This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
Author's personal copy
Quaternary International 343 (2014) 117e135
Contents lists available at ScienceDirect
Quaternary International
journal homepage: www.elsevier.com/locate/quaint
Cultural change or continuity in the late MSA/Early LSA of
southeastern Ethiopia? The site of Goda Buticha, Dire Dawa area
David Pleurdeau a, Erella Hovers b, *, Zelalem Assefa c, Asfawossen Asrat d,
Osbjorn Pearson e, Jean-Jacques Bahain a, Yin Man Lam f
a
Département de Préhistoire, UMR 7194, CNRS Muséum national d’Histoire naturelle, France
Institute of Archaeology, The Hebrew University, Mt. Scopus, Jerusalem, Israel
Department of Anthropology, Archaeobiology Program & Human Origins Program, National Museum of Natural History, USA
d
Department of Earth Sciences, Addis Ababa University, Ethiopia
e
Department of Anthropology, University of New Mexico, USA
f
Department of Anthropology, University of Victoria, Canada
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 27 March 2014
Goda Buticha is a newly discovered cave site in southeastern Ethiopia, containing MSA and LSA cultural
material, faunal remains, beads, and human skeletal remains. A 2.3 m-deep sedimentary sequence
records two occupational phases separated by a sharp chronological hiatus, in the Upper Pleistocene
(w 43e31.5 ka cal BP) and in the mid- Holocene (7.8e4.7 ka cal BP). Faunal remains suggest changes in
paleoecological conditions that are in agreement with patterns documented in regional speleothembased reconstructions. The lithic assemblage at the base of the sequence is clearly MSA, with Levallois
production, unifacial and bifacial points, relatively large debitage and use of local raw materials, associated with a microlithic component. The overlaying LSA assemblage contains diagnostic artifacts
(backed microliths and bladelet production), with ubiquitous use of obsidian and MSA elements that
appear in the Holocene. In the absence of indications for post-depositional mixture, the apparent cultural
continuity of MSA elements from the Upper Pleistocene into the Middle Holocene at Goda Buticha may
represent yet another variation of the elusive MSA/LSA transition. Goda Buticha is a key site for
reevaluating the dynamics and tempo of this transition in eastern Africa.
Ó 2014 Elsevier Ltd and INQUA. All rights reserved.
Keywords:
MSAeLSA transition
Upper Pleistocene
Eastern Africa
Southeastern Ethiopia
Microliths
1. Introduction
Intact, well-stratified archaeological deposits that preserve bone
and can be dated radiometrically are crucial for our understanding
of the Middle Stone Age (MSA) and earliest Later Stone Age (LSA) in
East Africa. These periods and geographic region are thought to be
highly important for the origin and spread of modern humans as
well as the emergence of important cultural innovations (Clark,
1988; Klein, 1995; Lahr and Foley, 1998; Ambrose, 1998a;
McBrearty and Brooks, 2000; Walter et al., 2000; Clark et al.,
2003; Bruggemann et al., 2004; Forster and Matsumura, 2005;
Marean and Assefa, 2005; McDougall et al., 2005; Beyin, 2006,
* Corresponding author.
E-mail addresses: dpleurd@mnhn.fr (D. Pleurdeau), hovers@mscc.huji.ac.il,
ehovers@gmail.com (E. Hovers), zassefa@gmail.com (Z. Assefa), asrata@geol.aau.
edu.et (A. Asrat), ompear@unm.edu (O. Pearson), bahain@mnhn.fr (J.-J. Bahain),
ymlam@uvic.ca (Y.M. Lam).
http://dx.doi.org/10.1016/j.quaint.2014.02.001
1040-6182/Ó 2014 Elsevier Ltd and INQUA. All rights reserved.
2011b; Behar et al., 2008; Klein, 2008; Tishkoff et al., 2009; Rose
et al., 2011; Delagnes et al., 2012). Unfortunately, only a handful
of stratified sites document the MSA and LSA in East Africa, for
example Mumba rockshelter in Tanzania (Mehlman, 1989; DiezMartín et al., 2009; Gliganic et al., 2012; Eren et al., 2013); Enkapune Ya Moto (Ambrose, 1998b) and Lukenya Hill in Kenya
(Merrick, 1975; Barut-Kusimba, 2001) and Porc-Epic Cave in
Ethiopia.
The Mesozoic limestone of the Antalo Formation, which crops
out extensively in the southern wall of the Rift Valley around Harar
(Ethiopia) (Bosellini et al., 2001), contains numerous karst caves
suitable for paleoecological work (Asrat et al., 2007) and for the
preservation of bone in archaeological deposits (Assefa, 2006).
Previous work on Paleolithic deposits in the region focused mainly
on exploration and excavations at the cave site of Porc Epic
(Teilhard de Chardin, 1930; Breuil, 1934; Teilhard de Chardin et al.,
1940; Breuil et al., 1951; Clark and Williams, 1978; Clark and
Williamson, 1984; Brandt, 1986; Pleurdeau, 2003; Assefa, 2006;
Assefa et al., 2008), with other sites mentioned briefly.
Author's personal copy
118
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
A survey conducted by Assefa et al. (2014) in 2007 and 2009
visited previously reported localities as well as new cave sites,
documenting the potential for datable speleothems, other paleoenvironmental archives and rock art. Caves were examined for MSA
remains that could contribute to the discussion about the dispersal
of modern humans out of Africa as well as cultural evolution in
Africa. Goda Buticha (Buticha cave, in the local Oromo language),
containing thick deposits with both LSA and MSA artifacts, faunal
bones and human remains, is a newly discovered site that had not
been reported by earlier workers in the region (Fig. 1).
2. The site
Buticha cave is located close to the hamlet of Kunama in the
administrative district of Serkama village (Dire Dawa district), situated high in a cliff made up of crystalline limestone of the
Gebredare Series (Fig. 1). It is a three-chambered karst cave with
two mouths opening to roughly NE- and NW-oriented galleries,
joined by a third small and dark SE-oriented chamber. The cave is a
remnant of a larger system, which has been long destroyed, along
NEeSW oriented faults and fractures related to the main Ethiopian
Rift. The main gallery opens to NortheNorth West onto the slope.
The site contains small rock paintings in dark pigment in the
northeastern gallery. In contrast, all the in situ MSA and LSA lithic
artifacts, fauna, human remains and ornaments were found in the
northwestern gallery (Fig. 2). The sediments containing the
Paleolithic material extend laterally into the SE chamber gallery. A
large pit close to the cave’s entrance excavated some years ago by a
local villager, and then used by porcupines, revealed a sediment
column ca. 1.5 m thick, in which fauna and lithic artifacts appeared
in situ. Cleaning of the pit revealed a long sequence of intact sediments. The southern wall of the pit was used as a stratigraphic
guide for the test excavation conducted at the site.
Save for the sediments from the very top 40 cm of the section,
which were excavated in square-meter units, excavation was conducted in 50 50 cm quadrants. For the purpose of the test excavation, arbitrary excavation spits of 5e10 cm each were used, with
constant evaluation of the lithostratigraphic conditions (changes in
color and texture). On the whole an area of 2 m2 was excavated
down to bedrock, which was exposed on the western side of the
excavation (Figs. 2 and 3) during two field seasons in 2008 and
2011. A 230-cm sedimentary sequence has been exposed, with all
archaeological remains >2 cm plotted in a 3D grid system. All the
excavated sediment was dry-sieved through a 2 mm mesh. In total
a sediment volume of 5 m3 was excavated during these two
seasons.
3. Geochronology and sedimentary history
Macroscopically, two major sedimentary complexes were
distinguished, within which 14 sub-horizontal layers as well as 2
large indurated pockets have been identified (Fig. 3).
Fig. 1. General location of Goda Buticha (Dire Dawa area, Ethiopia). Views of the cave from outside and within show its elevation relative to the local topography.
Author's personal copy
119
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
3.1. Complex I (upper complex)
A 50e60 cm thick series of 7 interstratified white, light pink, and
black fine-grained silts underlies a 10e30 cm thick layer of very
recent organic material (straw, wood) in which many hyena coprolites were found. This series is consistent with anthropogenic
ashes from fireplaces. The sediments are rich in long bone shaft
fragments. Remains of micromammals were found throughout the
ash layers, in some cases calcined and concentrated into dense
pockets. A single piece of pottery was found at an elevation of 180e
170 cm above datum.
At the base of the Upper Complex, there is a distinct 10e20 cm
thick horizon that contains a large number of micromammal remains and some artifacts.
Bedrock has been reached within square A1, where the cave wall
is inclined toward the central part, some 80 cm above the datum. A
limited test pit in square B1 revealed that artifact-bearing sediment
continues down to at least 30 cm.
Sediments of Complex II contain a rich lithic assemblage (MSA
and LSA components), faunal remains, human remains, and several
ostrich eggshell beads.
Forty-four charcoal samples were collected from in situ contexts,
at elevations ranging 230e50 cm above datum. Fourteen samples,
all derived from the stratigraphic Lower Complex, were dated using
Accelerator Mass Spectrometry (AMS) radiocarbon dating methods.
The dates were calibrated using the IntCal 13 timescale (Reimer,
2013) (Table 1, Fig. 3).
Table 1
14
C ages on charcoal from Buticha Cave according to elevation and grid.
Elevation
Grid square
A1a/A1b
1
170e160
2
3
160e150
150e140
4
140e130
5
130e120
6
120e110
7
8
110e100
100e90
9
70e60
10
60e50
A1c/A1d
B1a/B1b
B1c/B1d
4130 30
4676 144 cal BP
SacA#27480
(z ¼ 170)
6410 30
7347 71 cal BP
SacA#27479
(z ¼ 141)
*6200 30
7119 119 cal BP
SacA#35205
(z ¼ 132)
**6940 40
7797 118 cal BP
beta#269596
**6370 50
7300 122 cal BP
beta#269595
*6920 50
7791 126 cal BP
beta#255123
*6470 40
7375 82 cal BP
beta#269594
**5590 50
6380 88 cal BP
beta#255124
***6430 40
7350 74 cal BP
beta#269597
***6440 50
7351 80 cal BP
beta#255122
*29,680 230
33,843 405 cal BP
beta#269593
**31,590 270
35,497 595 cal BP
beta#255121
42,500 1000
46,273 2084 cal BP1
SacA#35207
(z ¼ 69)
38,870 670
42,997 1053 cal BP
SacA#35206
(z ¼ 54)
Calibrated dates are in boldface.
Calibration (IntCal 13 curve) at 1 sigma done with Online OxCal 4.2 (https://c14.arch.ox.ac.uk/login/login.php?Location¼/oxcal/OxCal.html).
*Asterisks correspond to the presentation of samples in Fig. 3.
3.2. Complex II (lower complex)
Underlying a 10 cm thick brown clay layer (Layer IIa/b) is a 170e
180 cm thick deposit of light brown clay. The increasing presence of
limestone clasts (5e10 cm large) from an elevation of w100 cm
above the datum led us to separate Complex II into two main layers
(IIc and IId).
The lowest part of Complex II (between 0 and 50e70 cm above
the datum) in square B1 is largely indurated and concreted (IIf).
Another large indurated pocket is distinguishable, with a distinctive orange color (IIe). The formation of these concretions is most
likely post-depositional.
The dates indicate episodes of human occupation in the
late Upper Pleistocene from 42,500 1000 14C BP
(46,273 2084 cal BP) to 29,680 230 (33,843 405 cal BP) in
Layers IIf and IId and during the mid-Holocene from 6920 50
(7791 126 cal BP) to 4130 30 (4676 144 cal BP) in layer IIc
and IIb. However, the available radiocarbon dates suggest a major
hiatus in the accumulation of the fine-grained sediments of the
Lower Complex (II), with the top of Layer IId dated to w29.5e
32 ka 14C BP (33.5e36 ka cal BP) and the lower part of layer IIc
dated to w5.5e7 ka 14C BP (6.3e7.8 ka cal BP). This gap corresponds to the sedimentological change between Layer IId and
Layer IIc.
Author's personal copy
120
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
OSL dating analyses are in progress at the Institut de Recherche sur
les matériaux, University of Bordeaux, France. The very preliminary
results are consistent with these radiocarbon dates and support this
general chronostratigraphy (C. Tribolo, personal communication).
4. The lithic artifacts
The characteristics of the lithic assemblage provide an essential
line of evidence from which to define the cultural stratigraphy at
the site and to assess changes in technological strategies.
A total of 5463 artifacts were collected in 2008 and 2011 (Table 2).
The average density is w1100/m3. However, 87% of the artifacts are
concentrated in the Lower Complex, 30e150 cm above datum. In this
part of the sequence, the lithic density reaches up to w2000 artifacts/
m3. The artifacts are well preserved. Some 200 large (>20 mm) chert
and basalt items show indications of fire damage. In the basal part of
the sequence, 15% of the large pieces are burned. Given the nearly
identical trends that we observed in the samples from 2008 to 2011,
we use the detailed analysis of the 2008 sample to describe the main
characteristics of the assemblage and their temporal trends.
Table 2
Vertical distribution of lithics in Goda Buticha according to excavation sub-squares. The part of the sequence with the main concentration of lithics is shaded. See text for
details.
Elevation above
datum (cm)
Upper Complex I
Lower Complex II
Layers IIaeIIc
Layers IIdeIIf
190e200
180e190
170e180
160e170
150e160
140e150
130e140
120e130
110e120
100e110
90e100
80e90
70e80
60e70
50e60
40e50
30e40
20e30
10e20
0e10
Total
Square A1
Square B1
a
b
c
d
4
a
1
10
60
26
7
17
43
106
34
89
98
95
77
103
111
85
18
34
11
5
22
56
85
30
41
55
59
121
120
190
21
22
10
21
28
53
103
38
97
31
27
24
74
30
35
5
14
61
57
23
29
17
9
27
65
159
136
55
55
6
675
790
159
58
961
905
49
104
54
43
25
43
24
19
25
417
Total
b
27
27
9
66
62
26
27
52
31
34
40
20
37
40
25
0
21
544
c
0
1
26
6
1
14
106
47
d
3
109
5
37
20
24
9
14
48
55
58
15
53
8
37
31
52
142
42
24
29
11
449
722
24
32
7
81
N
8
165
277
65
28
194
519
522
274
326
402
293
419
429
659
479
201
153
25
27
5463
%
0
3
5
1
1
4
10
10
5
6
7
5
8
8
12
9
4
3
0
0
100
Fig. 2. Plan and longitudinal profile of the SW gallery of Goda Buticha and location of the excavation. The bold line of the grid refers to the stratigraphic section of Fig. 3.
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
121
Fig. 3. Synthetic stratigraphic section EeW (transversal) of A/B and 1/2 bands of Goda Buticha. Radiocarbon samples are marked by numbers corresponding to Table 1, where the
dates and lab references are given.
Throughout the sequence, all stages of lithic production are
well represented, with the exception of the initial stage of
decortication. Debris (i.e., flakes smaller than 20 mm and all
chunks) are abundant and constitute 62% of the lithics (Table 3),
indicating that much of the last stages of the knapping
process were conducted on site, whereas the presence of various
types of core trimming elements suggests on-site core
rejuvenation.
Author's personal copy
122
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Table 3
Vertical distribution of typo-technological categories. The part of the sequence with the main concentration of lithics is shaded.
Elevation below
datum (cm)
Pebbles, nodules
N
190e200
180e200
170e180
160e170
150e160
140e150
130e140
120e130
110e120
100e110
90e100
80e90
70e80
60e70
50e60
40e50
30e40
20e30
10e20
0e10
Total
%
1
1
1
1
1
1
1
1
2
7
5
0
Debris
Cores
Chips <20 mm
Chunks
Sub-total
N
N
%
6
87
34
15
39
97
141
101
80
145
66
135
98
87
149
20
22
2
6
1330
4
1
6
30
16
6
8
25
23
12
4
19
53
2
1
210
55
66
63
89
45
45
60
59
50
63
57
72
67
72
77
83
55
21
100
62
N
Debitage (>20 mm)
%
5
1
4
2
1
13
7
6
6
5
4
3
1
4
5
1
5
3
3
3
2
3
1
1
3
2
61*
2
N
%
1
3
31
16
2
52
125
79
57
72
75
49
43
36
32
53
4
17
8
100
27
23
30
11
51
44
30
32
41
28
31
21
24
22
20
17
39
57
755
30
Retouched items
N
2
10
3
18
7
6
3
16
17
10
9
20
14
11
13
5
4
3
6
7
6
5
7
9
5
9
3
2
2
2
5
14
141
6
In this table only 61 cores are reported. Another 8 cores with retouch were included with tools. In other analyses 69 cores are discussed.
Fig. 4. Potential lithic sources in the area of Goda Buticha and Porc Epic.
%
Total
N
%
1
11
137
54
18
101
281
260
180
176
270
157
204
152
148
264
24
44
14
6
2502
100
100
100
100
100
100
100
100
100
100
100
100%
100
100
100
100
100
100
100
100
100
Author's personal copy
123
%
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
5
0
0
7
0
1
19
1
2
13
1
2
N
0
0
1
2
22
3
3
3
1
2
0
0
1
2
0
0
8
0
0
0
2
39
2
2
3
2
1
4
3
9
7
2
3
1
%
N
237
3
1
1
12
12
15
12
15
14
15
20
17
32
51
3
6
7
1
20
2
3
5
1
2
1
1
3
1
1
138
1
11
6
12
8
12
3
11
5
21
38
2
7
N
2
0
0
0
1
0
0
0
0
0
0
1
3
20
12
5
61
169
161
110
86
150
85
100
73
65
127
10
24
3
4
1265
4
2
33
5
2
0
5
1
6
5
6
1
28
190e200
180e200
170e180
160e170
150e160
140e150
130e140
120e130
110e120
100e110
90e100
80e90
70e80
60e70
50e60
40e50
30e40
20e30
10e20
0e10
Total
1
%
%
756
11
104
39
2
17
71
65
43
59
91
53
68
53
28
35
7
7
3
0
100
76
72
11
17
25
25
24
34
34
34
33
35
19
13
29
16
21
0
30
%
N
%
100
0
2
2
6
12
4
6
7
9
5
10
10
11
22
19
13
14
50
0
9
1
N
%
nb
0
0
1
0
0
2
1
2
1
1
0
1
1
0
0
0
0
0
0
17
1
0
0
0
0
0
1
4
2
7
5
4
2
5
3
14
14
8
16
0
17
6
%
N
N
0
0
15
22
28
60
60
62
61
49
56
54
49
48
44
48
42
55
21
67
51
Obsidian
Extrusive Rocks
Basalt
Limestone
Quartz
Endogenous Rocks
Quartzite
Flint
Sedimentary Rocks
Elevation above
datum (cm)
Table 4
Distribution of raw materials. The part of the sequence with the main concentration of lithics is shaded.
4.2.2. Débitage
The technological characteristics of the debitage are consistent
with observations on the cores. Throughout the sequence, lithic
reduction is geared towards the production of flakes. In general,
débitage items seem to have resulted from various unipolar production methods. The products are often elongated and include
laminar and lamellar items. Frequencies of blades and bladelets are
low in the basal part of the sequence (13%e30% of the débitage in
elevations below 70e80 cm) compared to over 50% towards the top
of the Lower Complex (at 90e150 cm above datum) (Table 5; Fig. 6).
This increased elongation of artifacts co-occurs with a decrease in
%
Indeterminate Rocks
Other volcanics.
4.2. Lithic technology
4.2.1. Cores
The 69 cores recovered from the sequence constitute 3% of all
artifacts (7% if chips and debris are excluded). The frequencies of
the various core types do not show a directional change across the
stratigraphic sequence (Table 5). Thirty-two (46%) of all cores are on
obsidian and 31 (45%) are on chert.
Levallois flaking is one of the main core reduction strategies
observed at Goda Buticha, consistent with our knowledge of MSA
technological practices. Levallois cores (n ¼ 34, Fig. 5:1e3),
including 9 flakes used as cores due to their suitable geometry (e. g.,
Delagnes, 1995; Hovers, 2007, Fig. 5:3), occur throughout the whole
sequence. In about half of the cases, the Levallois cores bear some
residual cortex, even though they tend to be small (average length
is 32 mm). Half of the cores show centripetal preparation for
preferential flake removals and the remainder were exploited using
a unipolar recurrent flaking method.
Another quantitatively important group consists of cores with
several flaking surfaces (n ¼ 30), which are by definition (Boëda,
1995) non-Levallois. It is difficult to identify the flaking methods
involved in their reduction. The majority of cores in this group are
very small. They are highly fragmented or heavily exploited (Fig. 5).
In a few cases it was possible to identify that a slightly convex
flaking surface was used for recurrent removals (i.e., discoidal),
which then was turned into a striking platform for exploiting an
opposite, convex surface.
Single platform (prismatic and pyramidal) cores, accounting for
7.2% of the cores (5/69), are more suggestive of Upper Paleolithic/
LSA technologies. This is corroborated by the presence of crested
blades in the assemblage (Fig. 5:6; see below). Some of the prismatic cores were semi-rotated in order to produce blades and
bladelets (Fig. 5:4). Pyramidal cores are small (<20 mm) and were
used for bladelet production from a flat prepared striking platform.
N
Total
The raw materials used throughout the sequence are dominated
by chert (51%), obsidian (30%), basalt (9%), and quartzite (6%). The
remainder consists of quartz and limestone (Table 4). Only a small
number of artifacts (n ¼ 30) show extensive cortical or natural
surfaces. As a result it is impossible to assess whether raw materials
derive from secondary (e.g., transported; river bed cobbles) or
primary (i.e., in situ geological exposures) sources.
Chert and sandstone-quartzite in Buticha Cave are most likely of
local origins. The cave is situated within chert-bearing limestone
beds, and the sandstone-quartzite is probably derived from the
conglomerate layers which cap the karstic limestone in the region
(Fig. 4). The Gara Adi silicic volcanic dome, located only about 15e
20 km to the north, is the closest possible primary source for the
obsidian. All other known potential sources are more than 100 km
away (Negash and Shackley, 2006; Negash et al., 2011; Vogel et al.,
2006); more accurate sourcing awaits geochemical analyses.
1
11
137
54
18
101
281
260
180
176
270
157
204
152
148
264
24
44
14
6
2502
4.1. Raw materials
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Author's personal copy
124
Table 5
Distribution of cores and blank types larger than 20 mm. The part of the sequence with the main concentration of lithics is shaded.
a
Total
Flakes
N
190e200
180e200
170e180
160e170
150e160
140e150
130e140
120e130
110e120
100e110
90e100
80e90
70e80
60e70
50e60
40e50
30e40
20e30
10e20
0e10
Total
Blanksa
Cores
6
3
1
14
8
6
6
6
4
4
1
4
5
1
69
%
N
1
3
23
14
Blades
%
0
0
14
15
0
2
9
8
8
7
6
6
7
2
10
8
0
5
0
37
72
42
32
34
52
40
34
35
29
48
4
13
7
100
75
52
70
0
66
47
42
45
40
52
60
60
71
71
77
100
68
70
7
520
55
N
Bladelets
%
3
2
0
25
2
10
100
18
18
19
28
16
20
18
11
12
0
3
0
16
20
147
16
1
1
2
2
10
27
19
20
14
20
12
6
6
2
N
Points
%
1
1
0
0
32
5
0
14
25
27
14
33
18
10
18
0
10
8
0
5
10
172
18
14
1
8
38
27
10
28
18
7
10
4
5
N
1
1
3
1
3
2
5
1
2
19
Indeterminates
%
N
0
0
0
0
0
0
1
1
4
1
3
3
0
10
2
3
0
0
0
2
%
1
4
2
1
2
3
2
3
1
19
N
%
0
0
0
0
0
0
1
4
0
2
1
3
5
4
7
0
0
5
0
1
4
44
20
2
56
153
101
71
85
100
67
57
49
41
62
4
19
10
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
2
946
100
Refers here to unretouched artifacts and to the blank types of retouched items. Indeterminate blanks correspond to he majority of unifacially- and bifacially-retouched points, (see text for details).
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Elevation above
datum (cm)
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Fig. 5. Cores. 1 e Levallois core; 2, 5 e Fragmented and exploited cores (technology unclear); 3, 4 e Single Platform cores.
Fig. 6. Elongated pieces, blades and bladelets.
125
Author's personal copy
126
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
suggesting that elongated blanks were preferentially selected for
secondary modification.
The frequency of backed microliths, often used as diagnostic
markers of the LSA, increases in the upper levels of the site (Fig. 9)
compared to the lower Complex (n ¼ 1 in Layer IId). The fluctuating frequencies of retouched points do not show clear temporal
trend. On the whole there are no clear differences in the typotechnological composition throughout the section. MSA elements (Levallois products, large blades, retouched points) cooccur or are interstratified with bladelets and backed microliths
(Fig. 9).
size. Frequencies of small (<20 mm) bladelets increase along the
Lower Complex. This trend is observed for all the raw materials, but
most notably in obsidian. The proportion of obsidian bladelets increases to 65% of the unretouched débitage in the 130e140 level.
These products are associated with recurrent unipolar flaking, from
either a single débitage surface (Levallois sensu lato) or from semirotated cores.
Several crested blades/bladelets and overshot flakes and blades
(Fig. 6:1, Fig. 7:12) first appear in Layer IIc. The overshot pieces tend
to be laterally convex with uni- or bipolar dorsal face scars indicating the removal of series of parallel-sided bladelets, yet the
guide-ridges on the dorsal faces and the lateral edges of the blades
do not follow a clearly distinct flaking, reflecting lack of standardization of flaking methods. These observations are consistent with
the low frequencies of Upper Paleolithic/LSA-type cores, the flaking
of which is more standardized (see above).
Levallois products (n ¼ 40), recognizable by their shape (often
oval), thinness, butt preparation and the organization of dorsal face
scars, appear in higher frequencies in the lower part of the
sequence (Layers IId and IIe), but are also present in Layers IIaeIIc,
which are dated to the mid-Holocene. These elements consist
mainly of flakes (n ¼ 28), blades (n ¼ 2) and points (n ¼ 5; Fig. 7).
Levallois flakes bear mostly centripetal and perpendicular scar
patterns (on 20 flakes and 2 blades), some showing evidence of
unipolar (n ¼ 4) and bipolar (n ¼ 4) recurrent flaking methods
(Fig. 7:7e12). In addition, éclats débordants and outrepassants (i.e.,
intentional core maintenance elements) indicate the rejuvenation
of Levallois cores by the re-shaping of lateral and distal convexities
(Fig. 7:9, 12).
5. Faunal remains
Some 340 mammalian faunal remains were collected
throughout the sequence at Goda Buticha. Of these, ca. 10% are
dental samples, another 10% are long bone shaft fragments, and the
rest are unidentifiable bone fragments. Many of the remains are in
friable condition, with poor surface preservation, which prevented
the identification of anthropogenic or carnivore modifications.
Therefore, analysis of the currently available restricted sample
focused on taxonomic identifications of sufficiently preserved elements. Future studies of an augmented sample will focus on the
quantitative aspects of the faunal assemblage.
A diversity of taxa has been identified, including Procaviidae,
Lagomorpha, Equus, Phacochoerus, small carnivore, and several
types of bovids including bovini, antilopini, Redunca, and Tragelaphus. No domesticated fauna (e.g., sheep and goat) were
recovered from any stratigraphic level, including the Upper
Complex. The small sample of identifiable fauna from Goda
Buticha contributes to our understanding of the local paleoenvironmental record, as it contains some open woodland indicators and suggests ecotone conditions including a source of
permanent water that could support edaphic grasslands. The
majority of taxa listed above are represented by one or two
4.2.3. Retouched artifacts
The retouched pieces (n ¼ 141) constitute w5% or w14% of the
assemblage (inclusive and exclusive of debris, respectively). These
include 8 retouched cores and 10 micro-flakes. Obsidian was the
preferred raw material for retouch (Table 6; Fig. 8).
Table 6
Distribution of raw materials by retouched types.
Flint
Quartzite
Basalt
Obsidian
Other Volcanic
rocks
Total
Side-scraper
Notch &
denticulate
Backed
microlith
16
2
4
27
1
1
8
1
1
49 (34.8)a
4 (2.8)a
Point
End-scraper
Burin
Truncation
Flat retouch
Retouched
flakes/blades
13
5
1
5
19
8
1
3
2
2
6
15
21 (14.9)a
30 (21.3)a
1 (0.7)a
3 (2.1)a
2 (1.4)a
2 (1.4)a
24 (20.6)a
Total
N
%
38
4
15
83
1
26
3
8
62
1
141
100.0
2
c ¼ 50.8881, p-value ¼ 0.01827.
a
Absolute frequencies and percents (in parentheses) out of the total number of tools.
The frequent formal tool categories are: side-scrapers (n ¼ 50),
usually with fine retouch; unifacial (n ¼ 19) and bifacial (n ¼ 11)
points, with one or two modified edges; and backed microliths
(n ¼ 21), with typically curved backed edges. The rest of the tools
are retouched flakes/blades exhibiting continuous retouch of
various styles. Some 88% of the tools occur Layers IIc and IId, in
elevations of 60e140 cm above datum (Fig. 9).
Besides the backed microliths (see also Leplongeon, 2014),
which are always made on bladelets, elongated blanks (i.e., blades
or bladelets) were used in another 38 cases. These counts do not
include the blanks for the retouched points, which in most cases
could not be identified due to extensive retouch. These frequencies
differ from those observed in the unretouched débitage (Table 5),
dental samples, the reduncines being the only exception with a
slightly higher representation in the middle-lower sections. This
sample is too small to assess the subsistence activities of the
occupants of the site.
Nearly one hundred fragments of ostrich eggshell (some with a
maximum dimension of 4e5 cm) occur throughout the sediment
column. Notably, their frequencies decline with depth. Similarly,
micro-faunal remains, sometimes heavily burned, appear
throughout the sequence, yet their relative frequencies as well as
proportion of burned specimens decrease with depth (see above).
At 197e194 cm above datum microfauna remains were densely
concentrated in a distinctive lens. A detailed analysis of this sample
is currently underway and will be published elsewhere.
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
127
Fig. 7. Levallois points and flakes.
6. Hominin remains
The excavations in 2008 and 2011 at Buticha Cave yielded ten
hominin specimens, which derive from the Lower Complex (Fig. 3).
One specimen, a fragment of the cranial vault, derives from layer IId
and therefore dates to at least 35 ka cal BP. It may be as old as 42 ka,
given the date obtained on a charcoal sample from this elevation in
an adjacent grid square (see Section 3), whereas the other nine
pieces originated from the Holocene deposits.
The hominins are represented by five permanent teeth or tooth
fragments, two small fragments of cranial vault, a proximal pollical
phalanx, two proximal pedal phalanges, and a metacarpal diaphysis
(Fig. 10). In the descriptions that follow, we selected a variety of
comparative samples. The teeth are compared to two samples from
Nubia, the Epipaleolithic (Jebel Sahaba and Wadi Halfa) and agricultural samples published by Calcagno (1989). The Nubian dentitions show a marked reduction across the transition to agriculture.
Additional comparisons were made to Jacobson’s (1982) data on
Author's personal copy
128
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Fig. 8. Retouched artifacts: 1, 3 e bifacial points; 2 e unifacial point; 4 e retouched crested blade; 5 e end-scraper on a retouched flake; 6 e thinned side-scraper; 7 e denticulate.
the teeth of native South Africans (as an example of a generalized,
modern sub-Saharan sample) and reports from the literature on
dentition of MSA hominins (most of whom had large teeth relative
to living Africans). The postcranial remains are compared to data
from Pearson (1997), who measured samples of skeletons of recent
humans from around the world (including four African
populations).
Tooth 4 (BC-188): a lower left permanent canine recovered in
2011 from sediments from square B1d at 146 cm above datum
(Layer IIc). Crown dimensions: 5.9 mm bucco-lingual (BL) by
6.3 mm mesio-distal (MD). These dimensions are quite small BL and
fairly small MD compared to Calcagno’s (1989) Nubians.
Tooth 1: a left I1 recovered in 2008 from square A1a from sediments between 130 and 127 cm (Layer IIc) above datum. Crown
dimensions: 7.1 mm (BL) by 9.6 mm (MD). These correspond to a
fairly wide MD width but unremarkable BL length.
Proximal Pollical Fragment: A small, slender proximal left
pollical phalanx recovered in 2008 from sediments from square
A1a at 127 to 120 cm above datum (Layer IIc). Measurements:
maximum length: 26.3 mm; maximum breadth of the proximal
end: 10.2 mm. The specimen fits within many modern
populations (Pearson, 1997), but indicates a small (female?)
individual.
Metacarpal Diaphysis (BC-510): Metacarpal diaphysis, probably
a left metacarpal IV, excavated in 2011 from square A1b at 127 cm
above datum (Layer IIc). The fragment is 37.5 mm long; dimensions
at its approximate midshaft are 6.7 mm dorso-volar by 6.1 mm
medio-lateral.
Tooth 2: a left M3 recovered in 2008 from square B1a 123 cm
above datum (Layer IIC). Crown dimensions: 10.3 mm BL by 9.7 mm
MD. These dimensions are small compared to all of Calcagno’s
(1989) samples.
Tooth 5 (BC-541): Fragment of a left upper canine recovered in
2011 from sediments from square A1b at 121 cm above datum
(Layer IIc). The crown is too fragmentary to measure its BL or MD
dimensions.
Cranial Vault Fragment 2 (BC-394): Parietal fragment, recovered
in 2011 from sediments from square B1b 119 cm above datum
(Layer IIc). Maximum dimensions: 34.5 19.3 mm. The fragment is
thin, ranging between 5.4 and 5.1 mm, and bears traces of the
sagittal suture. These facts suggest it came from a late adolescent or
young adult.
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
129
Fig. 9. Diachronic change of retouched pieces.
Tooth 3 (BC-UN-25): a heavily worn right P4 recovered in 2011
from sediments from square B1d from 116 cm above datum (Layer
IIc). The crown measures 8.1 mm BL by 7.1 mm MD. Relative to
Calcagno’s (1989) samples of Nubians, these dimensions are
somewhat small MD and quite small BL.
Proximal Pedal Phalanx 1 (BC-500): A complete right proximal
pedal phalanx, likely of ray III, excavated in 2011 from square A1b
130 cm above datum (Layer IIc). The specimen’s maximum length is
23.7 mm.
Proximal Pedal Phalanx 2: A complete left proximal pedal
phalanx, likely from ray IV, recovered in 2012 from sediments
excavated and sieved in 2011 from square A1d at 120e130 cm
above datum (Layer IIc). The maximum length of the phalanx is
19.3 mm.
Fig. 10. Hominin remains from Goda Buticha A) Tooth 1 (unnumbered); B) Tooth 2 (unnumbered); C) Tooth 3 (BC-UN-25); D) Tooth 4 (BC-188); E) Tooth 5 (BC-541); F) Cranial vault
fragment 1 (unnumbered); G) Cranial vault fragment 2 (BC-394); H) Proximal pollical phalanx 1 (unnumbered); I) Proximal pedal phalanx 1 (BC-500) [proximal pedal phalanx 2 is
not shown]; J) Metacarpal shaft (BC-510). All the teeth are scaled to the size of the upper scale bar; the fragments of cranial vault and postcranial bones are scaled to the size of the
lower scale bar.
Author's personal copy
130
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Cranial Vault Fragment 1: Fragment of cranial vault lacking the
internal table of bone, recovered in 2008 from sieved bones and
lithics from sediments from square A1a 50 to 40 cm above datum
(Layer IId). The maximum dimensions of the specimen are
32.6 23.4 mm, with a maximum preserved thickness of 6.2 mm.
With one exception of the BL breadth of tooth 1 (the left I1), the
teeth are small in size compared to South Africans from the Dart
Collection (Jacobson, 1982) as well as agricultural or Epipaleolithic
Nubians (Calcagno, 1989). Teeth from MSA contexts at other sites in
Africa tend to be large relative to these comparative samples (Grine
et al., 1991, 2000; Rightmire and Deacon, 1991, 2001; Grine and
Henshilwood, 2002; White et al., 2003; Verna et al., 2013),
although a small number of notably small-toothed individuals are
also known (Singer and Wymer, 1982; Bräuer and Mehlman, 1988;
McCrossin, 1992; Lam et al., 1996; Royer et al., 2009). The postcranial specimens are average to small in size relative to those of
recent Africans. Hominin postcranial fragments from the MSA (for
which the most numerous sample comes from South Africa, followed now by Ethiopia (Pearson et al., 2008a, 2008b) display a
large amount of variation in size, from medium- to small-sized
individuals (e.g., the first metatarsal from Klasies River (Rightmire
et al., 2006) and the new pedal phalanges from the Howiesons
Poort or late MSA from Diepkloof (Verna et al., 2013)) to large ones
(e.g., the second metatarsal from Klasies River main site (Rightmire
et al., 2006) and the femur from Blind River (Wang et al., 2008). The
medium to small sizes of the postcranial remains from Buticha fit
within this broad spectrum of variability.
7. Beads
Five ostrich eggshell (OES) beads were found in the upper part of
Complex II (Layer IIc) (Fig. 11). Some beads are well-executed and
finished, with a perfectly circular perforation and perimeter, when
others are made from thicker eggshell, with roughly shaped edges
and irregular central perforations. The differences in workmanship
can be attributed to the beads being discarded during different
stage of manufacture (for steps of OES bead manufacturing, see
Orton, 2008; also Kandel and Conard 2005).
8. Discussion
8.1. Goda Buticha in its immediate geographic context
Dating the dense sequences in cave sites in eastern Africa has been
far from a straightforward matter. Goda Buticha is no exception.
Nine stratigraphically consistent radiocarbon dates from Layer IIc,
which contains the richest archaeological remains, place this occupation within a short time range between 6940 40 14C BP
(7797 118 cal BP) and 5590 50 14C BP (6380 88 cal BP). The
archaeological material from Layer IIc has a clear LSA aspect, but includes also some MSA elements. The two younger dates from the top
of Layer IId indicate human presence in the cave around 34e35,000
years ago (Table 1). These dates represent a terminal age for the
associated MSA cultural finds, while dates from Layer IIf indicate MSA
occurrences around 42e46,000 ka cal BP. Since MSA deposits extend
down to bedrock (7 cm below datum, i.e., w80 cm below the dated
level), earlier dates may still be obtained from the basal section.
The taxonomic composition of large mammals at Goda Buticha
suggests a relatively closed and wet habitat around the cave in the
late Upper Pleistocene. Within the small faunal sample, Redunca is
the most common taxon throughout the sequence. Independent of
the type of transporting agent (i.e., human or carnivore), this suggests the presence in the cave’s vicinity of a perennial river (as
indeed is the case today) or any other permanent source of water
that can support moist edaphic grasslands. The presence of Equus
(mostly teeth) and Tragelaphus in different levels suggests a nearby
presence of an open woodland environment. Together, these lines
of evidence point to ecotone conditions that are preferable to
certain large mammals, including large bovines.
Our understanding of environmental conditions during the occupations of Goda Buticha stands to benefit from speleothem
studies in nearby cave sites (Assefa et al., 2014). These emerging
paleoclimatic archives are also useful for understanding the complex paleoclimatic record in eastern Africa (e.g., Anyah and Semazzi,
2007; Blome et al., 2012). In the larger Horn of Africa region, the
time around w50 ka marks a major wet event that continued until
31 ka (Gasse and Street, 1978; Gasse et al., 1980), followed by
generally dry conditions which persisted till the rapid onset of the
African Humid Period at about 11.6 ka cal BP (Foerster et al., 2012).
Humid conditions persisted during the early Holocene until they
gradually changed to dry conditions starting at about 5 ka cal BP.
Ongoing investigations on the speleothems from the Mechara area
(w170 km SW of Goda Buticha) are consistent with this trend,
showing warm and wet conditions in the SE Ethiopian region
during the Early to Middle Holocene followed by dry conditions
during the Late Holocene (Asrat et al., 2007, 2008; Baker et al.,
2010). Wetter conditions inferred from the sedimentological and
faunal records of Goda Buticha are generally consistent with such
climatic shifts in the region.
Goda Buticha was located in an ecological ecotone on the lower
flanks of the escarpment, at the relatively low elevation of
1382 m asl. This raises the hypothesis that the exploitation of lowelevation ecotones in the Late MSA was made possible through the
inferred climatic amelioration represented in the site’s record,
similar to the argument presented by Ambrose (1986) regarding
Eburran sites during the Holocene.
Within the site itself, the contact between Complex I and
Complex II (w180 cm above datum; Fig. 3) represents a sedimentological boundary separating anthropogenic ashes from geogenic
silts. At 170e160 cm above datum, a horizon of brown silt contained LSA archaeological material together with MSA components.
The main raw material represented is obsidian, which was used to
make small tools, including microliths with fine and backed
retouch (Leplongeon, 2014). Obsidian cores are correspondingly
small. At least some lithic reduction was carried out on site,
as evidenced by pieces of core trimming elements (e.g., éclats
outrépassants) indicative of core rejuvenation. Notably, in the upper
part of Complex II (150e110 above datum) small bladelet cores,
made on chert, begin to appear, associated with typical unifacial
and bifacial MSA points.
The sedimentary fine-grained matrix of Complex II is relatively
homogeneous and was deposited through a single depositional
regime, yet the radiocarbon dates point to a chronological hiatus
between Upper Pleistocene (IId) and mid-Holocene (IIc) deposition
(Fig. 3, Table 1). This hiatus, expressed granulometrically by a
higher frequency of large clasts in the matrix of Layer IIc, is
confirmed by preliminary results of ongoing OSL dating (Tribolo,
personal communication).
The cultural material in the lower part of Complex II (IId) is
dominated by MSA lithic artifacts with few LSA components. At
w70e60 cm above datum and down to bedrock, the small
component disappears almost entirely, obsidian becomes rare, and
the artifacts tend to be large (4e5 cm), clearly Levallois and almost
without retouch.
Thus the main tendencies observed in the lithic assemblages
throughout the Goda Buticha sequence can be summarized as
follows:
—Artifacts in the sequence of Goda Buticha are generally small,
and items longer than 60 mm are absent. The artifacts
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
131
Fig. 11. Three beads from the top of the Lower Complex (Layer IIc). The upper and lower rows show the outer and inner surfaces of the beads, respectively.
manufactured on basalt and quartzite, which are available
locally as large nodules/cobbles, are the largest ones in the
assemblage.
—The frequency of obsidian generally declines with depth,
although some types of MSA points are made on this raw
material.
—Obsidian cores appear to have been more intensively exploited
(average length of ca. 21 mm compared to 40 mm for chert
cores). Additionally, obsidian was the preferred raw material for
retouch (62% of the retouched artifacts are made on obsidian, vs.
only 30% of the total number of artifacts made of this raw material). These two traits are consistent with possible longdistance transport of obsidian as a preferred raw material (see
above).
Ostrich eggshell (OES) beads from Goda Buticha can now be
added to the finds known from only a handful of sites in eastern
Africa. Such beads constitute an important element in the context
of identifying cultural changes in the Late Pleistocene in the Horn of
Africa. At the rock shelter of Enkapune Ya Muto in Kenya OES beads
(Ambrose, 1998b: fig. 3) were dated directly or by associated
charcoal samples to w40 ka. At Kisese II rockshelter (Tanzania), OES
beads are associated with a transitional MSA/LSA industry (Inskeep,
1962) and were directly dated to 31,480 BP (Deacon, 1966). In
Mumba Cave (Tanzania) OES beads have been assigned to the LSA
with dates of w49 ka (Mehlman, 1989, 1991; Diez-Martín et al.,
2009; Gliganic et al., 2012). Direct AMS radiocarbon dating of
these beads has yielded multiple and slightly younger dates, between 29 ka and 33 ka (Conard 2004, Weiß, 2000, both cited in
Conrad 2005). In Goda Buticha OES beads derive from sediments
with a Holocene age, yet they are associated with both LSA artifacts
as well as stone tools with MSA affinities.
Two alternative hypotheses are considered to explain the
change in the relative proportions of MSA and LSA material
throughout the Goda Buticha sequence. One explanation implicates
natural formation processes and post-depositional mixture of
sediments and of MSA and LSA elements, thus creating a ‘phantom’
of a transitional cultural phase. However, no major disturbances
were observed macroscopically. We remain careful regarding the
occurrence of MSA features in Layer IIc, waiting for formal sedimentological and chronological analyses that will speak to
stratigraphic integrity and the potential effects of post-depositional
processes (including artifact migration through the sequence).
Nevertheless, our macroscopic observations, as well as the radiocarbon chronology, lead us to suspect that such effects, if existent,
have been very localized and that the archaeological elements of
Layer IIc (as well as the others) derive from an in situ context.
An alternative hypothesis is that the sequence of Goda Buticha
represents a facet of the material culture variability that is inherent
to the late MSA and the LSA in the southeastern Ethiopian region.
This hypothesis cannot be evaluated based on the record of a single
site. Porc Epic Cave, where a late MSA cultural sequence has been
described in detail (Clark and Williamson, 1984; Pleurdeau, 2003,
2004, 2005a, 2005b; Assefa, 2006; Assefa, 2006; Assefa et al.,
2008), provides an opportunity to compare the two cultural sequences on a regional scale, as only 30 km separate the two sites.
In both sites the same raw materials were used. Obsidian was
used more frequently in Goda Buticha (30% vs. 8% in Porc Epic), and
basalt is more common in Porc Epic (17% vs. only 9% in Goda
Buticha) even though it can be found overlying the sedimentary
succession in the region. Such differences suggest different raw
material-specific strategies of transport and use within and between localities. In both sites ‘provisioning of place’ (as defined by
Kuhn, 1995) seems to have taken place, with tested blocks, nodules
or partly decorticated cores imported into the sites, where they
were further reduced (as attested by the various types of core
trimming elements) to near-exhaustion (average core size in Porc
Epic is 40 mm vs. <30 mm in Goda Buticha). Obsidian was preferentially used for the production of bladelets at Porc Epic, while at
Buticha it was the preferred raw material for all types of retouched
artifacts (Table 6).
Technologically, both sites demonstrate the presence of the
same four technological schemes in similar proportion. These
include Levallois flaking methods (centripetal for preferential
flakes and unipolar recurrent), and very low frequencies of semirotated and full blade technologies. Throughout the Porc Epic
MSA sequence there are common technological (Levallois and
discoid schemes) and typological (e.g., retouched points) features.
With the exception of the lower unit at the site, these are associated
with elements which are argued to be “typical” LSA characteristics
such as backed bladelets, without any significant directional
changes throughout the sequence. In the absence of evidence for
Author's personal copy
132
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Fig. 12. Location map of MSA and LSA sites in the Horn of Africa mentioned in the text. (image modified from NASA Visible Earth)
significant bioturbation or mixing it was argued that MSA and LSA
characteristics coexisted in the late MSA assemblages of Porc Epic
(Pleurdeau 2005b; but see Leplongeon, 2013, 2014). The relatively
small sample size that is currently available for Goda Buticha precludes a robust identification of temporal trends throughout the
sequence. Nevertheless, the densest part of the sequence seems to
show a similar picture. A change was recorded only at the top of
Complex II, dated to the Holocene, where retouched points disappear while microliths and backed bladelets persist.
Both Porc Epic and Goda Buticha date to the late Upper Pleistocene. Age estimates for Porc Epic are controversial. They range
from MIS 4 (61,202 958, 61,640 1083, and 77,565 1575 a)
using obsidian hydration (Michels and Marean, 1984) to radiocarbon determinations >43,200 33,700 00 BP and
35,600 350 BP on opercula from various levels of the site (Assefa
et al., 2008). The latter dates are comparable to those from unit IId
of Goda Buticha. However, direct gamma-spectrometry dating of
the hominin mandible, discovered in 1933, yielded a date of w50 ka
(Y. Yokoyama and C. Falgueres, pers. comm.).
The local record represented by these two sites may be linked on
the basis of its microlithic and laminar products to comparable sites
at the regional scale in the Horn of Africa.
8.2. A broader perspective from the Horn of Africa
Only a handful of Upper Pleistoceneeearly Holocene sequences
in the Horn of Africa provide relevant regional and chronological
comparisons to the sequence of Goda Buticha (Fig. 12). Among
these occurrences are the open-air sites of Aladi Springs (Clark and
Williams, 1978; Gossa et al., 2012), K’one (Kurashina, 1978) and
Bulbula in Ziway-Shala area (Ménard et al., submitted) in Ethiopia,
as well as MSA and LSA sites from Buri Peninsula in Eritrea (Beyin,
2011a, 2013); the cave of Midhishi 2 (Brandt and Brook, 1984;
Gresham, 1984) and the rock shelter of Laas Geel 7, both in Somaliland (Diaz, 2009, and references therein; Gütherz et al.,
submitted); and the cave site of Mochena Borago (Brandt et al.,
2012, and references therein) in Ethiopia. With the exception of
Mochena Borago, chronological comparisons of these sequences
with Goda Buticha are tentative given that the relevant radiocarbon
dates were obtained prior to the 1980s, or due to excavators’ concerns regarding the stratigraphic integrity of the sites.
The lithic assemblages of these sites have been assigned to
different archaeological taxonomies. K’one is perceived as an MSA
assemblage, but the (potentially contemporaneous) sequence of
Mochena Borago, dated 54e43 ka, was not assigned to either a
MSA, LSA or a transitional entity due to the absence of marked
changes throughout the sequence (Brandt et al., 2012). Aladi
Springs and Midhishi 2 (stratigraphic unit CSUb) are believed to be
transitional MSA/LSA assemblages, whereas the lower complex at
Laas Geel 7 is assigned to the Hargeisian (as defined by Clark, 1954;
Gütherz et al., submitted). The sites of Lake Besaka which contain
the “Ethiopian Blade Tool Tradition” (Brandt, 1982, 1986) post-date
19,000 BP, as radiocarbon measurements on OES (22,080 305,
19,460 205, and 19,280 215) derive from a culturally sterile
layer underlying the archaeological material (Brandt, 1986:63).
The high degree of variability within these lithic assemblages
clearly has contributed to taxonomic confusion or even contradictions. It is important therefore to try and parse the sources of such
variability. While a detailed discussion of each assemblage is unwarranted here, some chronological (if the dates are accepted) as
well as typo-technological patterns seem to emerge in the context
of this broader overview.
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Typo-technologically, all the sites exhibit the use of Levallois
flaking methods in Pleistocene horizons, combined with a continuous presence of retouched points, particularly in K’one (Kurashina,
1978) and Midhishi 2 (Brandt and Brook, 1984). Midhishi 2 is
exceptional among these sites in that microliths and backed bladelets are not present in the older Pleistocene-age deposits and
only occur in CSUb (Brandt and Brook, 1984). Furthermore, Gossa
et al. (2012) note a diachronic trend of artifact size diminution in
Aladi Springs, similar to Goda Buticha. In one case, at Gobdera, were
typical (though not exclusive) MSA elements (centripetal cores and
facetted platforms) recognized in an assemblage defined as LSA
(Pillipson, 1977).
The largest and best described assemblage from K’one (K’one 5),
200 km west from Goda Buticha, provided a single radiocarbon date
of 14,670 200 BP, but Kurashina (1978:359) suggests an age between 40 and 30 ka based on stratigraphic extrapolation with
studied geological sections. This assemblage is not compared easily
with any of the others mentioned above, since it contains only 0.7%
retouched tools, the rest being debitage and cores (Kurashina,
1978:323). Kurashina identified the use of Levallois flaking systems, including an important presence of the type I Nubian technology that has recently been promoted as a cultural marker of Late
Pleistocene out of Africa dispersals (Beyin, 2013; Van Peer, 1998;
Rose et al., 2011; Crassard and Hilbert, 2013). There are no indications for microlithic elements in this assemblage.
Further away in the Horn of Africa, the best dated and reported
site is the cave of Mochena Borago (Brandt et al., 2012). Several
stratigraphic sequences were described and dated independently
due to the complex site stratigraphy. These indicate occupations
dated to 53,224 2662 cal BP (lower T-group), 45,164 982 cal BP,
43,480 443 cal BP, and 41,159 783 cal BP (upper T-group, Sgroup and R-group, respectively). The overlaying occupational
episode dates to the early Holocene (7589 689 cal BP). Thus
Mochena Borago, despite its distance from the eastern part of the
Horn, depicts the same temporal gap in human occupation.
Brandt et al. (2012) note the close typo-technological similarities across the T- and S-group assemblages. Flaking systems
consist of various Levallois methods inclusive of the Nubian
technology in the upper T-group assemblage. The majority of
cores belonged to the group dubbed “SDM” (single-double-multiple platform cores), but their frequencies decline in the S-group
assemblage concurrent with the emergence of bipolar technology.
The latter is comparable to the broadly contemporaneous
assemblage from layer V at Mumba Cave in Tanzania (Diaz-Martín
et al., 2009; Gilganic et al., 2012; Eren et al., 2013). Frequencies of
backed pieces increase through time since their first appearance
in upper T-group assemblages at the expense of retouched points,
resulting in an increase through time in richness and evenness of
the typological classes of retouched artifacts (including burins,
drills/awls, notches, non-standardized unifacial and fewer bifacial
points).
It appears that part of the variability of the lithic assemblages
discussed here derives from situational and specific aspects of the
assemblages, related to raw material availability, mobility patterns
and site functions. The prolific use of obsidian in K’one, Aladi
Springs and Mochena Borago reflects their proximity to the sources
of raw material, but assemblage compositions (i.e., frequencies of
debitage, cores, CTE and retouched items) differ according to site
functions. Whereas K’one may be an obsidian knapping station,
Mochena Borago displays the full range of lithic reduction and
secondary modification. At Porc Epic, Goda Buticha and Laas Geel,
on the other hand, obsidian is one of several raw materials, and is
used preferentially for some types of artifacts. Site provisioning
strategies also appear to differ according to the distances of a given
site from the various raw material sources. Specifically, Porc Epic
133
and Goda Buticha reflect site provisioning and on-site knapping
independently of specific raw materials.
Other aspects of the lithic assemblages seem to present broad
similarities, among which is the co-occurrence of Levallois with
semi-rotated and single platform cores early on in MIS 3 assemblages and through MIS 2, as well as the general trend (rather than
sudden appearance) of microlithization and the occurrence of
backed items in tandem with MSA points. The early dates assigned
recently to OES beads from Enkapune Ya Moto and Mumba, which
led to the definition of these assemblages as LSA, further confound
the issue. The shift from the Middle to the Later Stone Age remains
a poorly recognized process both in terms of its temporal extent
and of its telltale material culture markers.
9. Concluding remarks
In the context of the paleoenvironmental records from the Horn
of Africa, the faunal remains from Goda Buticha suggest that the
occupations of the site took place during relatively wetter periods.
The occupational hiatus between 35 ka and 8e6 ka corresponds to a
dry period in this region (Umer et al., 2004). A similar occupational
gap seems to have occurred in the majority of sites from this time
span, without evidence for human presence between w40 ka and
the late/terminal Upper Pleistocene (18,790 340 for CSUb in
Midhishi 2, 11,079 160 BP in Aladi Springs), or early Holocene (the
upper complex at the rock shelter 7 of Laas Geel (Gütherz et al.,
submitted) as well as Mochena Borago). The sequence from Goda
Buticha presents a rare case in which MSA elements are associated
with LSA materials dated to the Holocene. Based on current data,
site formation disturbances in Goda Buticha seem to have been
localized at most. Therefore, the pattern observed in the site’s
sequence needs to be considered when interpreting the shift from
the MSA to LSA in the Horn of Africa.
The limitations of the late Pleistocene record in the Horn of
Africa and dating ambiguities of the available sites must be borne in
mind. Having said that, we suggest that in view of the occupational
gap that appears to temporally disjoint LSA and MSA assemblages
in stratified sites, material culture similarities across this chronological gap may not represent homologies derived from a MIS 3
common ancestry or from different regional ‘anestral’ sites. Instead,
the potential effect of historical and demographic processes of
movement within, out of, and into Africa should be investigated as
possible causes for the general similarities and regional differences
during and after “The Dispersal” of modern humans from Africa.
Within this framework, Goda Buticha is a key site for reevaluating
the dynamics and tempo of the change from MSA to LSA lifeways in
this region.
Acknowledgments
We thank the Ethiopian Authority for Research and Conservation of Cultural Heritages (ARCCH) for permission to survey in the
area and to excavate in Goda Buticha. This research was supported
by grants to ZA from the National Geographic Society (grants #
8110-06 and 8510-08) and to ZA and DP from the Wenner-Gren
Foundation (Grant # ICRG e 102). We would like to acknowledge
the efforts of Tilahun G/Selassie, Workalemahu Bekele, Hadis,
DeJene Dendana Gulti, Alice, as well as the local people living near
the site for their assistance with the excavation. We are also very
thankful to Cécile Chapon, Simon Puaud, Marion Hernandez and
Chantal Tribolo, for their current support for sedimentological and
OSL dating analyses, and to Sébastien Nomade for providing some
radiocarbon analysis facilities. We are grateful to the French Center
for Ethiopian Studies for providing logistical support and for
funding part of the post-excavation analyses. We thank the Tourism
Author's personal copy
134
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
and Culture Offices of the Eastern Harerghe, Western Harerghe, the
Dire Dawa Administration, and Harari National Regional State for
fieldwork administrative support. We are grateful to the reviewers
of the paper for their very helpful suggestions and to Christian
Tryon for discussions and for directing us to some important
bibliographic references.
References
Ambrose, S.H., 1986. Hunteregatherers adaptations in non-marginal environments:
an ecological and archaeological assessment of the Dorobo model. Sprache und
Geschichte in Afrika. Herausgegeben an den Universitäten Köln and Bayreuth 7
(2), 11e42.
Ambrose, S.H., 1998a. Late Pleistocene human population bottlenecks, volcanic
winter, and differentiation of modern humans. Journal of Human Evolution 34,
623e651.
Ambrose, S.H., 1998b. Chronology of the Later Stone Age and food production in
East Africa. Journal of Archaeological Science 25, 377e392.
Anyah, R.O., Semazzi, F.H., 2007. Variability of East African rainfall based on
multiyear RegCM3 simulations. International Journal of Climatology 27, 357e
371.
Asrat, A., Baker, A., Leng, M., Gunn, J., Umer, M., 2008. Environmental monitoring in
the Mechara Caves, Southeastern Ethiopia: implications for speleothem paleoclimate studies. International Journal of Speleology 37, 207e220.
Asrat, A., Baker, A., Mohammed, M.U., Leng, M.J., Calsteren, P.V., Smith, C., 2007.
A high-resolution multi-proxy stalagmite record from Mechara, Southeastern
Ethiopia: palaeohydrological implications for speleothem palaeoclimate
reconstruction. Journal of Quaternary Science 22, 53e63.
Assefa, Z., 2006. Faunal remains from Porc-Epic: paleoecological and zooarchaeological investigations from a Middle Stone Age site in southeastern Ethiopia.
Journal of Human Evolution 51, 50e75.
Assefa, Z., Pleurdeau, D., Hovers, E., Asrat, A., Pearson, O.M., Duquesnoy, F.,
Lam, Y.M., Tsion, C., 2014. Survey and explorations of caves in southeastern
Ethiopia: MSA and LSA archaeology and rock art. Quaternary International.
http://dx.doi.org/10.1016/j.quaint.2013.07.132.
Assefa, Z., Yirga, S., Reed, K.E., 2008. The large-mammal fauna from the Kibish
Formation. Journal of Human Evolution 55, 501e512.
Baker, A., Asrat, A., Fairchild, I.J., Leng, M.J., Thomas, L., Widmann, M., Jex, C.N.,
Dong, B., Van Calsteren, P., Bryant, C., 2010. Decadal-scale rainfall variability in
Ethiopia recorded in an annually laminated, Holocene-age, stalagmite. The
Holocene 20, 827e836.
Behar, D.M., Villems, R., Soodyall, H., Blue-Smith, J., Pereira, L., Metspalu, E.,
Scozzari, R., Makkan, H., Tzur, S., Comas, D., 2008. The dawn of human matrilineal diversity. The American Journal of Human Genetics 82, 1130e1140.
Beyin, A., 2006. The Bab al Mandab vs the Nile-Levant: an appraisal of the two
dispersal routes for early modern humans out of Africa. African Archaeological
Review 23, 5e30.
Beyin, A., 2011a. Earlyemiddle Holocene human adaptations on the Buri peninsula
and Gulf of Zula, coastal lowland of Eritrea. Azania: Archaeological Research in
Africa 46 (2), 123e140.
Beyin, A., 2011b. Upper Pleistocene human dispersals out of Africa: a review of the
current state of the debate. International Journal of Evolutionary Biology.
Beyin, A., 2013. A surface Middle Stone Age assemblage from the Red Sea coast of
Eritrea: implications for Upper Pleistocene human dispersals out of Africa.
Quaternary International 300, 195e212.
Blome, M.W., Cohen, A.S., Tryon, C.A., Brooks, A.S., Russell, J., 2012. The environmental context for the origins of modern human diversity: a synthesis of
regional variability in African climate 150,000e30,000 years ago. Journal of
Human Evolution 62, 563e592.
Bosellini, A., Russo, A., Assefa, G., 2001. The Mesozoic succession of Dire Dawa, Harar
Province, Ethiopia. Journal of African Earth Sciences 32, 403e417.
Brandt, S.A., 1982. A Late Quaternary Cultural and Environmental Sequence from
Lake Besaka, Southern Afar, Ethiopia (Ph.D. thesis). University of California,
Berkeley.
Brandt, S.A., 1986. The Upper Pleistocene and early Holocene prehistory of the Horn
of Africa. African Archaeological Review 4, 41e82.
Brandt, S.A., Fisher, E.C., Hildebrand, E.A., Vogelsang, R., Ambrose, S.H., Lesur, J.P.,
Wang, H., 2012. Early MIS 3 occupation of Mochena Borago Rockshelter,
Southwest Ethiopian highlands: implications for Late Pleistocene archaeology,
paleoenvironments and modern human dispersals. Quaternary International
274, 38e54.
Brandt, S.A., Brook, G., 1984. Archaeological and paleoenvironmental research in
northern Somalia. Current Anthropology 25 (1), 119e121.
Bräuer, G., Mehlman, M.J., 1988. Hominid molars from a Middle Stone Age level at
the Mumba rock shelter, Tanzania. American Journal of Physical Anthropology
75, 69e76.
Breuil, H., 1934. Peintures rupestres préhistoriques du Harrar (Abyssinie). L’Anthropologie 44, 473e483.
Breuil, H., Teilhard de Chardin, P., Wernert, P., 1951. Le paléolithique du Harrar.
L’Anthropologie 55, 219e228.
Bruggemann, J.H., Buffler, R.T., Guillaume, M.M., Walter, R.C., Von Cosel, R.,
Ghebretensae, B.N., Berhe, S.M., 2004. Stratigraphy, palaeoenvironments and
model for the deposition of the Abdur Reef Limestone: context for an important
archaeological site from the last interglacial on the Red Sea coast of Eritrea.
Palaeogeography, Palaeoclimatology, Palaeoecology 203, 179e206.
Calcagno, J.M., 1989. Mechanisms of Human Dental Reduction: a Case Study from
Post-Pleistocene Nubia. Department of Anthropology, Univ of Kansas.
Clark, J.D., 1954. The Prehistoric Cultures of the Horn of Africa. An Analysis of the
Stone Age Culture and Climate Succession in the Somalilands and Eastern Parts
of Abyssinia. Cambridge University Press, Cambridge.
Clark, J.D., 1988. The Middle Stone Age of East Africa and the beginnings of regional
identity. Journal of World Prehistory 2, 235e305.
Clark, J.D., Beyene, Y., WoldeGabriel, G., Hart, W.K., Renne, P.R., Gilbert, H.,
Defleur, A., Suwa, G., Katoh, S., Ludwig, K.R., 2003. Stratigraphic, chronological
and behavioural contexts of Pleistocene Homo sapiens from Middle Awash,
Ethiopia. Nature 423, 747e752.
Clark, J.D., Williams, M.A.J., 1978. Recent archaeological research in Southeastern
Ethiopia. 1974e1975. Annales d’Ethiopie, 19e44.
Clark, J.D., Williamson, K.D., 1984. A Middle Stone Age occupation site at Porc
Epic Cave, Dire Dawa (east-central Ethiopia). African Archaeological Review 2,
37e71.
Conard, N., 2005. An overview of the patterns of behavioural change in Africa and
Eurasia during the Middle and Late Pleistocene. In: Backwell, L., d’Errico, F.
(Eds.), From Tools to Symbols. Witwatersrand University Press, Johannesburg,
pp. 294e332.
Crassard, R., Hilbert, Y.H., 2013. A Nubian complex site from Central Arabia: implications for Levallois taxonomy and human dispersals during the Upper
Pleistocene. PLoS ONE 8 (7), e69221.
Deacon, J., 1966. An annotated list of radiocarbon dates for sub-Saharan Africa.
Annals of the Cape Provincial Museum 5, 5e84.
Delagnes, A., 1995. Variability Within Uniformity: Three Levels of Variability Within
Levallois System. The Definition and Interpretation of Levallois Technology,
pp. 201e212.
Delagnes, A., Tribolo, C., Bertran, P., Brenet, M., Crassard, R., Jaubert, J., Khalidi, L.,
Mercier, N., Nomade, S., Peigné, S., 2012. Inland human settlement in southern
Arabia 55,000 years ago. New evidence from the Wadi Surdud Middle Paleolithic site complex, western Yemen. Journal of Human Evolution 63, 452e474.
Diez-Martín, F., Domínguez-Rodrigo, M., Sánchez, P., Mabulla, A.Z.P., Tarriño, A.,
Barba, R., Prendergast, M.E., Luque, L., 2009. The Middle to Later Stone Age
technological transition in East Africa: new data from Mumba Rockshelter Bed
V (Tanzania) and their implications for the origin of modern human behavior.
Journal of African Archaeology 7, 147e173.
Eren, M.I., Diez-Martin, F., Dominguez-Rodrigo, M., 2013. An empirical test of the
relative frequency of bipolar reduction in Beds VI, V, and III at Mumba Rockshelter, Tanzania: implications for the East African Middle to Late Stone Age
transition. Journal of Archaeological Science 40, 248e256.
Foerster, V., Junginger, A., Langkamp, O., Gebru, T., Asrat, A., Umer, M., Lamb, H.,
Wennrich, V., Rethemeyer, J., Nowaczyk, N., 2012. Climatic change recorded in
the sediments of the Chew Bahir basin, southern Ethiopia, during the last
45,000 years. Quaternary International 274, 25e37.
Forster, P., Matsumura, S., 2005. Did early humans go north or south? Science 308,
965e966.
Gasse, E., Street, F.A., 1978. Late Quaternary lake-level fluctuations and environments of the northern Rift Valley and Afar region (Ethiopia and Djibouti).
Palaeogeography, Palaeoclimatology, Palaeoecology 24, 279e325.
Gasse, F., Rognon, P., Street, F.A., 1980. Quaternary history of the Afar and Ethiopian
Rift lakes. In: Williams, M.A.J., Faure, H. (Eds.), The Sahara and the Nile,
pp. 361e400.
Gliganic, L.A., Jacobs, Z., Roberts, R.G., Domínguez-Rodrigo, M., Mabulla, A.Z., 2012.
New ages for Middle and Later Stone Age deposits at Mumba rockshelter,
Tanzania: optically stimulated luminescence dating of quartz and feldspar
grains. Journal of Human Evolution 62, 533e547.
Gossa, T., Shale, Y., Negash, A., 2012. A reassessement of the Middle and Later Stone
Age lithic assemblage from Aladi Springs, southern Afar Rift, Ethiopia. Azania 47
(2), 210e222.
Gresham, T., 1984. An Investigation of an Upper Pleistocene Archaeological Site in
Northern Somalia. Unpublished M.A. thesis. The University of Georgia, Athens, GA.
Grine, F.E., Henshilwood, C.S., 2002. Additional human remains from Blombos Cave,
South Africa:(1999e2000 excavations). Journal of Human Evolution 42, 293e
302.
Grine, F.E., Henshilwood, C.S., Sealy, J.C., 2000. Human remains from Blombos Cave,
South Africa:(1997e1998 excavations). Journal of Human Evolution 38, 755e
765.
Grine, F.E., Klein, R.G., Volman, T.P., 1991. Dating, archaeology and human fossils
from the Middle Stone Age levels of Die Kelders, South Africa. Journal of Human
Evolution 21, 363e395.
Gutherz, X., Diaz, A., Ménard, C., Bon, F., Douze, K., Léa, V., submitted. The Hargeisan
revisited: lithic industries from shelter 7 of Laas Geel, Somaliland and the
transition between the Middle and Late Stone Age in the Horn of Africa. Quaternary International.
Hovers, E., 2007. The many faces of cores-on-flakes: a perspective from the
Levantine Mousterian. In: McPherron, S.P. (Ed.), Cores or Tools? Alternative
Approaches to Stone Tool Analysis, pp. 42e74.
Inskeep, R., 1962. The age of the Kondoa rock paintings in light of recent excavations
at Kisese II rock shelter. In: Mortelmans, G., Nenquin, J. (Eds.), Actes du IVe
Congres Panafricain de Préhistoire de l’étude du Quaternaire, Sciences
Humaines, 40, pp. 249e256. Tervuren.
Author's personal copy
D. Pleurdeau et al. / Quaternary International 343 (2014) 117e135
Jacobson, A., 1982. The Dentition of the South African Negro. Higginbotham Inc,
Alabama.
Kandel, A.W., Conard, N.J., 2005. Production sequences of ostrich eggshell beads and
settlement dynamics in the Geelbek Dunes of the Western Cape, South Africa.
Journal of Archaeological Science 32 (12), 1711e1721.
Klein, R.G., 1995. Anatomy, behavior, and modern human origins. Journal of World
Prehistory 9, 167e198.
Klein, R.G., 2008. Out of Africa and the evolution of human behavior. Evolutionary
Anthropology: Issues, News, and Reviews 17, 267e281.
Kuhn, S.L., 1995. Mousterian Lithic Technology: an Ecological Perspective. Princeton
University Press Princeton.
Kurashina, J., 1978. An Examination of Prehistoric Lithic Technology in East-Central
Africa, Ph. D. University MIcrofilm International, Ann Arbor.
Kusimba, S.B., 2001. The Early Later Stone Age in East Africa: xcavations and
lithic assemblages from Lukenya Hill. The African Archaeological Review 18,
77e123.
Lahr, M.M., Foley, R.A., 1998. Towards a theory of modern human origins: geography, demography, and diversity in recent human evolution. Yearbook of Physical Anthropology 41, 137e176.
Lam, Y.M., Pearson, O.M., Smith, C.M., 1996. Chin morphology and sexual dimorphism in the fossil hominid mandible sample from Klasies River Mouth.
American Journal of Physical Anthropology 100, 545e557.
Leplongeon, A., 2013. La Transition Middle Stone Age/Later Stone Age en Afrique de l’Est
(Éthiopie). Unpublished Ph.D. thesis. Museum national d’histoire naturelle, Paris.
Leplongeon, A., 2014. Microliths and the Middle to Later Stone Age transition: new
data from Porc-Epic and Goda Buticha Cave Sites, Ethiopia. Quaternary International. http://dx.doi.org/10.1016/j.quaint.2013.12.002.
Ménard, C., Bon, F., Dessie, A., Bruxelles, L., Douze, K., Fauvelle-Aymar, F.X., Khaldi, L.,
Lesur, J. and Mensan, R., submitted. Late Stone Age variability in the Main
Ethiopian Rift: new data from the Bulbula river, Ziway-Shala basin. Quaternary
International.
Marean, C.W., Assefa, Z., 2005. The Middle and Upper Pleistocene African record for
the biological and behavioral origins of modern humans. African Archaeology,
93e129.
McBrearty, S., Brooks, A.S., 2000. The revolution that wasn’t: a new interpretation of
the origin of modern human behavior. Journal of Human Evolution 39, 453e563.
McCrossin, M.L., 1992. Human molars from Later Pleistocene deposits of Witkrans
Cave, Gaap Escarpment, Kalahari margin. Human Evolution 7, 1e10.
McDougall, I., Brown, F.H., Fleagle, J.G., 2005. Stratigraphic placement and age of
modern humans from Kibish, Ethiopia. Nature 433, 733e736.
Mehlman, M.J., 1989. Later Quaternary Archaeological Sequences in Northern
Tanzania. University of Illinois.
Mehlman, M.J., 1991. Context for the emergence of modern man in eastern Africa:
some new Tanzanian evidence. In: Cultural Beginnings: Approaches to Understanding Early Hominid Lifeways in the African Savanna, pp. 177e196.
Merrick, H.V., 1975. Change in Later Pleistocene Lithic Industries in Eastern Africa.
University of California, Berkeley.
Michels, J.W., Marean, C.A., 1984. A Middle Stone Age occupation site at Porc-Epic
Cave, Dire Dawa (east-central Ethiopia), Part II. The African Archaeological
Review 2, 37e71.
Negash, A., Shackley, M.S., 2006. Geochemical provenance of obsidian artefacts
from the MSA site of porc-Epic, Ethiopia. Archaeometry 48, 1e12.
Negash, A., Brown, F., Nash, B., 2011. Varieties and sources of artefactural obsidian in
the Middle Stone Age of the Middle Awash, Ethiopia. Archaeometry 53 (4),
661e673.
Orton, J., 2008. Later Stone Age ostrich eggshell bead manufacture in the Northern
Cape, South Africa. Journal of Archaeological Science 35, 1765e1775.
Pearson, O.M., 1997. Postcranial Morphology and the Origin of Modern Humans.
State University of New York.
Pearson, O.M., Fleagle, J.G., Grine, F.E., Royer, D.F., 2008a. Further new hominin
fossils from the Kibish Formation, southwestern Ethiopia. Journal of Human
Evolution 55, 444e447.
Pearson, O.M., Royer, D.F., Grine, F.E., Fleagle, J.G., 2008b. A description of the Omo I
postcranial skeleton, including newly discovered fossils. Journal of Human
Evolution 55, 421e437.
135
Phillipson, D.W., 1977. The excavation of Gobedra rockshelter. Azania 12, 53e82.
Pleurdeau, D., 2003. Le Middle Stone Age de la grotte du Porc-Épic (Dire Dawa,
Éthiopie): gestion des matières premières et comportements techniques.
L’Anthropologie 107, 15e48.
Pleurdeau, D., 2004. Gestion des matières premières et comportements techniques
dans le Middle Stone Age africain: les assemblages lithiques de la grotte du
Porc-Épic (Dire Dawa, Éthiopie). British Archaeological Reports Limited.
Pleurdeau, D., 2005a. Human technical behavior in the African middle stone age:
the lithic assemblage of Porc-Epic cave (Dire Dawa, Ethiopia). African Archaeological Review 22, 177e197.
Pleurdeau, D., 2005b. The lithic assemblage of the 1975e1976 excavation of the
Porc-Epic Cave (Dire-Dawa, Ethiopia). Implications for the East African Middle
Stone Age. Journal of African Archaeology 3, 117e126.
Reimer, P., 2013. IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0e
50,000 years cal BP. Radiocarbon 55, 1869e1887.
Rightmire, G.P., Deacon, H.J., 1991. Comparative studies of Late Pleistocene human
remains from Klasies River Mouth, South Africa. Journal of Human Evolution 20,
131e156.
Rightmire, G.P., Deacon, H.J., 2001. New human teeth from Middle Stone Age
deposits at Klasies River, South Africa. Journal of Human Evolution 41,
535e544.
Rightmire, G.P., Deacon, H.J., Schwartz, J.H., Tattersall, I., 2006. Human foot bones from
Klasies River main site, South Africa. Journal of Human Evolution 50, 96e103.
Rose, J.I., Usik, V.I., Marks, A.E., Hilbert, Y.H., Galletti, C.S., Parton, A., Geiling, J.M.,
\vCernỳ, V., Morley, M.W., Roberts, R.G., 2011. The nubian complex of Dhofar,
Oman: an African middle stone age industry in southern Arabia. PloS One 6,
e28239.
Royer, D.F., Lockwood, C.A., Scott, J.E., Grine, F.E., 2009. Size variation in early human
mandibles and molars from Klasies River, South Africa: comparison with other
middle and late Pleistocene assemblages and with modern humans. American
Journal of Physical Anthropology 140, 312e323.
Singer, R., Wymer, J., 1982. The Middle Stone Age at Klasies River Mouth in South
Africa. University of Chicago Press Chicago.
Teilhard de Chardin, P., 1930. Le Paléolithique en Somalie fran\ccaise et en Abyssinie. L’Anthropologie 40, 331e334.
Teilhard de Chardin, P., Breuil, H., Wernert, P., 1940. Les Industries lithiques de
Somalie française. L’Anthropologie 49, 497e522.
Tishkoff, S.A., Reed, F.A., Friedlaender, F.R., Ehret, C., Ranciaro, A., Froment, A.,
Hirbo, J.B., Awomoyi, A.A., Bodo, J.-M., Doumbo, O., Ibrahim, M., Juma, A.T.,
Kotze, M.J., Lema, G., Moore, J.H., Mortensen, H., Nyambo, T.B., Omar, S.A.,
Powell, K., Pretorius, G.S., Smith, M.W., Thera, M.A., Wambebe, C., Weber, J.L.,
Williams, S.M., 2009. The genetic structure and history of Africans and African
Americans. Science 324, 1035e1044.
Umer, M., Legesse, D., Gasse, F., Bonnefille, R., Lamb, H.F., Leng, M.J., Lamb, A.A.,
2004. Late Quaternary climate changes in the Horn of Africa. In: Battarbee, R.W.,
Gasse, F., Stickley, C.E. (Eds.), Past Climate Variability through Europe and Africa.
Developments in Paleoenvironmental Research 6, pp. 159e180.
Van Peer, P., 1998. The Nile corridor and the Out-of-Africa model: an examination of
the archaeological record. Current Anthropology 39, S115eS140 supplement.
Verna, C., Texier, P.-J., Rigaud, J.-P., Poggenpoel, C., Parkington, J., 2013. The Middle
Stone Age human remains from Diepkloof Rock Shelter (Western Cape, South
Africa). Journal of Archaeological Science 40, 3532e3541.
Vogel, N., Nomade, S., Negash, A., Renne, P.R., 2006. Forensic 40Ar/39Ar dating: a
provenance study of Middle Stone Age obsidian artifacts from Ethiopia. Journal
of Archaeological Science 33 (12), 1749e1765.
Walter, R.C., Buffler, R.T., Bruggemann, J.H., Guillaume, M.M., Berhe, S.M., Negassi, B.,
Libsekal, Y., Cheng, H., Edwards, R.L., Von Cosel, R., 2000. Early human occupation of the Red Sea coast of Eritrea during the last interglacial. Nature 405,
65e69.
Wang, Q., Tobias, P.V., Roberts, D.L., Jacobs, Z., 2008. A re-examination of a human
femur found at the Blind River Site, East London, South Africa: its age,
morphology, and breakage pattern. Anthropological Review 71, 43e61.
White, T.D., Asfaw, B., DeGusta, D., Gilbert, H., Richards, G.D., Suwa, G., Howell, F.C.,
2003. Pleistocene Homo sapiens from Middle Awash, Ethiopia. Nature 423,
742e747.