, J. É. Brochier, J. Chahoud,
M.-L. Chambrade, C. Chataigner, K. Madi, E. Régagnon,
H. Seba, and E. Vila. . Addressing the Desert Kites
Phenomenon and Its Global Range through a Multi-Proxy
Approach. Journal of Archaeological Method and eory
:–.
Crompton, A. . Confronting Marginality in the North
Atlantic: Archaeological and Historical Perspectives from
the French Colony of Plaisance, Newfoundland. Historical
Archaeology :–.
Dunbar, R. I. M., and R. Sosis. . Optimising Human
Community Sizes. Evolution and Human Behavior
:–.
Graf, D. F. . e Saracens and the Defense of the Arabian
Frontier. Bulletin of the American Schools of Oriental Research
:–.
. . Rome and the Saracens: Reassessing the Nomadic
Menace. In L’Arabie préislamique et son environnement histo-
rique et culturel, ed. T. Fahd, –. Leiden: Brill.
Haiman, M. . Cairn Burials and Cairn Fields in the Negev.
Bulletin of the American Schools of Oriental Research
:–.
Hassan, F. A. . Demographic Archaeology. New York: Academic
Press.
Helms, S. W. . Jawa: Lost City of the Black Desert. London:
Methuen.
Hill, R. . e Baghdad Air Mail. London: Arnold.
Holly, D. H. . Subarctic “Prehistory” in the Anthropological
Imagination. Arctic Anthropolog y :–.
Horseld, A. . Journey to Kilwa, Transjordan. e Geographical
Journal :–.
Huigens, H. O. . Preliminary Report on a Survey in the
Hazimah Plains: A Hamad Landscape in North-Eastern
Jordan. Palestine Exploration Quarterly :–.
. a. e Identication of Pathways on Harra Surfaces in
North-Eastern Jordan and eir Relation to Ancient Human
Mobility. Journal of Arid Environments :–.
, b. Mobile Peoples—Permanent Places. e Construction
and Use of Stone-Built Architecture by Nomadic Communities in
the Jebel Qurma Region of the Black Desert (Jordan) between
the Hellenistic and Early Islamic Periods. PhD diss., Leiden
University.
Kennedy, D. . Archaeological Explorations on the Roman Frontier
in North-East Jordan. BAR International Series . Oxford:
British Archaeological Reports.
. . Roman Roads and Routes in North-East Jordan.
Levant :–.
. . e Roman Army in Jordan. London: e Council for
British Research in the Levant.
. . Gerasa and the Decapolis: A “Virtual Island” in
Northwest Jordan. London: Duckworth.
. . e Roman Army and Frontier East of the Dead Sea.
Journal of Roman Archaeology :–.
. . e “Works of the Old Men” in Arabia: Remote
Sensing in Interior Arabia. Journal of Archaeological Science
:–.
Köhler, I. . Animal Remains. In Jawa: Lost City of the Black
Desert, ed. S. W. Helms, –. London: Methuen.
Lancaster, W., and F. Lancaster. . People, Land and Water
in the Arab Middle East: Environments and Landscapes
in the Bilâd ash-Shâm. Amsterdam: Harwood Academic
Publishers.
Lawrence, T. E. []. Seven Pillars of Wisdom. London: Cape.
Macdonald, M. C. A. . e Seasons and Transhumance in the
Safaitic Inscriptions. Journal of the Royal Asiatic Society :–.
. . Nomads and the Hawran in the Late Hellenistic and
Roman Periods: A Reassessment of the Epigraphic Evidence.
Syria :–.
. . Of Rock-Art, “Desert Kites” and Mesayid. In Arabia
Vitalis, ed. A. V. Sedov and I. M. Smulyanskaya, –.
Moscow: Rossijskaya Akademiya Nauk.
Meyer, J. C. . Palmyrena: Palmyra and the Surrounding
Territory from the Roman to the Early Islamic Period. Oxford:
Archaeopress.
Migowski, C., M. Stein, S. Prasad, J. F. W. Negendank, and A.
Agnon. . Holocene Climate Variability and Cultural
Evolution in the Near East from the Dead Sea Sedimentary
Record. Quaternary Research :–.
Müller-Neuhof, B. a. A “Marginal” Region with Many
Options: e Diversity of Chalcolithic/Early Bronze Age
Socio-Economic Activities in the Hinterland of Jawa. Levant
:–.
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms
JOURNAL OF EASTERN MEDITERRANEAN ARCHAEOLOGY AND HERITAGE STUDIES | 431
. b. Recent Research on the Late Prehistory of the Arid
Regions in Jordan. Levant :–.
Parker, S. T. . Romans and Saracens: A History of the Arabian
Frontier. Winona Lake, IN: Eisenbrauns.
. . e Roman Frontier in Central Jordan: Final Report
on the Limes Arabicus Project, 1980–1989. Washington, DC:
Harvard University Press.
Parr, P. J., J. Zarins, M. Ibrahim, J. Waechter, A. Garrard, C. Clarke,
M. Bidmead, and H. Al-Badr. . Preliminary Report on the
Second Phase of the Northern Province Survey /.
Atlal :–.
Philip, G., and J. Bradbury. . Pre-Classical Activity in the
Basalt Landscape of the Homs Region, Syria: Implications
for the Development of “Sub-Optimal” Zones in the
Levant during the Chalcolithic-Early Bronze Age. Levant
:–.
Rees, L. W. B. . e Transjordan Desert. Antiquity :–.
Roe, A. G. . Pastoral Livelihoods: Changes in the Role and
Function of Livestock in the Northern Jordan Badia. PhD diss.,
Durham University.
Rollefson, G. O. . Late Prehistoric Aggregation Patterns in
Jordan’s Eastern Badia. Syria :–.
. . Greener Pastures: th and th Millennia Pastoral
Potentials in Jordan’s Eastern Badia. In Palaeoenvironment
and the Development of Early Settlements, ed. M. Reindel,
K. Bartl, F. Lüth, and N. Benecke, –. Rahden:
Leidorf.
, Y. Rowan, and A. Wasse. . Neolithic Settlement at
Wisad Pools, Black Desert. Neo-Lithics /:–.
, Y. Rowan, A. Wasse, A. C. Hill, M. Kersel, B. Lorentzen, K.
Al-Bashaireh, and J. Ramsay. . Investigations of a Late
Neolithic Structure at Mesa , Wadi al-Qatta, Black Desert,
. Neo-Lithics /:–.
Rowan, Y. M., G. Rollefson, A. Wasse, A. “Chad” Hill, and M. M.
Kersel. . e Late Neolithic Presence in the Black Desert.
Near Eastern Archaeology :–.
, A. M. R. Wasse, G. O. Rollefson, M. M. Kersel, M. D.
Jones, and B. Lorentzen. . Late Neolithic Architectural
Complexity at Wisad Pools, Black Desert. Neo-Lithics
/:–.
Sartre, M. . e Middle East under Rome. Cambridge, MA:
Belknap Press of Harvard University Press.
Turner, S., and R. Young. . Concealed Communities: e
People at the Margins. International Journal of Historical
Archaeology :–.
Walsh, K., S. Richer, and J.-L. de Beaulieu. . Attitudes
to Altitude: Changing Meanings and Perceptions within
a “Marginal” Alpine Landscape: e Integration of
Palaeoecological and Archaeological Data in a High-Altitude
Landscape in the French Alps. World Archaeology :–.
Willcox, G. . Plant Remains. In Jawa: Lost City of the Black
Desert, ed. S. W. Helms, –. London: Methuen.
. . Charcoal Analysis and Holocene Vegetation
Historyin Southern Syria. Quaternary Science Reviews
:–.
Woolfenden, W. B., and L. Ababneh. . Late Holocene
Vegetation in the Azraq Wetland Reserve, Jordan.
Quaternary Research :–.
This content downloaded from
132.229.201.169 on Fri, 29 Nov 2019 13:39:58 UTC
All use subject to https://about.jstor.org/terms