Investigations of a Late Neolithic Structure at Mesa 7, Wadi al-Qattafi, Black Desert, 2015

Full text

Rollefson et al., Wadi al-Qatta
Neo-Lithics 1/16 3
represent two small hamlets distinguished by the pre-
sence or absence of attached animal enclosures. Char-
coal from the interior of SS-11 produced a date of 6450
± 40 B.P. (2σ 5480–5320 calBC [Beta-346614]). Betts
has suggested that houses with attached animal pens
are probably late developments in the Late Neolithic of
the Black Desert (Betts 2013: 189), which would mean
that the corbeled houses without animal pens could be
from earlier in the Late Neolithic period.
One of the highest densities of construction was on
the slopes and base of M-7, where more than 285 build-
ings could be identied from APAAME aerial photos
(Rollefson et al. 2014: 296 and Fig. 16). In addition to
the bewildering number of structures, there was also
clear architectural variability, with many buildings
erected using corbeling techniques and others charac-
terized by walls of basalt slabs standing on edge/end.
One structure in particular was of interest because it
Investigations of a Late Neolithic Structure at Mesa 7,
Wadi al-Qatta, Black Desert, 2015
Gary O. Rollefson, Yorke Rowan, Alexander Wasse, Austin C. Hill, Morag Kersel,
Brita Lorentzen, Khaled al-Bashaireh and Jennifer Ramsay
Introduction
The Eastern Badia Archaeological Project resumed its
research program in the Black Desert of Jordan’s pan-
handle with a three-week excavation and survey season
at Mesa 7 (hereafter M-7) in June, 2015. Earlier recon-
naissance of the more than 20 mesas capped with Late
Miocene Abed Olivine-Phyric basalt (Rabba’ 2005)
on both the eastern and western sides of the Wadi al-
Qatta demonstrated dense distributions of more than
600 structures (not including animal pens) from Mesa
3 to Mesa 10 (Figs. 1 and 2).
Excavation in 2012 of SS-11, a small dwelling adja-
cent to an animal corral at Mesa 4 („Maitland’s Mesa“),
showed it to be one of 20 or more corbeled houses on
the southern slope of M-4 (Wasse et al. 2012; Rollefson
et al. 2014); at least 11 corbeled houses had attached
animal corrals while nine did not. The buildings might
Fig. 1 Location of the mesas in the Wadi al-Qatta. The numbering sequence follows the one established by APAAME.
(© Google Earth)

Field Reports
Neo-Lithics 1/16
4
evidenced a combination of the two general approa-
ches. This building (SS-1, for „South Slope-1“) was
selected for excavation due to its unique structural
character as well as its potential for demonstrating ar-
chitectural renovation over time.
Structure SS-1
SS-1 is a curvilinear building with external dimen-
sions of c. 6.30 m NE-SW by 5.40 m NNW-SE (inte-
rior dimensions 4.60 x 4 m). Time constraints allowed
only the western half of the building to be excavated,
but this was enough for a general appraisal of its
structural history. There are two major construction
phases: the original circular structure that was aban-
doned eventually, followed by a remodeling phase
with a new function that witnessed many occupational
visits and cultural deposits until nal abandonment.
Finally, the collapsed structure was used as a small
burial mound.
Phase 1
Phase 1 is represented by Wall 001, a thick (ca.
90-95 cm) double-leaf construction with rubble ll
between the interior and exterior surfaces of the wall
that was built on bedrock (Wadi Shallala Chalk forma-
tion). No clear evidence of a doorway was found for
this phase, although an opening may have been made
in the northern part of the wall and reclosed at a later
time; it isn’t certain if this was during Phase 1 or dur-
ing the succeeding period. It is possible that the main
entrance was in the unexcavated part of SS-1. Wall 001
likely served as part of a corbeled structure. Almost
nothing can be said about the nature of the occupation
of SS-1 during this phase due to the activities of the
people who rst recongured the building in Phase 2.
Phase 2
Phase 2 appears to have been a long sequence
of rebuilding and subsequent renovations. A
new wall (Wall 002) was erected against the
interior face of Wall 001 after the Phase 1 ll
had been completely scoured down to bed-
rock, with thin basalt slabs that constituted
Wall 002 set vertically on end (Figs. 3 and 4)
beneath where the Phase 1 oor would have
been. Wall 002 is curious, not only because
there was a complete change in the orien-
tation of the basalt blocks (and therefore a
change in any kind of superstructure), but
also because Wall 002 seems only to have
continued around a semicircular arc inside
the western half of SS-1. A single pillar ca.
40 x 30 x 100 cm was erected against the
interior of Wall 001 at the northern end of Wall 002;
another pillar of similar dimensions was raised in the
center of the oor, and a third pillar at the southern
Fig. 2 Aerial view of the dense distribution of
buildings around M-7. (Photo by David Kennedy; ©
AAPAME, by permission)
Fig. 3 Top plan of the western half of Structure SS1. (Drawing
by M. Kersel)

Rollefson et al., Wadi al-Qatta
Neo-Lithics 1/16 5
end of Wall 002 had collapsed inwards. Another set of
pillars, one partially collapsed, occurs at the center of
Wall 002 at the southwestern edge of the room.
The pillars at both ends of Wall 002 and the one
in the building’s center form a straight line that bi-
sects the building into generally eastern and western
halves. While Wall 001 in the eastern half could still
serve as the base for a corbeled roof, the half con-
taining Wall 002 in the other half would not have
served the same purpose. It is likely, in fact, that the
western half of SS-1 was left unroofed, and that only
the eastern half had some form of protection from the
elements.
Phase 2 provided the bulk of information about the
use of SS-1. The earliest layer of this phase included
making a shallow oval plaster basin (65 x 50 x 10 cm)
on the newly re-exposed bedrock adjacent to Wall 002
in the western part of the structure. After a brief period
of abandonment represented by a thin (ca. 3 cm) layer
of sediment, a second, slightly smaller plaster basin
was made directly above the rst; traces of the use of
ngers to smooth the surface of the upper basin were
clearly visible (Figs. 5 and 6). The gypsum plaster
seems to be a ner and more durable material than the
plaster in the oor, basin, and alcove of house W-66
at Wisad Pools (Rollefson et al. 2012), although like
the plaster from Wisad W-66, the SS-1 plaster also
contained charcoal from burnt, small-diameter twigs
of the Chenopodioideae (goosefoot) sub-family of
Amaranthaceae. Like the charcoal from Wisad, these
Chenopodioideae twigs are most likely from Anab-
asis, which is a low woody shrub that grows in dry
steppe and desert environments, particularly on rocky
terrain (Cordova 2007).
Fire pits were common in the eastern (roofed?)
half of SS-1, including two that had been cut down
through earlier sediments of Phase 2 to bedrock. Ashy
deposits that lled and accumulated in the eastern half
were dense, thick, and widespread, again restricted
principally to the area east of the western and central
standing stones. Charcoal within these pits was abun-
dant. Of the 14 charcoal fragments larger than 2 mm
that were sampled, 12 could be identied, all of which
are from Amaranthaceae subfam. Chenopodioideae.
Chenopodioideae species in Jordan are all generally
low shrubs common in semi-arid and arid steppe and
desert, or degraded and saline environments. The
dense wood of Chenopodiodeae taxa is considered a
high-quality fuel by the badia’s inhabitants even today,
and their leaves have additional fodder and medicinal
uses (Lancaster and Lancaster 1999: 174). Based on
the curvature of the phloem bundles, all of the SS-1
charcoal is from small-diameter stems, which would
have been easier to collect and transport to the site.
Two radiocarbon dates from early in Phase 2 are
stratigraphically consistent: sample 15 came from
Fig. 4 Orthophoto image of SS-1. PB indicates the location
of the nested plaster basins, and Pi indicates the location of ve
pillars. (Photo: G. Rollefson)
Fig. 5 Nested gypsum plaster basins at the base of Wall 002.
Scale is 25 cm. (Photo: G. Rollefson)
Fig. 6 Close-up of nested plaster basins. Finger marks are
visible above and to the left of the arrow. (Photo: G. Rollefson)

Field Reports
Neo-Lithics 1/16
6
Locus 026, the ll of a pit cut into Locus 0290, and
sample 16 from Locus 029 yielded stratigraphically
consistent dates: 6455-6390 calBC (2ϭ, calBP 8405-
8340) for the rst and 6490-6430 calBC (2ϭ, calBP
8440-8380) for the second. Since both radiocarbon
dates are on short-lived, small-diameter stems, inbuilt
age-related biases should be relatively small for both
samples. Both radiocarbon dates are roughly a mill-
ennium later than structure SS-11 at M-4 (5480-5320
calBC, 2ϭ, calBP 7430-7280; cf. Wasse et al.2002) but
close to that from Wisad Pools structure W-66 (6600-
6460 calBC [Beta-346621: 2ϭ, calBP 8550-8400])
(Rollefson et al. 2014: 291). Structure W-80 at Wisad
Pools produced ve dates, again stratigraphically
consistent (6590-6580 and 6570-6440 calBC for the
lowest layer, just above the oor [Beta 366676: 2ϭ,
calBP 8540-8530 and 8520-8390] to 5710-5610 calBC
and 5590-5570 calBC [Beta 366675: 2ϭ, calBP 7660-
7560 and 7540-7520] to a layer near the top of the Late
Neolithic ll of the structure) (Rollefson et al. 2014:
291).
Flotation samples were also productive for the reco-
very of carbonized plant material from the site. From
the seven otation samples taken several specimens
of the genus Arnebia (Arnebia) were identied, which
are commonly desert annuals or perennials and have
roots that can produce a red dye (Feinbrun-Dothan
1978: 68-70). Seeds from the genus Astragulus (milk-
vetch) were noted and some species in the genus are
common pot herbs and can be found in desert envi-
ronments (Zohary 1972: 54-83; Townsend and Guest
1974: 231-442). Aizoon (stone crop) is a genus that has
many species that are generally herbs or low-shrubs.
These plants can be found in dry wadi stream beds and
oases in hot deserts (Zohary 1966: 74-75). These three
species of plants all ower in the spring, which points
to an occupation period of the site in the late spring.
Also found at the site was evidence of wild grasses
in the form of carbonized grains and culms (stalks).
The most interesting nd from an archaeobotanical
perspective is evidence of Ficus carica (g) at the site.
Not only have the achenes (seeds) of g been identi-
ed but also carbonized fragments of the esh (fruit).
Fig is the rst domesticated species in the Near East
dating to the Early Neolithic period. However it is dif-
cult to dierentiate the wild species (F. persica) from
the domesticated F. carica (Zohary 1966), but both
have signicant water requirements that may point to a
more verdant environment around Wadi al-Qatta du-
ring the Late Neolithic1. However, gs may not have
been grown locally since they have excellent storage
properties and can be transported over long-distances.
Phase 2 consisted of a sequence of superimposed
hearths and ashy deposits in the eastern half of SS-1
and a succession of int-rich sediments in the „open
air courtyard“ of the western part of the building. In
the eastern section there were two paving episodes
(Locus 011 above Locus 024) separated by c. 30 cm of
intervening sediment; paving stones were not coherent
(if present at all) in the courtyard sector2.
Phase 3
Locus 007, which might represent a ceiling collapse in
the eastern roofed part of SS-1, marks the end of the in-
tensive utilization of the building. Only a low remnant of
the walls would have remained by the time sediment and
basalt blocks had lled in the Phase 2 structure. Locus
006, adjacent to the east, is also probably part of the same
collapse, although the slabs may have been re-arranged
in post-abandonment Phase 3 to form a small expanse
of capstones over Locus 009, a shallow burial. (Only a
few bones were preserved, including a human tooth. It
is possible that a headstone exists at the eastern edge of
Locus 009 but it couldn‘t be excavated this season).
Phase 3 is also marked by some reworking of Wall
001 near the far northwest part of the building. In ad-
dition to altering the wall (rather poorly; perhaps hast-
ily in conjunction with the burial in Locus 009?), a
U-shaped organization of boulders (rather than rectan-
gular slabs) was built adjacent to Wall 001 in the north.
It appears to have been an ad hoc construction that saw
little duration of use.
Chipped Stone Artifacts
Tools
Chipped stone artifacts were numerous inside SS-1 as
well as externally adjacent to the building. Table 1 pre-
sents the distribution of tool types. Some layers were vir-
tual „burin sites“: it is notable that almost half (48.4%)
of the shaped tools were burins of various sorts: 26%
were simple or transverse burins, 11% were dihedral
burins, and the rest were versions of truncation burins.
Drills were the next most frequent tool at 11.3%. Scra-
pers made up 5.1% of the inventory, and knives of var-
ious sorts (including the „tabular/cortical tools”) made
up 4.8% of the tool kit. Notches and denticulates were
surprisingly numerous, with a combined total of 13%.
For tools other than utilized and irregularly re-
touched pieces, blades and bladelets were used for
59.0% of the tools, 30.7% were on akes, 2.7% on
cores, and 7.6% of the tool blanks could not be de-
termined. Among utilized and irregularly retouched
elements, 68.6% were on blades, 26.9% on akes, and
4.5% of the blanks were unclassiable.
Projectile points were relatively numerous
(Table 2). Of the identiable types, Badia points (rst
described by Betts in McCartney 1992: 44; cf. Betts
1998: Fig. 4.14) were strongly represented (Fig. 7).
The rst three types listed in Table 2 were relatively
light in weight (Fig. 8): Haparsa points averaged 0.3
grams, Nizzanim points 0.8 gm, and Herzliya points
0.6 gm. Badia points, on the other hand, ranged from
1.5 to 3.0 gm, averaging 2.2 gm. (One miniature Badia
point is an outlier at 0.4 gm and is not included in the
previous statistics). The bimodal distribution suggests
that Badia points were probably used for larger game
than the others.

Rollefson et al., Wadi al-Qatta
Neo-Lithics 1/16 7
There has been debate on the function of burins
based on the absence of use-wear on many of these
tools from burin sites (cf. Finlayson and Betts 1990).
One suggestion for the predominance of this tool type
is that they were not tools per se, but that they were
cores for the spalls that were used as bead drills. While
that observation might be true in some regard, Table 3
suggests that most bead drills were, in fact, made on
blades or bladelets, with burin spalls used for only 42%
of the sample. Although intensive analysis of the debi-
tage from the 2015 season has not yet been undertaken,
the impression from the initial sorting suggests that
unretouched burin spalls were not numerous, probably
much fewer than the number of burins would indicate.
This leaves at least one plausible alternative, that many
of the spalls may have been used as teeth in wooden or
bone handles, with the collective assortment used as
carding boards (Quintero et al. 2004: 209-210).
Tool type n % %'
Projectile point 72 4.4 5.4
Sickle 10.1 0.1
Burin 644 39.2 48.4
Truncation 45 2.7 3.4
Endscraper 16 1.0 1.2
Sidescraper 52 3.2 3.9
Notch 98 6.0 7.4
Denticulate 75 4.6 5.6
Perforator 10.1 0.1
Awl/borer 30 1.9 2.3
Drill 150 9.1 11.3
Biface 11 0.7 0.8
Axe/adze 20.1 0.2
Chopper 60.4 0.5
Wedge 28 1.7 2.1
Unifacial knife 28 1.7 2.1
Bifacial knife 80.5 0.6
Seam knife 50.3 0.4
Tuwailan knife 20 1.2 1.5
Tabular/cortical tool 2 1 0.2
Backed element 20.1 0.2
Tanged blade 10.1 0.1
Rectangular microlith 10.1 0.1
Bladelet, exterior retouch 80.5 0.6
Bladelet, interior retouch 10.1 0.1
Bladelet, abrupt backing 20.1 0.2
Other 22 1.3 1.7
Subtotal 1331 100.0
Retouched ake 39 2.4
Retouched blade 100 6.1
Utilized piece 148 9.0
Unclassiable 24 1.5
Total 1642 100.0
Table 1 Absolute and relative
frequencies of chipped stone tools in the
2015 M-7 inventory.
Type n % %'
Haparsa 18 26.9 38.3
Nizzanim 57.5 10.6
Herzliya 8 11.9 17.0
Badia 10 14.9 21.3
Byblos 46.0 8.5
Other 23.0 4.3
Subtotal (47)
Preform 10 14.9
Tang only 46.0
Unclassiable 69.0
Total 67 100.0 100.0
Table 2 Projectile point types in the
2015 M-7 inventory.
Type n % %'
Blade/let, symmetrical 52 35.1 38.8
Blade/let asymmetrical 25 16.9 18.7
Burin spall, symmetrical 24 16.2 17.9
Burin spall, asymmetrical 33 22.3 24.6
Subtotal 134 100.0
Mèche de forêt 32.0
Double-ended 10.7
Bit only 10 6.8
Total 148 100.0 100.0
Table 3 Absolute and relative
frequencies of drill types in the 2015 tool
assemblage from structure SS-1.
Type n % %’
Hammerstone 10.2 0.2
Radial core 9 1.6 2.0
Single face radial core 8 1.4 1.8
Micro-ake core 18 3.2 4.1
Core on a ake 22 3.9 5.0
Single platform, single face ake core 71 12.7 16.1
Single face, multiface ake core 12 2.1 2.7
Single face, multiplatform ake core 25 4.5 5.7
Multiface, multiplatform ake core 69 12.3 15.6
Pyramidal core 50.9 1.1
Semi-pyramidal 17 3.0 3.8
90º change-of-orientation core 52 9.3 11.8
Single platform, single face blade core 78 13.9 17.6
Opposed platform, non-naviform blade core 11 2.0 2.5
Other blade core 21 3.7 4.8
Bladelet core 23 4.1 5.2
Subtotal (442) 100.0
Casual core/tested piece 34 6.1
Unclassiable 85 15.2
Total 561 100.0
Table 4 Absolute and relative frequencies of core types in the M-7 SS-1 sample.
Among the knives were ve examples that were bi-
facially retouched on pieces of seam int. In addition,
31 large fragments of seam int were cached against an
interior corner of a low wall (Locus 019) just outside to
the northeast of SS-1 (Fig. 9), and seven other pieces of
unretouched seam int were found inside the structure.
The blanks and the knives averaged about 4.5 mm in
thickness at midpoint, and 13 of the blanks had unpat-
terned scratch marks in the cortex.
Cores
Cores (Fig. 10) were relatively numerous (Table 4), re-
ecting intensive tool manufacture inside the structure.
Cores producing akes made up 56.5% of the classi-
able cores (not counting the hammerstone), 33.4%
were blade/bladelet cores, and 10.1% produced both
akes and blades (especially pyramidal, semi-pyra-

Field Reports
Neo-Lithics 1/16
8
qa north of the mesas (including one source of excel-
lent quality material dark brown to black in color; the
hilltop was also the location of a burin site c. 60 x 40m,
with numerous burins of large size), as well as outcrops
eroding out of WSC hills to the southwest of the mesas.
Ground Stone
Groundstone artifacts were not particularly abundant
(Table 5), but they nevertheless demonstrate plant
processing inside the Phase 2 structure; four of the
grinding slabs had been re-used as paving stones. The
midal, and 90° change-of-orientation types). Most of
the cores were heavily reduced, with average length at
43.6 mm, width at 36.1 mm, and thickness at 25.1 mm.
There appears to be a high correlation between excel-
lent int quality and the degree of reduction. Several
of the larger cores were converted to pecking stones to
shape grinding stones.
Flint of medium to ne quality – mostly of brown
color – was found on pedestrian survey eroding out
of Wadi Shallala Chalk (WSC) hills just opposite
the mesas on the eastern side of the wadi. Other int
sources were also found on hills just beyond the large
Fig. 7 Badia points from SS-1. (Photo: G. Rollefson) Fig. 8 Haparsa points from SS-1. (Photo: G. Rollefson)
Fig. 9 Cache of 31 pieces of seam int. b: one of the ve seam knives from SS-1. (Photo: G. Rollefson)

Rollefson et al., Wadi al-Qatta
Neo-Lithics 1/16 9
(Fig. 11). In addition, ve pieces of clinopyroxene,
a black, shiny crystalline mineral associated with
volcanic activity were found, but the material, although
supercially similar to obsidian, is essentially useless
for tool production (Betts 1985).
Aerial Survey and 3D Recording
As part of the 2015 season, a limited aerial survey of
m-7, m-8, and m-9 was undertaken using an Unmanned
Aerial Vehicles (UAV). A series of ights with a xed
wing autonomous drone created a set of aerial images
that could be post processed using Agisoft Photoscan
Pro and ArcGIS. These images were used to create a
high-resolution map of the landscape around M-7 and
served as a pilot project for a larger survey of the grea-
ter Wadi Qatta area in order to document the extant
structures on the surface (see Fig. 12).
Additionally, SS-1 was recorded throughout the ex-
number of grinding elements is in stark contrast to the
case of SS-11 at Mesa 4 (“Maitland’s Mesa”), where
only two small fragments were recovered from an area
and volume of sediments even greater than the situa-
tion at SS-1 (Wasse et al. 2012). However, the intensity
of plant processing at SS-1 pales in comparison to the
near industrial scale of grinding at structure W-80 at
Wisad Pools (Rollefson et al. n.d.), though it is possible
that many more grinding stones will be recovered from
the other half of SS-1.
Small Finds
Table 6 lists the small nds from the 2015 season. The
number of beads of all types (n=30), including both
whole and broken pieces, is far below the number of
drills (see Table 1), which might mean bead production
was intended for trade. Outcrops of Dabba marble and
a coarse red stone were found on exposed WSC hills
on the eastern side of Wadi al-Qatta and at the north-
ern edge of the large qa (“Landing Ground E” in the
nomenclature of the British Royal Airmail service; Hill
1929) to the north of the mesas.
Four pieces of obsidian were recovered, includ-
ing one bifacially retouched ake 56 x 37 x 12 mm
Fig. 10 . Cores from SS-1. a: 90° change-of-orientation
(blade:blade). b: bladelet core. c: single face, single platform blade
core. d: bladelet core. (Photo: G. Rollefson)
Fig. 11 Obsidian ake with bifacial retouch. (Photo: G. Rollefson)
Item n
Grinding slabs 10
Mortar 1
Hand stones 26
Scoria/pumice fragments 2
Sandstone palette fragments 5
Table 5 Ground stone artifacts from
structure SS-1, M-7, 2015.
Item n
Shell beads 17
Dabba marble beads 6
Redstone beads 6
"Other" stone bead 1
Spindle whorl 1
Obsidian pieces 4
Clinopyroxene pieces 5
Carnelian fragment 1
Animal gurine fragment 1
Gizzard stones 3
Limestone "nger ring" 1
Drilled limestone fragment 1
Bone awl12
Seam int cache21
1 More bone tools are expected as faunal analysis proceeds.
2 The seam int cache contained 31 pieces.
Table 6 Small nds from SS-1, Mesa-7, 2015.

Field Reports
Neo-Lithics 1/16
10
at M-7 is more than 20 times the tool total
from the interior of the M-4 building and
its immediate surroundings. Broadly spea-
king, the relative importance of particular
tool types is similar in the two inventories
(e.g., high percentages of burins, notches
and denticulates, and a low percentage of
arrowheads), so one interpretation is that
the M-7 structure was used over a longer
period than the one at M-4, and that each
visit to M-7 during Phase 2 was of a longer
duration than what appear to be shorter term
stays at M-4. If this were so, it might be the
case that the landscape around the mesas
had changed substantially by the middle of
the 6th millennium BC.
A comparison with another, generally contempora-
neous assemblage also shows considerable dierences.
The excavation of structure W-80 at Wisad Pools
(Rollefson et al. 2013; Rowan et al. 2015) produced a
large inventory of chipped stone tools (Rollefson et al.
n.d.). W-80 was a large, complex oval building (c. 5 x
6 m) that began as a corbeled dwelling, but after the
original roof collapsed the walls were used as a wind-
break for various activities. The volume of sediments
dug from W-80 is well over twice the amount exca-
vated during the 2015 season at SS-1. The number of
tools at SS-1 is relatively higher than at W-80 (1642 for
the western half of SS-1 vs. 2346 from all of W-80),
although the number of cores is roughly proportionally
equal (561 for the west half of SS-1 vs. 1066 for all
of W-80). Perhaps the most striking dierences lie
among the tools: Projectile points account for 27.2%
of the W-80 tools (n=629), while they reach only 5.4%
at SS-1 (Table 1). Among the arrowhead types, 85.2%
are transverse types at Wisad Pools, while not a single
example was recovered at Mesa 7; on the other hand,
large and heavy Badia points make up almost 15% at
SS-1, while none have been found at W-80 or any other
excavated structure at Wisad Pools. Badia points occur
in the earlier levels at Site 27000 at Burqu, approxi-
mately 65 km north of Wisad Pools, but they disappear
and are replaced entirely by transverse arrowheads in
the later Late Neolithic layers (Betts et al. 2013:93).
Other principal dierences include burins, where
the heavy representation at SS-1 (48.4%) far exceeds
the weak gure of 2.5% at Wisad W-80; burin classes
were also very dierent, with truncation types at 66.3%
at SS-1 vs. 40.4% at W-80 (cf. Rollefson 1995). Despite
a large number (n=80) of beads recovered from W-80,
bead drills were scarce (4.1%); at SS-1, there were 30
beads but bead drills were 9.1% of the tools.
The dierences of the tool inventories between M-7
SS-1 and M-4 SS-11 are probably heavily inuenced
by dierences in activities undertaken at the two lo-
cations and the large amount of time between the two
cavations with a Pole Aerial Photography (PAP) setup
to create a time series of low elevation image sets.
These image sets were also processed with Agisoft
Photoscan Pro and ArcGIS in order to produce a series
of spatially accurate 3D models of the structure as it
was excavated (see gure 4).
Discussion
One of the reasons SS-1 was selected for excavation was
that its architectural dierences compared to structure
SS-11 at Mesa-4 (“Maitland’s Mesa”) suggested there
may have been coincident cultural variability associ-
ated with temporal change and/or activity distinctions.
Unfortunately, there is no clear evidence for how the
rst conguration (Phase 1) of SS-1 at M-7 was used
due to the removal of original occupational deposits in
Phase 2, but based on the dimensions of Wall 001, it
is likely that Phase 1 represents a residential building,
perhaps one that was corbeled. Phase 2, on the other
hand, appears to be a workshop that was the locus of
several dierent kinds of undertaking.
Orthophotos from the UAV survey, recorded from
an altitude of 100m and at a resolution of approximately
4cm/pixel, provide some tantalizing indications that
the structures on the slopes of M-7 are not randomly di-
stributed but instead might in many instances be spaced
in clusters of four or more closely spaced buildings oc-
cupied by cooperative economic units. SS-1 and ve
other buildings are grouped in close proximity over
an area of 18m E-W by 15m N-S (Fig. 12). Notably,
SS-1 is the only building with a wall constructed of
perpendicular slabs, and perhaps the activities carried
out inside it were for the benet of the residents of the
other ve structures.
Table 1 and Table 7 show how dierently M-7 SS-1
was used compared to M-4 SS-11 (cf. Wasse et al.
2012). In general, the volume of sediment excavated
during the 2012 and 2015 seasons are about the same,
yet the total number of tools in only half of the structure
Fig. 12 Orthophoto of a cluster of buildings in the
SS-1 area. North is at the top of the picture. (Photo:
A.C. Hill)

Rollefson et al., Wadi al-Qatta
Neo-Lithics 1/16 11
Alexander Wasse
University of East Anglia
alexander.wasse@uea.ac.uk
Austin C. Hill
University of Connecticut
austin.hill@uconn.edu
Morag Kersel
DePaul University, Chicago
mkersel@depaul.edu
Brita Lorentzen
Cornell University, Ithaca
bel9@cornell.edu
Khaled al-Bashaireh
Yarmouk University
khaledsm@email.arizona.edu
Jennifer Ramsay
State University of New York, Brockport
jramsay@brockport.edu
References
Bar-Matthews M., Ayalon A., and Kaufman A.
2000 Timing and Hydrological Conditions of Sapropel
Events in the Eastern Mediterranean, as Evident from
Speleothems, Soreq Cave, Israel. Chemical Geology
169: 145 -156.
Betts A.
1985 Black Desert Survey, Jordan. Third Preliminary
Report. Levant 17: 29-52.
1998 The Harra and the Hamad. Excavations and Surveys in
Eastern Jordan 1. Sheeld Archaeological
Monographs 9. Sheeld: Sheeld Academic Press.
2013 The Later Prehistory of the Badia. Excavations and
Survey in Eastern Jordan 2. Oxford: Oxbow
Books.
Cordova C.E.
2007 Millennial Landscape in Jordan. Tucson: University of
Arizona Press.
Feinbrun-Dothan N.
1978 Flora Palaestina: Part III. Jerusalem: Israel Academy
of Science and Humanities.
Finlayson B. and Betts A.
1990 Functional Analysis of Chipped Stone Artefacts from
the Late Neolithic site of Ǧabal Na’ja, Eastern Jordan.
Paléorient 16: 13-2 0.
Hill R.
1929 The Baghdad Air Mail. London: Longmans, Green and
Co mpa ny.
occupations. Furthermore, the visits to M-7 SS-1 were
probably much more frequent and each of longer dura-
tion than the presence at M-4 SS-11. The comparison
of the SS-1 inventory with W-80 does not involve
much time between the two occupational sequences,
so the distinctions probably lie more in the kinds and
intensities of activities carried out at the two buildings:
both buildings appear to have served as workshops, but
SS-1 was apparently more closely directed to whatever
pursuits are represented by burins, whereas W-80 (with
its strong focus on grinding stones) might have been
more of a domestically oriented occupation.
Concluding Remarks
Excavations will be resumed in the summer of 2016 in
order to nish the exposure of SS-1 and to investigate
its relationship to other buildings in the immediate vi-
cinity. Additional drone mapping of the mesas will be
carried out in the spring of 2016, and additional aerial
survey will be resumed in the region the following
summer.
Acknowledgments: We are indebted to the Department
of Antiquities personnel for their continued enthusiasm
for the goals of the Eastern Badia Archaeological Pro-
ject. We have received considerable support and assis-
tance from Dr. Barbara Porter and the sta at the Ame-
rican Center of Oriental Research (ACOR) and from
the Council for British Research in the Levant (CBRL).
Institutional support and nancial assistance has been
provided by Whitman College, the Oriental Institute of
the University of Chicago, and East Anglia University.
Field work was facilitated by the participation of Kath-
leen Bennallack, Blair Heidkamp, Jackson Somerville,
Chris Yeoh, and Mohammad Nasser.
Endnotes
1 Notably, there is evidence from Soreq Cave speleothems
for high precipitation from 8,500 to 7,000 bp (Bar-Matthews
et al. 2000).
2 It is not imp ossi ble that Ph ase 1 includ ed the construct ion of
Wall 002, in which case Phases 1 and 2 should be considered
to be the same phase. However, the incongruity of the
architectural ensemble renders this an unlikely scenario.
Gary Rollefson
Whitman College, Walla Walla
rollefgo@whitman.edu
Yorke Rowan
University of Chicago
ymrowan@uchicago.edu

Field Reports
Neo-Lithics 1/16
12
2014 The Late Neolithic Colonization of the Eastern Badia
of Jordan. Levant 46(2): 285-301.
Rollefson G., Rowan Y., Wasse A., Kersel M., Hill A.C., and
Lorentzen B.
n.d. Excavation of Structure W-80, a Complex Late
Neolithic Building at Wisad Pools, Black Desert.
Annual of the Department of Antiquities of Jordan. (in
review).
Rowan Y., Wasse A., Rollefson G., Kersel M., Jones M., and
Lorentzen B.
2015 Late Neolithic Architectural Complexity at Wisad
Pools, Black Desert. Neo-Lithics 1/15: 3-10.
Townsend C.C. and Guest E.
1974 Flora of Iraq: vol. 3. Leguminales. Baghdad: Ministry
of Agriculture and Agrarian Reform.
Wasse A., Rowan Y., and Rollefson G.
2012 A Seventh Millennium BC Late Neolithic Village at
Mesa 4 in Wadi al-Qatta, Eastern Jordan. Neo-Lithics
1/12: 15-24.
Zohary M.
1966 Flora Palaestina: Part I. Jerusalem: Israel Academy of
Science and Humanities.
1972 Flora Palaestina: Part II. Jerusalem: Israel Academy
of Science and Humanities.
Lancaster W. and Lancaster F.
1999 People, Land and Water in the Arab Middle East:
Environments and Landscapes in the Bilâd ash-Shâm.
Amsterdam: Harwood Academic Publishers.
McCar tney C.
1992 Preliminary Report of the 1989 Excavations at Site 27
of the Burqu’/Ruweishid Project. Levant 24: 33-54.
Quintero L., Rollefson G., and Wilke P.
2004 Highland Towns and Desert Settlements: Origins of
Nomadic Pastoralism in the Jordanian Neolithic.
In: H.-D. Bienert, H.G.K. Gebel and R. Neef (eds.),
Central Settlements in Neolithic Jordan: 201-213.
Berlin: ex oriente.
Rabba’ I.
2005 The Geology of Umm Nukhayla and Wadi al Qatta
Areas Map Sheets No. 3453-II and 3453-I. Bulletin 62.
Amman: Natural Resources Authority.
Rollefson G., Rowan Y., Perry M., and Abu-Azizeh W.
2012 The 2011 Season at Wisad Pools, Black Desert:
Preliminary Report. Annual of the Department of
Antiquities of Jordan 56: 29-44.
Rollefson G., Rowan Y., and Wasse A.
2013 Neolithic Settlement at Wisad Pools, Black Desert.
Neo-Lithics 1/ 13: 11-23 .