Journal of Human Evolution 65 (2013) 266e281
Contents lists available at SciVerse ScienceDirect
Journal of Human Evolution
journal homepage: www.elsevier.com/locate/jhevol
Origins of the Iberomaurusian in NW Africa: New AMS radiocarbon
dating of the Middle and Later Stone Age deposits at Taforalt Cave,
Morocco
R.N.E. Barton a, *, A. Bouzouggar b, J.T. Hogue a, S. Lee c, S.N. Collcutt d, P. Ditchfield c
a
Institute of Archaeology, University of Oxford, 36 Beaumont Street, Oxford OX1 2PG, UK
Institut National des Sciences de l’Archéologie et du Patrimoine, Hay Riad, Madinat Al Irfane, Angle rues 5 et 7, Rabat-Instituts, 10 000 Rabat, Morocco
Research Laboratory for Archaeology and the History of Art, University of Oxford, Dyson Perrins Building, South Parks Road, Oxford OX1 3QY, UK
d
Oxford Archaeological Associates Ltd., 1 Divinity Road, Oxford OX4 1LH, UK
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 February 2013
Accepted 1 June 2013
Available online 24 July 2013
Recent genetic studies based on the distribution of mtDNA of haplogroup U6 have led to subtly different
theories regarding the arrival of modern human populations in North Africa. One proposes that groups of
the proto-U6 lineage spread from the Near East to North Africa around 40e45 ka (thousands of years
ago), followed by some degree of regional continuity. Another envisages a westward human migration
from the Near East, followed by further demographic expansion at w22 ka centred on the Maghreb and
associated with a microlithic bladelet culture known as the Iberomaurusian. In evaluating these theories,
we report on the results of new work on the Middle (MSA) and Later Stone (LSA) Age deposits at Taforalt
Cave in Morocco. We present 54 AMS radiocarbon dates on bone and charcoals from a sequence of late
MSA and LSA occupation levels of the cave. Using Bayesian modelling we show that an MSA nonLevallois flake industry was present until w24.5 ka Cal BP (calibrated years before present), followed
by a gap in occupation and the subsequent appearance of an LSA Iberomaurusian industry from at least
21,160 Cal BP. The new dating offers fresh light on theories of continuity versus replacement of populations as presented by the genetic evidence. We examine the implications of these data for interpreting
the first appearance of the LSA in the Maghreb and providing comparisons with other dated early blade
and bladelet industries in North Africa.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Maghreb
North Africa
Microlithic bladelet industry
Introduction
Recent phylogenetic studies of mtDNA haplogroups M1 and U6
have proposed that modern human populations in North Africa
originated from groups that had migrated into this region from
Southwest Asia (Maca-Meyer et al., 2003; Olivieri et al., 2006;
Gonzalez et al., 2007). However, the nature, timing, and geographical spread of such a back-migration are still a matter of considerable
debate (Pennarun et al., 2012). On the one hand, some studies propose an early dispersal of M1 and U6 lineages into North Africa at
w40e45 ka (thousands of years ago) (Olivieri et al., 2006), while
others suggest multiple events with a major expansion of the U6
lineages in the Maghreb w22 ka (Maca-Meyer et al., 2003; Pereira
et al., 2010). Bound up with these models is the proposal that the
geographical patterns of the haplogroups can be shown to coincide
* Corresponding author.
E-mail address: nick.barton@arch.ox.ac.uk (R.N.E. Barton).
0047-2484/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jhevol.2013.06.003
with major technological shifts in the archaeological record. One of
these relates to sub-haplogroup U6a1 and its posterior clade U6a1a,
with coalescence ages of w22 ka, which may be associated with the
appearance of a culture known as the Iberomaurusian (Maca-Meyer
et al., 2003). This microlithic bladelet industry is significant because
it represents potentially the earliest Later Stone Age technology in
the Maghreb (Morocco, Algeria, Tunisia). The genetic studies
therefore also highlight the issue of whether the Iberomaurusian
was a truly indigenous development to the Maghreb or whether it
reflects a general spread of people and traditions from Cyrenaica
with older roots in Southwest Asia.
While the published genetics research provides useful models
for understanding the early peopling of North Africa by modern
humans, considerable caution must be exercised in interpreting
these data. One issue concerns underlying assumptions regarding
the timing of dispersal events that are heavily dependent on the
methodology used to estimate molecular divergence values and
DNA mutation rates (Endicott et al., 2009; Scally and Durbin, 2012).
Indeed there is still a huge disparity between the age of U6 and
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
some of its individual clades, such that the age of U6a7 is consistently estimated as being older than that of U6 overall (Olivieri
et al., 2006; Pereira et al., 2010). A further challenge is to assess
whether any of the proposed demographic models can be
demonstrated by testing them against empirical evidence in the
archaeological and palaeontological records. For example, did the
arrival of modern humans in the Maghreb result in abrupt changes
in the archaeological record (replacement model) or were changes
brought about independently and within the context of long-term
population continuity (Debénath et al., 1986; Linstädter et al.,
2012)? Until now, it has been difficult to assess such claims,
because of the absence of high precision dating records for key
regions such as the Maghreb and the scarcity of well-stratified
archaeological sequences with associated human fossil remains. A
site of major significance that can help remedy this situation is
that of Grotte des Pigeons at Taforalt. The cave is located in the
Beni-Snassen Mountains, in northeastern Morocco (Fig. 1), and has
been the subject of recent excavations that provide a long and
largely unbroken sequence of archaeological deposits from w12 to
110 ka (Bouzouggar et al., 2007; Clark-Balzan et al., 2012), covering
the period of the proposed arrival in the region of modern humans.
The sediments include both extensive Aterian and Iberomaurusian
occupation, with cemetery evidence (Bouzouggar et al., 2006, 2007,
2008; Barton et al., 2007; Taylor et al., 2011; Humphrey et al., 2012).
In this paper we report on the upper part of the sequence
comprising the terminal Middle (MSA) and Later Stone (LSA) Age
deposits of the cave. The 54 AMS (accelerator mass spectrometry)
radiocarbon dates provide the first well constrained record for the
appearance of the Iberomaurusian in northwest Africa. They also
provide a basis for comparing the genetics-derived chronology and
enable an independent test of the timing of the transition from the
MSA in this region.
Among the most intriguing elements in this research are those
that concern the nature and origins of the Iberomaurusian. The
techno-complex is found very widely distributed across North Africa and is associated with cemeteries containing skeletal remains
of robust modern humans attributed to Mechta-Afalou types
(Camps, 1974; Lahr, 1996; Irish, 2000; Humphrey and Bocaege,
2008). The Iberomaurusian is particularly well documented in
cave, rock-shelter, and open-air sites in the Mediterranean coastal
zone of the Maghreb with a distribution that potentially extends
267
into Cyrenaica (McBurney, 1967; Barker et al., 2008) and Egypt
(Phillips, 1972). The Iberomaurusian lithic industry is typified by
microlithic backed bladelets and, apart from its geographically
wide distribution, is significant because it marks a diagnostically
clear change from Middle Palaeolithic/Middle Stone Age technologies in the Maghreb (Lubell, 2001; Bouzouggar et al., 2008). Many
specialists assign the Iberomaurusian to the Epipalaeolithic (Roche,
1963; Barton et al., 2007; Olszewski et al., 2011) but, despite the
extraordinary wealth and density of findspots in the Maghreb and
over a century of research, relatively little is known about how or
where it originated.
Various theories have been proposed for the cultural origins of
the Iberomaurusian. The term itself derives from the fusion of two
words ‘Ibero’ (meaning Spanish) and ‘Maurusian’ (referring to
Mauretania tingitana, the name first given by the Romans to
northern Morocco and western Algeria). The definition was introduced by Pallary, who used it to draw attention to similarities between lithic industries in Spain and Morocco that contained “une
profusion de très petites lames à dos retouché et à pointe très
aiguë” (Pallary, 1909). The implied link with southern Europe was
dismissed by later archaeologists who recognised stronger African
affinities and adopted alternative names reflecting regional sources
such as ‘Oranian’ (Gobert and Vaufrey, 1932) and ‘Mouillian’ (Goetz,
1941) from type locations in Algeria. However, the term Iberomaurusian has always persisted in the literature and, for reasons
of taxonomic priority, we shall continue to use it here. Divergent
with these views was an idea put forward by McBurney (1967), that
the Oranian/Iberomaurusian had arisen out of an ‘Upper Palaeolithic’ industry known as the Dabban, represented at the Cyrenaican site of Haua Fteah and which may be of Near Eastern origin.
But a major anomaly in this scheme was that the Iberomaurusian
appeared to be earlier in the Maghreb than for the rest of North
Africa (McBurney, 1977; Close, 1986). More recently, it has been
suggested that the development of the Iberomaurusian was part of
a much wider, pan-regional phenomenon resulting in the appearance of backed bladelet technologies across much of North Africa
and the Near East around 20e23 ka BP (Close and Wendorf, 1990;
Vermeersch, 1992; Godfrey-Smith et al., 2003; Goring-Morris and
Belfer-Cohen, 2003). However, this theory neither adequately
addressed the possibility of an early Iberomaurusian in the Maghreb nor inherent differences in the tool typologies at Upper Nile
Figure 1. Distribution of Iberomaurusian sites. 1. Cap Rhir, 2. El Khenzira, 3. Contrebandiers, 4. El Harhoura II, 5. Dar es-Soltan I, 6. Ghar Cahal, 7. Kehf El Hammar, 8. Hattab II, 9. Ifri El
Baroud, 10. Ifri n’Ammar, 11. Kifan Bel Ghomari, 12. Taforalt, 13. La Mouillah, 14. Rachgoun, 15. Columnata, 16. Cap Ténès, 17. Rolland, 18. Rassel, 19. Oued Kerma, 20. El Hamel, 21. ElOnçor, 22. Afalou Bou Rhummel, 23. Tamar Hat, 24. Taza, 25. Ouchtata localities, 26. Horizon Collignon.
268
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
sites such as Wadi Halfa, Gebel Silsila and Deir el Fakhouri, which
would make such comparisons less likely (Vermeersch, 1992). A
more radical proposal visualised links between the Epigravettian
industries of the Italian peninsula and the Iberomaurusian (Camps,
1974; Ferembach, 1985; Debénath, 2003). Although some similarities exist in the typology of the industries, there are many other
lines of evidence (palaeontological, genetic, dating) that disfavour
an origin of the Iberomaurusian in the Epigravettian of Italy or Sicily
(Mannino et al., 2011, 2012).
Part of the problem in assessing the Iberomaurusian have been
continuing doubts and ambiguities over the correct dating of
samples and the integrity of their cultural associations. According
to published studies, the earliest conventional radiocarbon ages for
the Iberomaurusian come from Grotte des Pigeons (Taforalt),
Morocco and Tamar Hat in Algeria. At Taforalt, Roche (1976)
recorded two very early ages from ‘terre charbonneuse’ (charcoalrich sediments) of 21,900 400 BP (Gif-2587) and 21,100 400 BP
(Gif-2586). But, for reasons that will be discussed below, both of
these are now regarded as highly doubtful. Elsewhere in the
Maghreb, the oldest radiocarbon date recorded for the Iberomaurusian is (MC-822) 20,600 500 BP from Layer 84/5 at Tamar
Hat (Saxon et al., 1974). However, despite the fact that this was one
of seven dates ranging from 20,600 500 to 16,100 360 BP, five of
which came from stratified contexts of the 1973 excavations,
scepticism remains over the use of bulked materials, which may be
susceptible to cross-contamination. This is a common problem with
other dated sites in the Maghreb whose ages are based on bulked
samples and thus subject to mixing of carbonised material of
potentially different ages. Outside the Maghreb, the best dating for
the oldest Iberomaurusian still comes from the Haua Fteah in
Cyrenaica, where layers excavated by McBurney can be shown to be
no older than two radiocarbon dates of 16,070 100 BP (GrN-2586)
and 18,620 150 BP (GrN-2585) (Close, 1986). Nonetheless the
dating was again based on bulked charcoal samples and therefore
susceptible to similar doubts over reliability.
Allied to the uncertainties with chronology have been questions concerning the stratigraphic relationship between the Iberomaurusian and older industries. In caves in the Maghreb with wellpreserved sedimentary sequences, the Iberomaurusian can often be
shown to overlie Aterian deposits. However, the nature of continuity or discontinuity between these two techno-complexes has
been a matter of longstanding debate. Early archaeologists such as
Antoine (1937) saw no appreciable gap between the Aterian and
Iberomaurusian but this was not widely accepted and was gradually
replaced by a consensus in favour of a hiatus separating the Aterian
from the Iberomaurusian of between five to ten thousand years
(Close, 1980, 1988; Debénath et al., 1986). Even so, such interpretations relied on questionable or minimum radiocarbon estimates that seemed to show the Aterian occupying a relatively short
chronology from 40 to 20 ka BP (Bordes, 1976e1977; Debénath et al.,
1986; Texier et al., 1988; Debénath, 2000). Newer studies based on
luminescence, uranium series and AMS dating have now led to a
drastic revision of this timescale with much earlier ages for both the
oldest and latest Aterian occurrences (Bouzouggar et al., 2007; Roset
and Harbi-Riahi, 2007; Richter et al., 2010; Clark-Balzan et al., 2012;
Jacobs et al., 2012). For areas outside the Maghreb, a slightly
different picture has emerged. In Cyrenaica, the Iberomaurusian can
be demonstrated to lie directly above the Dabban ‘Upper Palaeolithic’ industry (McBurney, 1967; Barker et al., 2008), while a
similar industry is also documented in northwestern Libya (Garcea,
2004). Thus, two potentially contrasting cultural-historical models
have been proposed for geographically adjacent areas: one for the
Maghreb exemplified by an Aterian-Iberomaurusian succession, and
one outside this region to the east with an ‘Upper Palaeolithic’ industry as a cultural forerunner of the Iberomaurusian. In this paper,
we re-examine the evidence for the dating of the sequence in the
Maghreb and consider some of its wider implications.
The aims of this paper, therefore, are four-fold:
To provide a first high precision Accelerator Mass Spectrometry
(AMS) radiocarbon dating record for Taforalt in the Maghreb,
spanning the Iberomaurusian (LSA) and the most recent prebladelet technology (MSA).
To identify (dis)continuities in the stratified occupation sequence,
using a combination of sedimentological data and depth-age
modelling based on Bayesian analyses and a Poisson process
deposition model.
To assess relationships of any gaps in settlement and cultural
shifts at Taforalt with environmental fluctuations by comparing
their timing to existing archaeological and palaeoclimatic records in North Africa and globally.
To use these data to examine more widely the implications for
theories of modern human dispersal in North Africa proposed
in the genetic evidence.
Site setting, stratigraphy and archaeological context
Grotte des Pigeons at Taforalt (34 480 3800 N, 2 240 3000 W) is
located at 720 m above mean sea level overlooking the Zegzel Valley
in the Beni Snassen mountain range (Fig. 2). The bedrock in this area
comprises steeply folded Permo-Triassic dolomitic limestones, with
the cave itself having formed by rekarstification in a zone of earlier
travertines and fluvial conglomerates, constituting a more ancient
deep karstic fill. The currently accessible cave, with a large entrance
opening to the northeast, has a floor area within the drip line of
w400 m2. Today the site lies w40 km from the Mediterranean coast,
currently within the ‘thermo-Mediterranean’ biozone (Blondel and
Aronson, 1999).
Major excavations were undertaken at Taforalt Cave in 1944e
1947, 1950e1955, and 1969e1977 (Roche, 1953, 1963, 1967, 1969,
1976), with further investigations taking place during the 1980s
(Raynal, 1980; Courty et al., 1989) and a new phase of excavations
that was begun in 2003 (Bouzouggar et al., 2006, 2007, 2008; Barton
et al., 2007; Taylor et al., 2011). The new phase of work involved
excavating from standing profiles left by Roche and other previous
excavators. The aim was to collect fresh dating samples and to
investigate the archaeological deposits that spanned a combined
depth of over 10 m, and which contained rich Aterian hearth layers
overlain by a 4-metre thick sequence of Iberomaurusian deposits.
The latter included midden layers and an assemblage of partial
skeletons recovered from two burial areas investigated between
1952 and 1955 by Roche (Ferembach et al., 1962; Roche, 1963).
Part of the new work involved the investigation of an area
contiguous with long sections dug by Roche on the south side of the
cave, which we refer to as Sector 8 (Fig. 2). Other new trenches
were located on the north side of the cave (Sector 9) and in the
cemetery area (Sector 10). The sedimentary sequence for Sector 8 is
shown in Fig. 3 and summarised schematically in Fig. 4. In the main
part of the cave, Roche (1963, 1976) identified at least 17 sedimentary units (layers I-XVII) containing archaeological finds, which
he subdivided into cendreuses (ashy) and argillo-sableuses (clayey
sands), with a distinct separation between the upper grey ashy
series (levels IeVIIII) and a lower series of reddish brown to yellow
clayey sands (IXeXVII).
Like Roche, we recognise the same major stratigraphic division
between the Grey Series (ashy deposits) and the underlying Yellow
Series (clayey sands) but we differ in the finer descriptive detail of the
sedimentary succession. The Grey Series comprises an approximately
4 m thickness of dominantly anthropogenic ‘midden’ deposits (ash,
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
269
Figure 2. Plan of Taforalt showing earlier Roche excavations and main sectors of the recent field campaigns.
charcoal, bone and snail shell debris, burnt limestone, etc). Locally
within Sector 8, it is possible to maintain precise and detailed lithostratigraphic control, despite the dominantly lenticular sedimentation mode resulting from constantly shifting human activity in
occupation areas across the site. The results of new excavation and
sampling are reported in two separate but broadly equivalent sequences, some 3 m laterally apart, and excavated in 2003e05 (units
G88-100) and 2009e10 (layers 1e29), the latter located slightly
deeper into the cave (Fig. 4).
In contrast to the Grey Series, the Yellow Series is finely laminated throughout, indicating emplacement by wash processes,
with greater lateral continuity in stratigraphic units. These sediments are always dominated by fine to medium sands, with
varying amounts of dolomitic limestone debris and minor peaks
of quartzitic ‘grit’. Whilst there is an archaeological presence at
most levels, only rarely does the anthropogenic input reach
concentrations high enough to be reflected in the actual lithostratigraphy (e.g., the persistent traces of burning in the distinctive
horizons of Units Y3 and Y5). Three of the many points of interest
in this approximately 2 m thickness of Yellow Series deposits are
singled out for comment here. First, there is an increasingly
strong quartzitic coarse silt component upwards in Unit Y2, with
plastic deformation phenomena present below the sharp upper
boundary. Second, the middle portion of Unit Y4 has a more
massive structure (lacking clear boundaries and significant
erosion planes), with few stones and, again, a peak in quartzitic
coarse silt. Third, from the top boundary of Y5 downwards (Y13
being the lowest unit reported here), it can be observed that
erosion planes (often irregular) between units and the lamination
within units are more strongly characterised than in overlying
deposits. The possible implications of these features of the Yellow
Series will be discussed below.
270
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
this has been done on strictly defined typological and technological criteria and the phases can be correlated with specific
chronostratigraphic units. The bladelet assemblages that make up
the Iberomaurusian are markedly different from the flake assemblages that underlie them. These bear no obvious resemblance
either in raw material or technology to the Iberomaurusian. A full
study of the lithic assemblages will be published elsewhere
(Hogue, in prep.) but the preliminary findings are summarised
below.
Grey Series IB3 Iberomaurusian lithic assemblages (layers 2e29)
Figure 3. Main section on the south side of the cave (Sector 8) showing the Grey Series
overlying the Yellow Series. Scale is 2 m long. (Photo: Ian Cartwright; copyright: Institute
of Archaeology).
In his monograph of the site, Roche (1963) described the archaeological sequence for the equivalent of Sector 8. The Iberomaurusian finds were analysed by layer but little attempt was made
to explain variability across the sequence that was only alluded to
briefly (Roche, 1963). In a subsequent publication, Roche subdivided
the Iberomaurusian into two main phases: a phase ancienne and a
phase classique (Roche, 1976). This followed the tripartite model
developed for the Iberomaurusian by Camps (1974), though the latest
phase evoluée of the Iberomaurusian was missing at Taforalt. Roche’s
subdivisions of ancienne and classique were based on variation in the
gross categories of tools (i.e., relative proportions of end-scrapers,
burins, notches, backed bladelets, etc.) (Roche, 1963), with little, if
any, attention given to variability within the artefact classes themselves. His scheme was also difficult to replicate because his classification did not take account of the more widely accepted typology
devised by Tixier (1963) for the Epipalaeolithic of the Maghreb. In all,
Roche described Iberomaurusian artefacts from eight of the grey ashy
layers (couches cendreuses) and from nine of the underlying clayey
sand layers (couches argilo-sableuses) (Roche, 1976). The ages on
charcoals obtained from these stratigraphic units ranged from
10,800 400 BP to 16,420 190 BP, and included one from the
cemetery in the grey ashy layers, which yielded an age of 11,920
240 BP. None of these dates overlapped with the two appreciably
older ages on burnt sediments from lower down in the clayey sands
e also attributed by him to the Iberomaurusian e with ages of
21,100 400 BP and 21,900 400 BP, respectively. Unfortunately
there is no record of the artefacts so the association remains a matter
of conjecture. Equally, our own studies of the same sequence excavated and drawn by Roche (actually surviving in our Area 3, see Fig. 2)
have failed to verify any of his suggested ‘cultural’ boundaries. In
addition, neither of the two radiocarbon dates were on charcoals and
it is questionable whether determinations obtained on burnt earth
would be sufficiently free of extraneous sources of carbon to have
provided a reliable age for the deposit. Elsewhere in the same paper,
Roche (1976) briefly refers to the direct superposition of the Iberomaurusian over the Aterian and a suggestion that the contact between these two cultural horizons could be dated by a radiocarbon
date on Helix snail shells of (Gif-2276) 32,370 þ 1890/-2470 BP. Again,
it is impossible to verify the stratigraphic position of any of these
samples as no field notes survive.
According to the analysis of lithics from the new excavations in
our Sector 8, we would now subdivide the Iberomaurusian (sensu
lato) into three distinctive phases. In contrast to the earlier work,
Only a small degree of variation is present in the lithic assemblages recovered from the sedimentary layers of the Grey Series
ashy deposits. Unretouched flakes and bladelets made with a soft
hammer, and struck from single platform and opposed platform
bladelet cores, are recorded. However, they occur in lower proportions than in the underlying deposits. There is a high frequency
of burning within the assemblage. The raw materials consist of small
fine-grained siliceous river cobbles known to derive from locations
no further than about 25 km from the site. Sources closer to the site
may have been exploited but these do not seem to include the
streambed immediately below and in front of the cave. The
retouched tools (Fig. 5) are dominated by backed bladelet types to
the exclusion of almost all other tool classes. These pieces fall most
commonly in the categories of curve-backed bladelets (Tixier types
56 through 59) and longer and more elongated pointed straightbacked bladelets (types 45e51). There is also evidence for the use
of the microburin technique. End-scrapers, notched and denticulated pieces, and simple retouched flakes and blades are observed in
small numbers, whilst burins and most other tool forms are absent
from the assemblage.
Upper Yellow Series IB2 Iberomaurusian lithic assemblages (Unit Y1)
In contrast to the younger phase, bladelet and flake debitage is
more common from this unit, which also includes greater numbers
of cores. Similar small river cobbles were brought to the cave for
knapping. The preferred method was to split the cobbles longitudinally before preparing a unidirectional crest down one edge of
the split piece. Removal of blanks proceeded from one end of the
core, with knapping sometimes switched to the other end, rather
than regularly alternating between the two platforms. Bladelets
show low levels of platform abrasion or other forms of preparation.
The most noticeable difference from the assemblages from phase 3
is extensive use of the microburin technique. Microburins are very
common in these levels (Fig. 5); some of them are quite large, and
comprise predominantly distal types with the notch formed on the
left lateral margin. Evidence for the use of the microburin technique is also observed in the high proportions of La Mouillah points,
a backed bladelet form that retains the microburin facet either
distally or proximally. Previous debates have focused on whether
this type represents a formal tool in itself or is an intermediate
stage in the manufacture of backed bladelets (Neeley and Barton,
1994; Olszewski et al., 2011). With the exception of the La Mouillah points, the range of retouched pieces within the Upper Yellow
Series is similar to that found in the overlying Grey Series and there
is clear technological continuity across the GreyeYellow Series
transition.
Lower Yellow Series IB1 Iberomaurusian lithic assemblages (Units
Y2eUpper Y4)
The assemblages from these layers are also characterised by
bladelets in a range of fine-grained siliceous raw materials knapped
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
271
Figure 4. Schematic section through the Grey and Yellow Series deposits in Sector 8, illustrating different excavation columns and AMS radiocarbon dates (see Tables 1 and 2).
from river cobbles. There is a high degree of similarity with phase 2
in the core-reduction procedures and in the use of soft hammer
percussion but a strong divergence can be seen in the method of
retouching bladelet tools from those in the phases above. There is no
evidence of the microburin technique in the Lower Yellow Series.
The distal ends of retouched bladelets are also rarely modified.
Instead, the most common tool forms are obtuse-ended backed
bladelets (Tixier’s type 67), with semi-abrupt or abrupt backing
down one margin, and Ouchtata bladelets (Tixier’s type 70), which
carry fine retouch down all or part of one edge (often the proximal
portion). Pointed backed forms also occasionally occur (Fig. 5).
Lower Yellow Series N-LF non-Levallois flake assemblages (units
lower Y4eY11)
There is a change in the nature of the lithic assemblage(s) from
Lower Y4, which becomes much clearer from below this in Y6. The
raw materials in these layers are represented by mainly coarsergrained metamorphic rocks. There are no blades or bladelets and
backed tools are all but absent except in Y4. The debitage consists of
flakes detached from non-Levallois cores and the method of percussion is dominated by non-organic hammer techniques (hard
and soft stone). Amongst the few tools are side scrapers and
272
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Figure 5. Three phases of the Iberomaurusian. Top row: IB3 Curve-backed and elongated straight-backed bladelets; Middle: IB2 curve-backed points and ‘La Mouillah points’
(right); Bottom: IB1 Backed point (far left), Ouchtata bladelets (second and third from right), and obtuse-ended backed bladelets. Scale in cm. (Drawings by J.T Hogue).
thinning flakes that would appear to derive from large bifacial tools
(Fig. 6).
Methods and results of a high precision radiocarbon
chronology
AMS radiocarbon samples
Samples for AMS radiocarbon dating for this study were taken
from Sector 8, the southeastern section of the central part of the cave,
where both Iberomaurusian and underlying layers with a nonLevallois industry were present. A total of 52 radiocarbon determinations of cut-marked bones (collagen fraction) and charcoals
were therefore available from the same section (Sector 8) of the cave.
The majority of samples were individual large charcoals, with
recorded spot heights and all can be confidently identified to stratigraphic unit (Table 1). Four were taken from sediment blocks of 10e
15 cm thickness. The entire sequence runs from 0.22 m to 5.77 m
below Site Datum. Three radiocarbon determinations were duplicates, measured as quality control at ORAU (Oxford Radiocarbon
Accelerator Unit). All samples were pre-treated to remove potential
contaminants using standard ORAU protocols for each sample type
(Brock et al., 2010). All radiocarbon determinations are cited as ‘BP’,
where BP stands for years before 1950 AD; and calibrated radiocarbon or modelled ages on the calendar scale are cited as ‘Cal BP’.
Two additional determinations (shown in Table 2 but not in
Table 1), from the separate lithostratigraphic section in Sector 9 at
Taforalt, represent the earliest occurrences of the Iberomaurusian
yet known from this site. These were sampled and prepared in
exactly the same manner as for the other determinations. However,
they are excluded from the age/depth model because they come
from a separate area of the site but are mentioned here to add
weight to the overall study and to render the test of the ‘gap hypothesis’ as robust as possible.
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
273
Figure 6. Non-Levallois flake industry at Taforalt. Side scrapers (TAF 4056 in quartzite and TAF 6544 in chert), retouched flake (Taf 4163 in quartzite). Scale in cm. (Drawings by M.
Grenet).
Bayesian analysis
Bayesian analysis is used in this study to combine archaeological
information (prior probability) with calibrated dates (likelihoods) to
improve the precision and accuracy of the chronology (posterior
probability). The computer program OxCal v.4.1 (Bronk Ramsey,
2009a) was used for chronology construction. Two priors for modelling the archaeological information, i.e., the deposition sequence,
were employed. These were a Poisson process model (P_Sequence;
Bronk Ramsey, 2008) and a uniform phase model. The first was used
with depth (in metres) information of individual samples and the
second was used to model samples (from sediment blocks) without
exact spot heights, but with their positions constrained by using the
known depth limits of the sediment blocks. Note that, in Fig. 4
‘apparent’ depth at the section face is shown as the vertical dimension. However, in the Baysesian analysis (cf. Table 1) this has been
corrected to ‘true’ depth to reflect the position of samples in a vertical
stack, with each unit represented as the typical average value of its
observed thickness range. The two different deposition models were
cross-linked at the height limits of the sediment blocks. The model
averaging approach was used to estimate an appropriate value for k in
the P_Sequence model (Bronk Ramsey and Lee, 2013). Duplicate
measurements were combined using the R_Combine function in
OxCal to produce a weighted average.
Formal outlier analysis was also utilised to account for outliers
in the radiocarbon scale and the calendar scale. The outlier models
are specified to allow the possible shifts in the specified scale to be
drawn from a long tail Student t distribution. The outliers can be in
the scale of anywhere between 100 and 104 years. These are the
models recommended by Bronk Ramsey (2009b) for general purposes when the scale of the possible offsets is unknown. When
employed, the overall model is not affected by the odd extreme
outlier. Each measurement is assigned a prior probability of 5% of
being an outlier.
Posterior chronology
The posterior chronology for the Iberomaurusian sequence in
Sector 8 is shown in Fig. 7 and Table 2. The chronology was
modelled using multiple P_Sequence functions due to the presence
of different types of sediments and hiatuses in the sequence (Fig. 8).
In terms of the outlier analyses applied, no outliers were found in
the radiocarbon scale, and only one (OxA-23411) was found to yield
95% probability of being an outlier in the calendar scale. Such
findings support the validity of the individual radiocarbon determinations themselves (Staff et al., 2011).
The model suggests that the boundary between the Yellow and
Grey Series occurred in the range of 15,190e14,830 Cal BP (95.4%
probability) (Table 2). In the Yellow Series, the difference in age
between samples OxA-22788 (Unit Y1) and OxA-16267 (Unit Y2)
was calculated to be between 1245 and 2102 years, and 2677 and
3830 years between samples OxA-16273 (upper Unit Y4) and OxA16271 (lower Unit Y4) (95.4% probability; Figs. 9 and 10).
Discussion
The AMS record of 54 dates for Taforalt provides the largest
coherent set of radiocarbon determinations yet available for this
period in the Maghreb and is an important baseline for understanding the development of this LSA technology and its relationship with stratigraphically older industries here and across North
Africa. The unmodelled ages indicate a timespan of at least 9000
calendar years for Iberomaurusian occupation, beginning abruptly
and with no obvious antecedents at 22,093e21,420 Cal BP (the
earliest Sector 9 sample at two s) and ending in this cave (Sector 8)
at 12,698e12,548 Cal BP (at two s), though younger ages at other
sites indicate a prolonged existence in the region (Bouzouggar et al.,
2008; Linstädter et al., 2012).
One of the significant implications of the new results is the fresh
light they cast on the chronological relationship between the Iberomaurusian and an earlier technology characterised by flakes made
using a non-Levallois technique. In the past, the assumption had
been that pre-Iberomaurusian industries in this region shared a
uniquely Aterian affinity. This idea was directly endorsed by Roche,
who believed that there was a “superposition directe entre l’Atérian
final .et l’Epipaléolithique très ancien.” at Taforalt (Roche, 1976:
157e158), with a break possibly of relatively short duration in between. The observation was repeated by Debénath et al. (1986), who
recognised the same succession in the Témara caves in western
Morocco but with a potentially longer hiatus of up to 5000 years. In
our excavations at Taforalt, we could find no evidence of such a
superimposition and we now suspect that, even if Roche’s observations were consistent, he failed to recognise both the vertical and
lateral complexity of the sedimentary sequences. We would suggest
that such an interpretation is also no longer tenable for sites in the
Témara area (Contrebandiers, El Mnasra, Dar es-Soltane) where redating would imply a much wider gap between the Aterian and the
Iberomaurusian (Schwenninger et al., 2010; Jacobs et al., 2012). At
274
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Table 1
Un-calibrated Radiocarbon determinations from Grotte des Pigeons (Sector 8) at Taforalt, Morocco.
Unit
Grey series
Yellow series
Sample
OxA-24111
OxA-13479
OxA-23404
OxA-13480
OxA-13516
OxA-24112
OxA-13517
OxA-24113
OxA-23405
OxA-23407
OxA-23406
OxA-23408
OxA-23409
OxA-23410
OxA-13477
OxA-23411
OxA-13478
OxA-22902
OxA-22904
OxA-22787
OxA-22785
OxA-22784
OxA-24109
OxA-22786
OxA-22903
OxA-22905
OxA-22788
OxA-16267
OxA-22907
OxA-22906
OxA-22908
OxA-16268
OxA-13519
OxA-22909
OxA-16272
OxA-16269
OxA-16270
OxA-13518
OxA-16242
OxA-16273
OxA-16271
OxA-16274
OxA-22910
OxA-16275
OxA-24110
OxA-13607
OxA-16243
OxA-16244
OxA-13556
OxA-16276
OxA-16278
OxA-16277
14
C determination
10,680
10,935
10,870
10,950
11,065
11,165
10,990
11,540
11,615
11,465
11,445
11,545
11,890
12,405
12,675
13,060
12,495
12,370
12,490
12,545
12,500
12,660
12,605
12,200
13,045
12,665
12,850
14,005
14,230
14,135
14,110
14,515
13,905
14,140
14,630
15,790
16,285
17,085
16,630
17,515
20,420
20,630
20,030
20,560
20,800
22,200
22,890
25,860
25,760
26,550
29,310
29,160
Uncertainty
Av depth (m)
Layer
Species
Ref.
45
40
45
45
45
45
45
50
50
50
55
55
55
55
50
65
50
50
50
55
55
70
55
55
50
50
55
60
55
55
55
60
55
55
60
60
65
65
75
75
90
90
90
90
120
90
120
150
140
140
160
160
0.22
0.36
0.36e0.48
0.36e0.48
0.48
0.53
0.54
0.69e0.61
0.69e0.61
0.94
0.94
1.14
1.56
2.96
3.50
3.51
3.88
3.97e3.88
3.97e3.88
3.97e3.88
3.97e3.88
3.97e3.88
3.98
4.02
4.05
4.08
4.12
4.16
4.17
4.18
4.25
4.26
4.30
4.35
4.36
4.55
4.68
4.72
4.77
4.90
4.96
5.11
5.11e5.16
5.16
5.17
5.24
5.34
5.48
5.61
5.63
5.77
5.77
L2
G88
L3
G89
G89
L4
G90
L6
L6
L8
L8
L11
L15
L25
G97
L28
G99
G99
G99
G99
G99
G99
G100
Y1
Y1
Y1
Y1
Y2
Y2
Y2
Y2
Y2
Y2
Y2
Y2
Y2
Y3
Y3
Y4
Y4
Y4
Y5
Y6
Y6a
Y6
Y6
Y6b
Y6d
Y?
Y7
Y12
Y12
Ammotragus
Pinus sp.
Pinus sp.
Pinus sp.
Pinus sp.
Ammotragus
Dicotyledonous
Juniperus/Tetraclinus
Gazella
Juniperus/Tetraclinus
Juniperus/Tetraclinus
Pinus sp.
Pinus sp.
Juniperus/Tetraclinus
Conifer
Juniperus/Tetraclinus
Juniperus/Tetraclinus
Conifer
Conifer
Conifer
cf. Juniperus
cf. Juniperus
Bos
cf. Cedrus
cf. Juniperus
cf. Arbutus
Conifer
Tetraclinus articulata
cf. Juniperus
Conifer
cf. Arbutus
Tetraclinus articulata
Juniperus/Tetraclinus
Conifer
Quercus sp.
Juniperus sp.
Pinus sp.
Quercus sp.
Dicot unidentified
Pinus sp.
Pinus sp.
Conifer
cf. Cedrus
Pinus sp.
Panthera
Taxus sp.
Juniperus sp.
Pinus sp.
Quercus sp.
Pinus sp.
Cupressus sempervirens
Cupressus sempervirens
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
The dating samples are presented by averaged depths from the surface; where ranges shown, this indicates that the samples came from sediment blocks of known thickness.
many of these and other sites the Iberomaurusian is separated from
earlier (‘Middle Palaeolithic’/MSA) occupation by archaeologically
sterile layers (Linstädter et al., 2012).
The nature of the first ‘radiocarbon hiatus’ in the cultural sequence
at Taforalt is of particular interest because of the implications for the
presence/absence of human populations in northwest Africa at that
particular time. According to the modelled Sector 8 radiocarbon ages,
the gap in occupation between the Iberomaurusian (IB1) and underlying non-Levallois flake (N-LF) industry seems to have been around
3800 years at two s confidence interval. However, this separation may
have been of shorter duration because of two slightly older ages for the
Iberomaurusian in Sector 9 of Taforalt. These age estimates cannot be
incorporated directly into the same age-depth model for the deposits
of Sector 8 (although they can be linked stratigraphically within the
Yellow Series). Taking the calibrated ages for the oldest Iberomaurusian of 22,093e21,420 Cal BP (2s) and the uppermost levels of
the non-Levallois flake industry of 24,769e23,940 Cal BP (2s)
considerably reduces the break to potentially no more than 1900
years. These observations on the hiatus are also underpinned by
sedimentological data. As referred to above, from lower Y4 downwards, the units that contain the non-Levallois flake industry are
characterised by strong bedding features, both in unit boundaries and
internal lamination. The immediately overlying layers display no
major disconformities and (even though there are small-scale signs of
bioturbation) the sediments are more massive in structure and lack
distinct boundaries. This part of the sequence covers a thickness range
of 10e40 cm and is largely devoid of lithic artefacts. More concentrated signs of occupation with identifiably Iberomaurusian lithics
occur above this in upper Y4 that feature fine sands, slightly gritty in
places, with smaller stones present in bands. These observations do
not suggest major gaps in the depositional sequence, certainly not one
corresponding with the ‘radiocarbon hiatus’.
275
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Table 2
Radiocarbon determinations, their un-modelled calibrated ages, (IntCal09) and the modelled posterior age highest probability density (HPD) ranges at 68.2% and 94.5%.
Sample
Sector 8
Sector 9
OxA-24111
OxA-13479
OxA-23404
OxA-13480
OxA-13516
OxA-24112
OxA-13517
OxA-24113
OxA-23405
OxA-23407*
OxA-23406*
OxA-23406/07**
OxA-23408
OxA-23409
OxA-23410
OxA-13477
OxA-23411
OxA-13478
OxA-22902
OxA-22904
OxA-22787
OxA-22785*
OxA-22784*
OxA-22784/85**
OxA-24109
YG
OxA-22786
OxA-22903
OxA-22905
OxA-22788
OxA-16267
OxA-22907
OxA-22906
OxA-22908
OxA-16268
OxA-13519
OxA-22909
OxA-16272
OxA-16269
OxA-16270
OxA-13518
OxA-16242
OxA-16273
OxA-16271
OxA-16274
OxA-22910
OxA-16275
OxA-24110
OxA-13607
OxA-16243
OxA-16244
OxA-13556
OxA-16276
OxA-16278*
OxA-16277*
OxA-16277/78**
OxA-16260
OxA-16240
14
C date
Uncertainty
Un-modelled (Cal BP)
Modelled (Cal BP)
68.2%
95.4%
68.2%
95.4%
Ref
12,662e12,581
12,765e12,665
12,868e12,735
12,893e12,759
13,031e12,897
13,065e12,935
13,076e12,937
13,361e13,281
13,374e13,286
12,709e12,554
12,835e12,636
12,913e12,684
12,944e12,699
13,085e12,841
13,098e12,884
13,100e12,886
13,423e13,161
13,427e13,098
13,425e13,349
13,476e13,380
13,796e13,646
14,540e14,291
14,755e14,596
14,759e14,600
14,866e14,691
14,951e14,771
14,974e14,785
14,986e14,794
13,473e13,299
13,579e13,346
13,856e13,505
14,622e14,166
14,844e14,444
14,849e14,450
14,935e14,571
15,051e14,670
15,055e14,664
15,063e14,669
14,991e14,797
15,096e14,932
15,110e14,940
15,144e14,987
15,204e15,022
15,372e15,050
15,515e15,076
17,126e16,974
17,161e17,020
17,190e17,055
17,479e17,241
17,530e17,242
17,727e17,311
17,962e17,409
18,002e17,625
18,943e18,797
19,544e19,407
19,987e19,542
20,076e19,786
20,763e20,430
24,400e23,963
24,688e24,345
24,764e24,367
24,840e24,479
24,903e24,555
26,814e26,292
28,023e27,561
30,670e30,358
31,016e30,707
31,186e30,996
15,065e14,670
15,171e14,831
15,190e14,830
15,233e14,888
15,342e14,955
15,476e14,995
15,686e15,010
17,204e16,898
17,237e16,950
17,268e16,979
17,532e17,131
17,587e17,158
17,763e17,195
17,985e17,350
18,032e17,459
19,208e18,721
19,801e19,310
20,181e19,445
20,264e19,597
21,160e20,360
24,529e23,859
24,852e24,236
24,890e24,279
24,943e24,378
25,020e24,439
27,374e26,146
28,112e26,962
30,943e29,707
31,120e30,504
31,287e30,880
1
2
3
4
5
6
7
8
9
10
11
10/11
12
13
14
15
16
17
18
19
20
21
22
21/22
23
34,430e33,461
21,560e21,370
21,794e21,480
34,522e33,322
21,746e21,242
22,040e21,431
10,680
10,935
10,870
10,950
11,065
11,165
10,990
11,540
11,615
11,465
11,445
11,456
11,545
11,890
12,405
12,675
13,060
12,495
12,370
12,490
12,545
12,500
12,660
12,562
12,605
45
40
45
45
45
45
45
50
50
50
55
37
55
55
55
50
65
50
50
50
55
55
70
44
55
12,644e12,568
12,879e12,710
12,808e12,646
12,894e12,718
13,085e12,890
13,135e12,962
12,940e12,738
13,432e13,315
13,566e13,362
13,385e13,272
13,378e13,254
13,373e13,271
13,440e13,314
13,830e13,673
14,631e14,164
15,183e14,886
16,135e15,261
14,935e14,255
14,560e14,141
14,924e14,251
15,015e14,568
14,949e14,256
15,175e14,788
15,010e14,606
15,089e14,659
12,698e12,548
12,942e12,649
12,899e12,615
13,051e12,646
13,108e12,747
13,220e12,865
13,070e12,684
13,527e13,258
13,636e13,310
13,445e13,200
13,433e13,173
13,430e13,211
13,562e13,263
13,906e13,492
14,958e14,110
15,447e14,606
16,406e15,190
15,057e14,201
14,910e14,068
15,051e14,196
15,125e14,237
15,066e14,201
15,494e14,524
15,160e14,245
15,204e14,261
12,200
13,045
12,665
12,850
14,005
14,230
14,135
14,110
14,515
13,905
14,140
14,630
15,790
16,285
17,085
16,630
17,515
20,420
20,630
20,030
20,560
20,800
22,200
22,890
25,860
25,760
26,550
29,310
29,160
29,236
18,005
18,185
55
50
50
55
60
55
55
55
60
55
55
60
60
65
65
75
75
90
90
90
90
120
90
120
150
140
140
160
160
114
75
75
14,140e13,957
16,045e15,251
15,180e14,870
15,505e15,095
17,157e16,920
17,487e17,147
17,400e17,009
17,381e16,974
17,861e17,550
17,062e16,863
17,402e17,017
17,939e17,673
19,240e18,795
19,547e19,397
20,386e20,182
19,888e19,568
21,126e20,536
24,512e24,207
24,812e24,459
24,150e23,770
24,744e24,375
24,960e24,556
26,857e26,311
28,017e27,106
30,850e30,483
30,744e30,395
31,208e31,032
34,461e33,660
34,428e33,432
34,445e33,560
21,560e21,343
21,893e21,499
14,482e13,827
16,372e15,183
15,431e14,589
15,873e14,973
17,416e16,819
17,615e16,991
17,523e16,921
17,500e16,901
17,933e17,256
17,179e16,785
17,527e16,925
18,046e17,492
19,305e18,740
19,807e18,965
20,505e19,955
20,094e19,474
21,283e20,465
24,769e23,940
24,970e24,345
24,310e23,546
24,940e24,252
25,121e24,431
27,501e26,173
28,104e26,936
30,994e30,342
30,926e30,290
31,291e30,936
34,567e33,422
34,509e33,297
34,520e33,406
21,804e21,194
22,093e21,420
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
51/52
60
61
Samples marked with an asterisk (*) are duplicates and two asterisks (**) are their combined ages. The sample named “YG” indicates the modelled age for the Yellow/Grey
series transition at 4 m.
Another critical issue in evaluating the archaeological evidence
concerns the attribution of the non-Levallois industry and whether
it differs significantly from the Aterian, found in deeper layers of
the site. In reality, the differences are reasonably clear-cut and are
helped by consistencies in the Aterian typology and technology in
various sites throughout the Maghreb. For example, a particular
feature of the Aterian is the use of the Levallois technique and
associated tools, which usually include a range of scrapers,
pedunculate points and bifacial foliates made on diagnostically
Levallois products including blades (Bouzouggar and Barton, 2012).
In contrast, such items are notably absent in the industry in the
lower sequence in Sector 8, which includes occasional flake tools,
but none of them are made on obvious Levallois blanks. Unfortunately, as there are no layers with Aterian artefacts in Sector 8, a
direct stratigraphic relationship with the non-Levallois industry
cannot be demonstrated for this part of the site. However, in Sector
9, accessible sections have revealed a thick sequence of sediments
that includes Iberomaurusian layers near the top (with the reported
276
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Figure 8. Apparent sedimentation rates, Yellow Series in Sector 8. The modelled C14
date ranges (95.4%) are plotted by the two lines against depth in the sequence, so that
sedimentation rate approximates to the slope of the lines. Human input tends to increase the slope (sedimentation rate), particularly at the strongly anthropogenic Units
Y3, Y5 and Y7. Most instances of apparent low slope are likely to be ‘jumps’ (as indicated in the diagram) caused by demonstrable erosion events. In this respect, the Y8e
11 sequence could be shown as a series of small steps. Only the markedly decreased
slope within Unit Y4 appears to represent a truly significant slowing of sedimentation
(see main text for discussion).
Personal communication). For the moment, Taforalt provides the
only securely dated sequence for a clear cultural succession preceding the Iberomaurusian and showing that an Aterian Levalloisdominated technology was probably replaced by one of nonLevallois type. We believe this model can be tested more widely
and has major ramifications for interpreting other lithic artefact sequences in the Maghreb.
Turning to the Iberomaurusian, although the deposits at Taforalt
represent a thick and fairly continuous record of human occupation, there are in fact subtle variations in the cultural sequence. The
clearest example is the switch from IB1 with marginally backed
(‘Ouchtata’) blades and bladelets to IB2 dominated by microlithic
backed bladelets. The actual transition between the two phases is
marked by a sharp sedimentary contact between Units Y2/Y1. This
is clearly an erosive boundary so it is impossible to know whether
Figure 7. Age-depth model for Taforalt cultural sequence. Posterior AMS chronology is
modelled with one P Sequence function for samples with spot heights.
AMS dates). The layers immediately beneath this are not archaeologically rich but include undiagnostic flakes in raw materials
similar to those in the non-Levallois industry of Sector 8. At the base
of the Sector 9 excavation trench and separated from the lowest
Iberomaurusian by at least 130 cm are a series of charcoal-rich
laminated hearth deposits. These contain clear examples of
Aterian retouched tools (including a bifacial foliate) and Levallois
debitage. An OSL date of 37.57 3.42 BP (Clark-Balzan, Personal
communication) has been obtained from the deposits and represents potentially the youngest age so far recorded for the Aterian at
this site. No radiocarbon dating has yet been undertaken on the
charcoals, but if the age estimate is correct it may give greater
credence to an OSL date of 30.9 2.5 ka years for the Aterian at
Wadi Noun in southern Morocco (Weisrock et al., 2006; Wengler,
2010), and to ESR dating of around 39 4 ka years on the later
Aterian at Mugharet el ‘Aliya in northern Morocco (Wrinn and Rink,
2003), where new radiocarbon dates are also anticipated (Tuross,
Figure 9. Modelled chronological break between MSA non-Levallois flake technology
and the oldest Iberomaurusian.
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Figure 10. Modelled chronological break between Iberomaurusian IB1 and IB2 industries in Y2 and Y1, respectively.
the time gap of 857 years (between 15,686e15,010 BP and 17,204e
16,898 BP at 2s) in the radiocarbon model is more apparent than
real. Yet, observations of the sediments towards the top of Unit Y2
reveal some unusual geological phenomena, which may provide
clues about the environmental conditions during the later formation of this unit. A raised silt content would suggest that wash input
was capturing aeolian dust from the exterior surroundings. It appears too that the signal of regional aridity implied by these processes evidently came to a brusque end with a phase of instability
indicated by plastic deformation (wetness and possibly even transient ground freezing) in the uppermost part of this unit. The wider
consequences of these observations will be explored further below.
Also of relevance is the temporal relationship between the IB2 and
IB3 assemblages. Here the techno-typological changes are relatively minor and relate mostly to the highly varied appearance of
the microlithic bladelet forms in IB3. These introductions seem to
have been made during an uninterrupted period of cave use, and
we would therefore hypothesise that such differences in typology
may represent a drift in the intensity and nature of activities across
the site.
One of the key aims of this paper has been to assess whether
gaps in settlement and shifts in lithic cultural phases can be
correlated with phases of rapid climatic change. Palaeoclimatic
records available from Atlantic and the Alboran Sea marine cores
(Combourieu Nebout et al., 2002; Sánchez-Goñi et al., 2002;
Fletcher and Sánchez-Goñi, 2008) confirm that major climatic
shifts occurred in the western Mediterranean during the Late
Pleistocene, and these could have had a significant impact on human populations. Of particular interest are so-called Heinrich
Events marked by the incursion of cold polar surface waters into
the Mediterranean (Moreno et al., 2005). These appear to have been
associated with phases of greater aridity in Iberia and northwest
Africa as documented in pollen sequences from the Alboran Sea
cores (Combourieu Nebout et al., 2002; Sánchez-Goñi, 2006;
Jiménez-Espejo et al., 2007; Fletcher and Sánchez-Goñi, 2008)
and increased input of windborne dust from the Sahara (Moreno
et al., 2002; Moreno, 2012). In contrast, the periods following
Heinrich episodes appear to have been relatively warm and humid,
as for example the ‘Last Glacial Maximum’ phase after HE2 (Penaud
et al., 2010) and the phase of major warming at the beginning of
Greenland Interstadial 1e, the latter marked by a rise in sea surface
temperatures and the diversion westward of moisture bearing
winds bringing higher precipitation to the Maghreb (Moreno et al.,
2005; Rodrigo-Gámiz et al., 2011). Climate modelling also suggests
that annual rainfall may have fallen below 100 mm per year during
277
certain Heinrich Events (Sepulchre et al., 2007). A further source of
information comes from the direct comparison between the
radiocarbon dated sequence at Taforalt with the NGRIP ice-core
timescale (Fig. 7), which allows broad correlation with DeO
(DansgaardeOeschger) stadial-interstadials of the Greenland record (Bond et al., 1993; Dansgaard et al., 1993; Rodrigo-Gámiz et al.,
2011).
Beginning with the Grey Series, the radiocarbon model confirms
that this unit began accumulating close to the start of Greenland
Interstadial 1e and continued throughout most of GI1, a phase of
relatively humid conditions. However, it is also clear that the Grey
ash deposit is heavily anthropogenic (midden) and that the accumulation rate was extremely rapid (at c. 1.7 m/kyr) so that any
obvious palaeoenvironmental signal may have been effectively
swamped. Only the presence of cedar (Cedrus sp.) charcoals upwards from Unit G89 (Ward, 2007) would imply more ‘montane’
(cooling) conditions, and this could indicate the onset of Younger
Dryas or Greenland Stadial 1 at the very top of this sequence (Fig. 7).
One other potential sign of variation occurs near the base of the
Grey Series where the remains of Barbary ground squirrels (Atlantoxerus getulus) suggest cooler, drier conditions (Bouzouggar et al.,
2008). Moving downward (backwards in time) into the Yellow
Series, clearer evidence of climatic variation (instability) occurs in
the sediments of upper Y2 and here it is interesting to note that the
dating model places this period of deposition within the same time
span as Heinrich Event 1 (HE1). It also marks the point at which IB1
assemblages are replaced by those of IB2, and may be evidence that
the dislocation in the cultural signature was influenced by climatic
change. However, it is difficult to ascertain the length of break
between the two Iberomaurusian phases because of the erosive
unconformity. A more marked example of climate-cultural change
occurs earlier in the Yellow Series sequence. Here there is a clear
time lag between the earliest Iberomaurusian and the N-LF industry. The stratigraphic expression of this ‘gap’ is not, as in the
younger example, an erosion plane. Rather, there is a body of
sediment that is generally finer than the norm and which contains
significant silt (dust). It would therefore appear that the sedimentation rate dropped drastically (to less than 0.05 m/kyr), probably in
a significantly arid and cool period. It is perhaps relevant that the
‘gap’ calculated here includes the time span of HE2, marking the
onset of Greenland Stadial 2 (GS2) and following the more variable
and probably often more moist environmental conditions normally
reconstructed for the preceding GS3. The latter may be reflected in
the strong bedding structure present in the underlying Sector 8
Units Y13eY5 at Taforalt. Some indication of cooler climatic conditions is also implied by a rise in cedar (Cedrus sp.) charcoals in Y65, which disappears by the middle of Y4 (Ward, Personal
communication).
Lastly, what are the implications for the genetic theories of
modern human dispersal in North Africa? Two of the most recent
scenarios put forward in those studies have suggested that the
Iberomaurusian either: 1) originated in populations that gradually
spread westward from Cyrenaica, or 2) developed independently
somewhere in the Maghreb and was subsequently transmitted
eastwards (and possibly west) via an expansion of haplogroups
U6a1 and U6a1a lineages. Archaeologically, in order to satisfy the
conditions of the first hypothesis, one would expect logically that
the oldest expression of this ‘culture’ should occur in the east.
Despite new claims for an earliest age at Haua Fteah of w19 ka
(Barker et al., 2010), we would suggest caution in the interpretation
as this is so far based on dates on two shells from so-called transitional Layers XIVeXV of the old excavation. Equally, suggestion of
early dating of sites in Upper Egypt (Close, 2002) are based on old
charcoal dates from open sites, whilst the oldest age estimate from
western Libya at Ain Shakshuk of 16,750 60 BP (Barich and
278
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Garcea, 2008) is still substantially younger than the oldest Iberomaurusian recorded in the Maghreb. Secondly, if the progenitor of
the Iberomaurusian was an industry with retouched blades and
chamfered pieces like the Dabban then it follows that, given the
earlier presence of Iberomaurusian in the Maghreb, there should
also be signs of the Dabban or similar blade industry in this region.
Here again, the criteria are not clearly met as there are no immediately pre-Iberomaurusian blade industries yet known in the
Maghreb. So far the most westerly findspot of a blade industry that
might fit this description is in the Jebel Gharbi of northwestern
Libya, with estimated ages of w30 ka BP (Garcea, 2004). At the
Haua Fteah, the earliest Dabban can now be firmly anchored by
tephro-chronology to just before w40 ka BP (Lowe et al., 2012). The
dating is significant because it brings the Dabban more closely into
line with other early blade technologies such as the Ahmarian of
the Near East (Belfer-Cohen and Goring-Morris, 2003; Marks, 2003)
and at Nazlet Khater in Egypt (Vermeersch, 2010), which are of
broadly similar age. However, it is unlikely on technological
grounds that there was any demographic connection between
these regions (Marks, 1975; Iovita, 2009). It also follows that any
interactions between early Upper Palaeolithic blade technologies
did not extend very far if at all into the Maghreb.
The second scenario, which envisages the earliest occurrence of
the Iberomaurusian in the Maghreb, offers a better fit with the
archaeological data but raises other important issues. One of these
concerns the relationship of the Iberomaurusian with older technologies in this region. As has been shown above, there are no
obvious antecedents for the blade and bladelet industries in the
Maghreb and it is generally agreed that sterile layers separate the
Iberomaurusian from anything earlier, with no suggestion of cultural continuity (Bouzouggar et al., 2008; Linstädter et al., 2012). A
question therefore arises whether the replacement of non-Levallois
flake assemblages by the Iberomaurusian is equivalent to the
Middle Stone Age (MSA) to Late Stone Age (LSA) transition seen in
other parts of Africa (Clark, 1959; McBrearty and Brooks, 2000;
Mitchell and Barham, 2008). Unfortunately, the only useful comparisons that can be made are with the earliest bladelet technologies in regions that are geographically distant from the Maghreb.
For example, in northeast Africa and the Nile Valley, bladelet industries become prevalent after w25 ka (Schild and Wendorf,
2010). But in these regions, the situation is made more complex
because the Middle Palaeolithic/MSA includes both Levallois flake
assemblages (Nazlet Safaha 2) and blade technologies (Taramsa
Hill, Nazlet Khater), which were present from as early as MIS 3 (Van
Peer, 2004; Vermeersch, 2010). Nonetheless, it is significant that
there are no reported cases of either backed bladelets or ‘ouchtata’
forms in the MSA blade assemblages (Vermeersch, 2010). Further
afield, the systematic manufacture of bladelets seems to have
begun somewhat earlier. For instance, in East Africa, assemblages
with a major bladelet component can be dated to 45 ka BP at
Mochena Borago rockshelter (southwest Ethiopian Highlands)
(Brand et al., 2012) and to 46 ka BP at Enkapune ya Muto (Kenya)
(Ambrose, 1998). In other areas of sub-Saharan Africa at Mumba
rockshelter, northern Tanzania, newly reported OSL ages suggest
that the production of bladelets became abundant around
49.1 4.3 ka BP (Gliganic et al., 2012), while at Border Cave microliths made on opposed platform cores are common from 44,000
to 42,000 Cal BP (Villa et al., 2012). At each of these sites, there are
signs of continuous sedimentation and occupation with underlying
layers, which seem to imply gradual cultural transitions. The only
area that might offer some parallels with the Maghreb is in the
southern margins of southern Africa, where from the end of Marine
Isotope Stage 3 (w25 ka) microlithic assemblages made on
quartzites become manifest and are succeeded by systematic bladelet production in Robberg-type assemblages from around w19 ka
(Klein, 1974; Wadley, 1997; Mitchell, 2002). Thus, over much of
Africa standardised microlithic bladelet production (LSA) can
generally be seen to supercede technologies characterised by more
variable flake and blade manufacture (MSA), although the adoption
of changes was by no means synchronous across the whole continent (Mitchell, 2008). On the basis of these considerations, we
would suggest that the MSAeLSA template can also be applied to
North Africa but that the introduction of bladelet technologies was
considerably delayed in the Maghreb and their appearance by
w22 ka Cal BP and subsequent spread may partly be explained by
the demographic expansion of sub-clades of U6. However, it does
leave several questions unanswered: why did these innovations
emerge when they did in the Maghreb, and did they arise in
response to palaeoclimatic shifts (e.g., Greenland Stadial 2), or were
these innovations linked to subsequent demographic rise or the
result of the influx of new peoples into this region following the
disappearance of the MSA? These and other questions can ultimately only be answered by focusing research on other sites in the
Maghreb similar to Taforalt, which have long sequences that cover a
comparable time span.
Conclusions
In this paper we have presented the first high precision record of
AMS dates for the Late Pleistocene Maghreb, providing a framework for understanding the development of the Iberomaurusian,
the oldest backed bladelet LSA technology in northwest Africa. In
examining the dating evidence at Taforalt, several gaps in the
sequence were noted, including one of possibly as little as 1900
calendar years separating the first appearance of the Iberomaurusian at 22.0e21.4 ka Cal BP from the underlying nonLevallois flake technology, tentatively attributed to the MSA. A
further gap in dating (but this time also coinciding with an erosive
unconformity) can be seen between the earliest Iberomaurusian
industry with ‘Ouchtata’ retouched blades (IB1) and one above
containing microlithic backed bladelets (IB2). The duration of this
gap may have been of the order of one to two thousand years and
confirms that fully developed microlithic components had emerged
in the Iberomaurusian by 15.5e15.0 ka Cal BP. A major accumulation of ashy midden deposits can be identified at Taforalt at 15.2e
14.2 ka Cal BP and use of the cave in the Iberomaurusian was shown
to have continued until about 12.6 ka Cal BP (Tables 1 and 2).
Assessing the relationship of these gaps to potential environmental shifts has been possible with reference both to cave sedimentological data and palaeoclimatic records for the western
Mediterranean. Using broad comparisons with available oceanic
and atmospheric records, it has been observed that the disappearance of the latest MSA at around 24 ka Cal BP may have been
coincident with pronounced cooling of Atlantic and western
Mediterranean waters in Heinrich Event 2 (Penaud et al., 2010),
which would have produced a marked increase in continental
aridity (Moreno et al., 2005). In contrast, the emergence of the
Iberomaurusian seems to have taken place against the relatively
warm and moist conditions indicated for the LGM at these latitudes
(Penaud et al., 2010). At Taforalt, finer silts were observed in the
intervening archaeologically sterile sediments between the proposed MSA and LSA levels that have been interpreted as a signal of
increased regional aridity. Higher in the sequence, the erosive hiatus separating the Iberomaurusian phases IB1 and IB2 can be
shown to coincide with Heinrich Event 1, although oceanic studies
have suggested that this was not necessarily as severely arid as the
earlier Heinrich episode (Penaud et al., 2010) and might be correlated with increased marine palaeoproductivity (Rodrigo-Gámiz
et al., 2011). The latter could have had major beneficial consequences for populations living on and near the coast. Finally, we do
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
not believe it purely fortuitous that the timing of midden accumulation at Taforalt (Grey Series) occurred so close to the beginning
of Greenland Interstadial 1e. During this period, the projected
northwards migration of the Inter-Tropical Convergence Zone and
increased monsoon activity would have brought much wetter
conditions to this area (Rodrigo-Gámiz et al., 2011). Such circumstances might explain or may have given rise to some behavioural
changes that occurred in the Iberomaurusian in this period (e.g.,
evidence of increased dietary breadth, greater sedentism, use of the
deeper part of the cave site as a cemetery).
In conclusion, while present genetic models for the dispersal of
humans in North Africa are largely inconclusive, we would suggest
that the new dating evidence supports an independent origin of the
Iberomaurusian in the Maghreb (certainly by 22 ka Cal BP if not
earlier). The latter implies innovation and transmission of new
ideas that may have arisen in the Maghreb at a time of environmental instability or that were transmitted via rapid population
movements from an area fringing the Maghreb. However, it does
not seem likely that the impetus for change came from areas in the
south on the Atlantic margin where only younger phases of the
Iberomaurusian are so far known (e.g., the Agadir region: Bouzouggar, Personal observation). Parallel developments may have
led to the appearance of backed bladelet technologies in Libya and
Cyrenaica, but here these seem to have been more deeply rooted in
the Dabban. Clearly this picture may change in the light of fresh
evidence but our proposal for the MSA-LSA transition in the
Maghreb provides a robust model that is capable of extensive
testing.
Acknowledgements
We would like to thank Alison Wilkins and Michael Athanson
for their help in producing Figs. 1 and 2, and to Michel Grenet for
the artefact drawings in Fig. 6. We are also grateful to Peter Mitchell
and Phillip Endicott for their comments on earlier drafts of this
paper and to two anonymous reviewers for useful points of clarification and criticism. This project has been funded by the British
Academy, Oxford University, Protars P32/09-CNRST (Morocco),
NERC (NER/T/S/2002/00700 and NE/E015670/1) and the Leverhulme Trust (F/08 735/F).
References
Ambrose, S.H., 1998. Chronology of the Later Stone Age and food production in East
Africa. J. Archaeol. Sci. 25, 377e392.
Antoine, M., 1937. Notes de préhistoire marocaine: XIII e la question atéroibéromaurusienne au Maroc: historique et mise au point, 11è année. Bull. Soc.
Préhist. Maroc., 45e58.
Barker, G., Basell, L., Brooks, I., Cartwright, C., Cole, F., Davison, J., Farr, L., Grün, R.,
Hamilton, R., Hunt, C., Inglis, R., Jacobs, Z., Legge, T., Leitch, V., Morales, J.,
Morley, I., Morley, M., Pawley, S., Pryor, A., Roberts, R., Reynolds, T., el-Rishi, H.,
Simpson, D., Twati, M., van der Veen, M., 2008. The Cyrenaican Prehistory
Project 2008: the second season of investigations of the Haua Fteah cave and its
landscape, and further results from the initial 2007 fieldwork. Libyan Stud. 39,
175e223.
Barker, G., Antoniadou, A., Armitage, S., Brooks, I., Candy, I., Connell, K., Douka, K.,
Drake, N., Farr, L., Hill, E., Hunt, C., Inglis, R., Jones, S., Lane, C., Lucarini, G.,
Meneely, J., Morales, J., Mutri, G., Prendergast, A., Rabett, R., Reade, H.,
Reynolds, T., Russell, N., Simpson, D., Smith, B., Stimpson, C., Twati, M.,
White, K., 2010. The Cyrenaican Prehistory Project 2010: the fourth season of
investigations of the Haua Fteah cave and its landscape, and further results
from the 2007e2009 fieldwork. Libyan Stud. 41, 63e88.
Barich, B.E., Garcea, E.A.A., 2008. Ecological patterns in the Upper Pleistocene and
Holocene in the Jebel Gharbi, Northern Libya: chronology, climate and human
occupation. Afr. Archaeol. Rev. 25, 87e97.
Barton, R.N.E., Bouzouggar, A., Bronk-Ramsey, C., Collcutt, S.N., Higham, T.F.G.,
Humphrey, L.T., Parfitt, S., Rhodes, E.J., Schwenninger, J.L., Stringer, C.B.,
Turner, E., Ward, S., 2007. Abrupt climatic change and chronology of the Upper
Palaeolithic in northern and eastern Morocco. In: Bar-Yosef, O., Mellars, P.,
Stringer, C., Boyle, K. (Eds.), Rethinking the Human Revolution: New Behavioural and Biological Perspectives on the Origins and Dispersal of Modern
279
Humans. Research Monographs of the Macdonald Institute, Cambridge,
pp. 177e186.
Belfer-Cohen, A., Goring-Morris, A.N., 2003. Current issues in Levantine Upper
Palaeolithic research. In: Goring-Morris, A.N., Belfer-Cohen, A. (Eds.), More than
Meets the Eye. Studies on Upper Palaeolithic Diversity in the Near East. Oxbow
Books, Oxford, pp. 1e12.
Blondel, J., Aronson, J., 1999. Biology and Wildlife of the Mediterranean Region.
Oxford University Press, Oxford.
Bond, G., Broecker, W.S., Johnsen, S.J., McManus, J., Labeyrie, L., Jouzel, J., Bonani, G.,
1993. Correlations between climate records from North Atlantic sediments and
Greenland ice. Nature 365, 143e147.
Bordes, F., 1976e1977. Moustérien et Atérien. Quaternaria 19, 19e34.
Bouzouggar, A., Barton, R.N.E., 2012. The identity and timing of the Aterian in
Morocco. In: Hublin, J.-J., McPherron, S. (Eds.), Modern Origins: a North African
Perspective. Springer, Dordrecht, pp. 93e105.
Bouzouggar, A., Barton, R.N.E., Collcutt, S.N., Parfitt, S., Higham, T.F.G., Rhodes, E.,
Gale, R., 2006. Le Paléolithique Supérieur au Maroc: apport des sites du NordOuest et de l’Oriental. In: Sanchidrián, J.-L., Márquez, A., Fullola, J.M. (Eds.), La
Cuenca Mediterránea durante el Paleolítico Superior (38.000 e 10.000 años). IV
Simposio de Prehistoria Cueva de Nerja. Fundación Cueva de Nerja & Málaga
UISPP, Málaga, pp. 138e150.
Bouzouggar, A., Barton, N., Vanhaeren, M., d’Errico, F., Collcutt, S., Higham, T.,
Hodge, E., Parfitt, S., Rhodes, E., Schwenninger, J.L., Stringer, C., Turner, E.,
Ward, S., Moutmir, A., Stambouli, A., 2007. 82,000 year-old shell beads from
North Africa and implications for the origins of modern human behavior. Proc.
Natl. Acad. Sci. 104, 9964e9969.
Bouzouggar, A., Barton, R.N.E., Blockley, S., Bronk-Ramsey, C., Collcutt, S.N., Gale, R.,
Higham, T.F.G., Humphrey, L.T., Parfitt, S., Turner, E., Ward, S., 2008. Reevaluating the age of the Iberomaurusian in Morocco. Afr. Archaeol. Rev. 25, 3e19.
Brand, S.A., Fisher, E.C., Hildebrand, E.A., Vogelsang, R., Ambrose, S.H., Lesur, J.,
Wang, H., 2012. Early MIS 3 occupation of Mochena Borago rockshelter,
Southwest Ethiopian Highlands: Implications for Late Pleistocene archaeology,
paleoenvironments and modern human dispersals. Quatern. Int. 274, 38e54.
Brock, F., Higham, T., Ditchfield, P., Bronk Ramsey, C., 2010. Current pretreatment
methods for AMS radiocarbon dating at the Oxford Radiocarbon Accelerator
Unit (ORAU). Radiocarbon 52, 103e112.
Bronk Ramsey, C., 2008. Deposition models for chronological records. Quatern. Sci.
Rev. 27, 42e60.
Bronk Ramsey, C., 2009a. Bayesian analysis of radiocarbon dates. Radiocarbon 51,
337e360.
Bronk Ramsey, C., 2009b. Dealing with outliers and offsets in radiocarbon dating.
Radiocarbon 51, 1023e1045.
Bronk Ramsey, C., 2010. OxCal Program, vol. 4.1.7. Radiocarbon Accelerator Unit,
University of Oxford, UK. <http://c14.arch.ox.ac.uk/embed.php?File¼oxcal.
html>.
Bronk Ramsey, C., Lee, S., 2013. Recent and planned developments of the program
OxCal. Radiocarbon. in press.
Camps, G., 1974. Les Civilisations Préhistoriques de l’Afrique du Nord et du Sahara.
Doin, Paris.
Clark, J.D., 1959. The Prehistory of Southern Africa. Pelican, London.
Clark-Balzan, L.A., Candy, I., Schwenninger, J.-L., Bouzouggar, A., Blockley, S.,
Nathan, R., Barton, R.N.E., 2012. Coupled U-series and OSL dating of a Late
Pleistocene cave sediment sequence, Morocco, North Africa: Significance for
constructing Palaeolithic chronologies. Geochronology 12, 53e64.
Close, A.E., 1980. Current research and recent radiocarbon dates from northern
Africa. J. Afr. Hist. 21, 145e167.
Close, A.E., 1986. The place of the Haua Fteah in the Late Palaeolithic of North Africa.
In: Bailey, G.N., Callow, P. (Eds.), Stone Age Prehistory. Studies in Memory of
Charles McBurney. Cambridge University Press, Cambridge, pp. 169e180.
Close, A.E., 1988. Current research and recent radiocarbon, dates from northern
Africa, III. J. Afr. Hist. 29, 145e176.
Close, A.E., 2002. Backed bladelets are a foreign country. Archeol. Pap. Am.
Anthropol. Assoc. 12, 31e44.
Close, A.E., Wendorf, F., 1990. North Africa at 18,000 BP. In: Gamble, C., Soffer, O.
(Eds.), The World at 18,000 BP. Unwin Hyman, London, pp. 41e57.
Combourieu Nebout, N., Turon, J.-L., Zahn, R., Capotondi, L., Londeix, L., Pahnke, K.,
2002. Enhanced aridity and atmospheric high-pressure stability over the
western Mediterranean during the North Atlantic cold events of the past 50 k.y.
Geology 30, 863e866.
Courty, M.-A., Goldberg, P., Macphail, R.I., 1989. Soils and Micromorphology in
Archaeology. Cambridge University Press, Cambridge.
Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.S.,
Hammer, C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjornsdottir, A.E., Jouzel, J.,
Bond, G., 1993. Evidence for general instability of past climate from 250-kyr icecore record. Nature 364, 218e220.
Debénath, A., 2000. Le peuplement préhistorique du Maroc: données récentes et
problèmes. L’Anthropologie 104, 131e145.
Debénath, A., 2003. Le Paléolithique supérieur de Maghreb. Praehistoria 3, 259e280.
Debénath, A., Raynal, J.-P., Roche, J., Texier, J.-P., Ferembach, D., 1986. Stratigraphie,
habitat, typologie at devenir de l’Atérien Marocain: Donneés récentes. L’Anthropologie 90, 233e246.
Endicott, P., Ho, S.Y.W., Metspalu, M., Stringer, C., 2009. Evaluating the mitochondrial timescale of human evolution. Trends Ecol. Evol. 24 (9), 515e521.
Ferembach, D., 1985. On the origin of the Iberomaurusians. A new hypothesis.
J. Hum. Evol. 14, 393e397.
280
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Ferembach, D., Dastugue, J., Poitrat-Targowla, M.J., 1962. La Nécropole Epipaléolithique de Taforalt (Maroc Oriental): Étude des Squelettes Humains. Edita
Casablanca, Rabat.
Fletcher, W.J., Sánchez-Goñi, M.F., 2008. Orbital- and sub-orbital-scale climate impacts on vegetation of the western Mediterranean basin over the last 48,000 yr.
Quat. Res. 70, 451e464.
Garcea, E.A.A., 2004. Crossing deserts and avoiding seas: Aterian North Africane
European relations. J. Anthropol. Res. 60, 27e53.
Gliganic, L.A., Jacobs, Z., Roberts, R.G., Domínguez-Rodrigo, M., Mabulla, A.Z.P., 2012.
New ages for Middle and Later Stone Age deposits at Mumba rockshelter,
Tanzania: optically stimulated luminescence dating of quartz and feldspar
grains. J. Hum. Evol. 62, 533e547.
Gobert, E.G., Vaufrey, R., 1932. Deux gisements extrêmes de l’Ibéromaurusian.
L’Anthropologie XVII, 449e490.
Godfrey-Smith, D.I., Vaughan, K.B., Gopher, A., Barkai, R., 2003. Direct luminescence
chronology of the Epipalaeolithic Kebaran site of Nahal Hadera V, Israel. Geoarchaeology 18, 461e475.
Goetz, C., 1941. Note d’Archéologie préhistorique Nord-Africaine sur un foyer oranien de la sablière d’El-Kçar. Bull. Soc. Préhist. Fr 38, 262e265.
Gonzalez, A.M., Larruga, J.M., Abu-Amero, K.K., Shi, Y., Pestano, J., Cabrera, V.M.,
2007. Mitochondrial lineage M1 traces an early human backflow to Africa. BMC
Genomics 8, 223.
Goring-Morris, A.N., Belfer-Cohen, A. (Eds.), 2003. More than Meets the Eye: Studies
on Upper Palaeolithic Diversity in the Near East. Oxbow Books, Oxford.
Humphrey, L.T., Bocaege, E., 2008. Tooth evulsion in the Maghreb: chronological
and geographical patterns. Afr. Archaeol. Rev. 25, 109e123.
Humphrey, L., Bello, S.M., Turner, E., Bouzouggar, A., Barton, N., 2012. Iberomaurusian funerary behaviour: evidence from Grotte des Pigeons, Taforalt,
Morocco. J. Hum. Evol. 62, 261e273.
Iovita, R.P., 2009. Re-evaluating connections between the Early Upper Paleolithic of
Northeast Africa and the Levant: technological differences between the Dabban
and the Emiran. In: Shea, J.J., Lieberman, D. (Eds.), Transitions in Prehistory:
Essays in Honor of Ofer Bar-Yosef. Oxbow Books, Oxford, pp. 125e142.
Irish, J.D., 2000. The Iberomaurusian enigma: North African progenitor or dead end?
J. Hum. Evol. 39, 393e410.
Jacobs, Z., Roberts, R.G., Nespoulet, R., El Hajraoui, M.A., Debénath, A., 2012. Singlegrain OSL chronologies for Middle Palaeolithic deposits at El Mnasra and El Harhoura 2, Morocco: Implications for Late Pleistocene human environment interactions along the Atlantic coast of northwest Africa. J. Hum. Evol. 62, 377e394.
Jiménez-Espejo, F.J., Martínez-Ruiz, F., Finlayson, C., Paytane, A., Sakamoto, T.,
Ortega-Huertasa, M., Finlayson, G., Iijimaf, K., Gallego-Torres, D., Fa, D., 2007.
Climate forcing and Neanderthal extinction in southern Iberia: insights from a
multiproxy marine record. Quatern. Sci. Rev. 26, 836e852.
Klein, R.G., 1974. Environment and subsistence of prehistoric man in the southern
Cape Province, South Africa. World Archaeol. 5, 249e284.
Lahr, M.M., 1996. The Evolution of Modern Human Cranial Diversity: a Study in
Cranial Variation. Cambridge University Press, Cambridge.
Linstädter, J., Eiwanger, J., Mikdad, A., Weniger, G.-C., 2012. Human occupation of
Northwest Africa: a review of Middle Palaeolithic to Epipalaeolithic sites in
Morocco. Quatern. Int. 274, 158e174.
Lowe, J., Barton, N., Blockley, S., Ramsey, C.B., Cullen, V.L., Davies, W., Gamble, C.,
Grant, K., Hardiman, M., Housley, R., Lane, C.S., Lee, S., Lewis, M., MacLeod, A.,
Menzies, M., Müller, W., Pollard, M., Price, C., Roberts, A.P., Rohling, E.J.,
Satow, C., Smith, V.C., Stringer, C.B., Tomlinson, E.L., White, D., Albert, P.,
Arienzo, I., Barker, G., Bori
c, D., Carandente, A., Civetta, L., Ferrier, C., Guadelli, J.,
Karkanas, P., Koumouzelis, M., Müller, U.C., Orsi, G., Pross, J., Rosi, M., Shalamanov-Korobar, L., Sirakov, N., Tzedakis, P.C., 2012. Volcanic ash layers illuminate the resilience of Neanderthals and early modern humans to natural
hazards. Proc. Natl. Acad. Sci. 109, 13532e13537.
Lubell, D., 2001. Late Pleistocene Early Holocene Maghreb, Africa. In: Peregrine, P.N.,
Ember, M. (Eds.), 2001. Encyclopedia of Prehistory, vol. 1. Kluwer, New York,
pp. 129e149.
Maca-Meyer, N., González, A.M., Pestano, J., Flores, C., Larruga, J.M., Cabrera, V.M.,
2003. Mitochondrial DNA transit between West Asia and North Africa inferred
from U6 phylogeography. BMC Genet. 4, 15.
Mannino, M.A., Thomas, K.D., Leng, M.J., Di Salvo, R., Richards, M.P., 2011. Stuck to the
shore? Investigating prehistoric hunter-gatherer subsistence, mobility and territoriality in a Mediterranean coastal landscape through isotope analyses on marine
mollusc shell carbonates and human bone collagen. Quatern. Int. 244, 88e104.
Mannino, M.A., Catalano, G., Talamo, S., Mannino, G., Di Salvo, R., Schimmenti, V.,
Lalueza-Fox, C., Messina, A., Petruso, D., Caramelli, D., Richards, M.P., Sineo, L.,
2012. Origin and diet of the Prehistoric hunter-gatherers on the Mediterranean
island of Favignana (Egadi Islands, Sicily). Plos One 7, e49802.
Marks, A.E., 1975. The current status of Upper Palaeolithic studies from the Maghreb
to the northern Levant. In: Wendorf, F., Marks, A.E. (Eds.), Problems in Prehistory: North Africa and the Levant. Southern Methodist University Press, Dallas,
pp. 439e458.
Marks, A.E., 2003. Reflections on Levantine Upper Palaeolithic studies: past and
present. In: Goring-Morris, A.N., Belfer-Cohen, A. (Eds.), More than Meets the
Eye. Studies on Upper Palaeolithic Diversity in the Near East. Oxbow Books,
Oxford, pp. 249e264.
McBrearty, S., Brooks, A.S., 2000. The revolution that wasn’t: a new interpretation of
the origin of modern human behavior. J. Hum. Evol. 39, 453e563.
McBurney, C.B.M., 1967. The Haua Fteah (Cyrenaica) and the Stone Age of the SouthEast Mediterranean. Cambridge University Press, Cambridge.
McBurney, C.B.M., 1977. Archaeology and the Homo sapiens sapiens Problem in
Northern Africa. Derde Kroon-Voordracht. Gehouden voor de stichtung
Nederlands Museum voor Anthropologie en Praehistorie, Amsterdam.
Mitchell, P.J., 2002. The Archaeology of Southern Africa. Cambridge University Press,
Cambridge.
Mitchell, P.J., 2008. Developing the archaeology of marine isotope stage 3. S. Afr.
Archaeol. Soc. Goodwin Ser. 10, 52e65.
Mitchell, P.J., Barham, L.S., 2008. The First Africans: African Archaeology from the
Earliest Toolmakers to Most Recent Foragers. Cambridge University Press,
Cambridge.
Moreno, A., 2012. A multiproxy paleoclimate reconstruction over the last 250 kyr
from marine sediments: the northwest African margin and the western Mediterranean sea. In: Hublin, J.-J., McPherron, S.P. (Eds.), Modern Origins: a North
African Perspective. Vertebrate Paleobiology and Paleoanthropology. Springer,
Dordrecht, pp. 3e18.
Moreno, A., Cacho, I., Canals, M., Prins, M.A., Sánchez-Goñi, M.F., Grimalt, J.O.,
Weltje, G.J., 2002. Saharan dust transport and high-latitude glacial climatic
variability: the Alboran Sea record. Quatern. Res. 58, 318e328.
Moreno, A., Cacho, I., Canals, M., Grimalt, J.O., Sánchez-Goñi, M.F., Shackleton, N.,
Sierro, F.J., 2005. Links between marine and atmospheric processes oscillating
on a millennial timescale. A multi-proxy study of the last 50,000 yr from the
Alboran Sea (Western Mediterranean Sea). Quatern. Sci. Rev. 24, 1623e1636.
Neeley, M.P., Barton, C.M., 1994. A new approach to interpreting late Pleistocene
microlith industries in Southwest Asia. Antiquity 68, 275e288.
Olivieri, A., Achilli, A., Pala, M., Battaglia, V., Fornarino, S., Al-Zahery, N., Scozzari, R.,
Cruciani, F., Behar, D.M., Dugoujon, J.-M., Coudray, C., SantachiaraBenerecetti, A.S., Semino, O., Bandelt, H.-J., Torroni, A., 2006. The mtDNA legacy
of the Levantine early Upper Palaeolithic in Africa. Science 314, 1767e1770.
Olszewski, D., Schurmans, U., Schmidt, B., 2011. The Epipaleolithic (Iberomaurusian)
from Grotte des Contrebandiers, Morocco. Afr. Archaeol. Rev. 28, 97e123.
Pallary, P., 1909. Instructions pour la Recherche Préhistorique dans le Nord-Ouest de
l’Afrique. In: Mémoires de la Société Historique Algérienne, vol. 3. Jourdan,
Algiers.
Penaud, A., Eynaud, F., Turon, J.L., Blamart, D., Rossignol, L., Marret, F., LopezMartinez, C., Grimalt, J.O., Malaizé, B., Charlier, K., 2010. Contrasting paleoceanographic conditions off Morocco during Heinrich events (1 and 2) and the
Last Glacial Maximum Original. Quatern. Sci. Rev. 29, 1923e1939.
Pennarun, E., Kivisild, T., Metspalu, E., Metspalu, M., Reisberg, T., Behar, M.D.,
Jones, C.S., Villems, R., 2012. Divorcing the Late Upper Palaeolithic demographic
histories of mtDNA haplogroups M1 and U6 in Africa. BMC Evol. Biol. 12, 234.
Pereira, L., Silva, N.M., Franco-Duarte, R., Fernandes, V., Pereira, J.B., Costa, M.D.,
Martins, H., Soares, P., Behar, D.M., Richards, M.B., Macaulay, V., 2010. Population expansion in the North African Late Pleistocene signalled by mitochondrial
DNA haplogroup U6. BMC Evol. Biol. 10, 390.
Phillips, J., 1972. North Africa, the Nile Valley, and the problem of the Late Palaeolithic. Curr. Anthropol. 13, 587e590.
Raynal, J.-P., 1980. Taforalt. Mission préhistorique et paléontologique française au
Maroc: rapport d’activité pour l’année. Bull. Archéol. Marocaine 12, 69e71.
Richter, D., Moser, J., Nami, M., Eiwanger, J., Mikdad, A., 2010. New chronometric
data from Ifri n’Ammar (Morocco) and the chronostratigraphy of the Middle
Palaeolithic in the Western Maghreb. J. Hum. Evol. 59, 672e679.
Roche, J., 1953. Note préliminaire sur les fouilles de la grotte de Taforalt (Maroc
Oriental). Hespéris 40, 89e116.
Roche, J., 1963. L’Epipaléolithique Marocaine. Fondation Calouste Gulbenkian,
Lisbon.
Roche, J., 1967. L’Aterian de la grotte de Taforalt (Maroc oriental). Bull. Archeol.
Marocaine 7, 11e56.
Roche, J., 1969. Les industries paléolithiques de la grotte de Taforalt (Maroc oriental). Quaternaria 11, 89e100.
Roche, J., 1976. Cadre chronologique de l’Epipaléolithique marocain. In: Actes du IXè
Congrès de l’UISPP: Chronologie et Synchronisme dans la Préhistoire CircumMéditerranéenne, pp. 153e167.
Rodrigo-Gámiz, M., Martínez-Ruiz, F., Jiménez-Espejo, F.J., Gallego-Torres, D., NietoMoreno, V., Romero, O., Ariztegui, D., 2011. Impact of climate variability in the
western Mediterranean during the last 20,000 years: oceanic and atmospheric
responses. Quatern. Sci. Rev. 30, 2018e2034.
Roset, J.-P., Harbi-Riahi, M., 2007. El Akarit: Un Site Archéologique du Paléolithique
Moyen dans le Sud de la Tunisie. Editions Recherches sur les Civilisations, Paris.
Sánchez-Goñi, M.F., 2006. Intéractions végétation-climat au cours des derniers 425
000 ans en Europe occidentale. Le message du pollen des archives marines.
Quaternaire 17, 3e25.
Sánchez-Goñi, M.F., Cacho, I., Turon, J.-L., Guiot, J., Sierro, F.J., Peypouquet, J.-P.,
Grimalt, J.O., Shackleton, N.J., 2002. Synchroneity between marine and terrestrial responses to millennial scale climatic variability during the last glacial
period in the Mediterranean region. Clim. Dyn. 19, 95e105.
Saxon, E.C., Close, A., Cluzel, C., Morse, V., Shackleton, N.J., 1974. Results of recent
excavations at Tamar Hat. Libyca 22, 49e91.
Scally, A., Durbin, R., 2012. Revising the human mutation rate: implications for
understanding human evolution. Nat. Rev. Genet. 13, 745e753.
Schild, R., Wendorf, F., 2010. Late Palaeolithic hunter-gatherers in the Nile Valley
of Nubia and Upper Egypt. In: Garcea, E.E.A. (Ed.), South-Eastern Mediterranean Peoples between 130,000 and 10,000 Years Ago. Oxbow Books, Oxford,
pp. 89e125.
Schwenninger, J.-L., Collcutt, S.N., Barton, R.N.E., Bouzouggar, A., El Hajraoui, M.A.,
Nespoulet, R., Debénath, A., Clark-Balzan, L., 2010. Luminescence chronology for
R.N.E. Barton et al. / Journal of Human Evolution 65 (2013) 266e281
Aterian cave sites on the Atlantic coast of Morocco. In: Garcea, E.E.A. (Ed.),
South-Eastern Mediterranean Peoples between 130,000 and 10,000 Years Ago.
Oxbow Books, Oxford, pp. 18e36.
Sepulchre, P., Ramstein, G., Kageyama, M., Vanhaeren, M., Krinner, G., SánchezGoñi, M.-F., D’Errico, F., 2007. H4 abrupt event and late Neanderthal presence in
Iberia. Earth Planet. Sci. Lett. 258, 283e292.
Staff, R.A., Bronk Ramsey, C., Bryant, C.L., Brock, F., Payne, R.L., Schlolaut, G.,
Marshall, M.H., Brauer, A., Lamb, H.F., Tarasov, P., Yokoyama, E., Haraguchi, T.,
Gotanda, K., Yonenobu, H., Nakagawa, T., Suigetsu 2006 Project Members, 2011.
New 14C determinations from Lake Suigetsu, Japan: 12,000 to 0 cal BP.
Radiocarbon 53, 511e528.
Taylor, V.K., Barton, R.N.E., Bell, M., Bouzouggar, A., Collcutt, S., Black, S., Hogue, J.T.,
2011. The Epipalaeolithic (Iberomaurusian) at Grotte des Pigeons (Taforalt),
Morocco: a preliminary study of the land Mollusca. Quatern. Int. 244, 5e14.
Texier, J.-P., Huxtable, J., Rhodes, E., Miallier, D., Ousmoi, M., 1988. Nouvelles données sur la situation chronologique de l’Atérien du Maroc et leurs implications.
C.R. Acad. Sci. Paris 307, 827e832.
Tixier, J., 1963. Typologie de L’Épipaléolithique du Maghreb. Arts et Métiers
Graphiques, Paris.
Van Peer, P., 2004. Did Middle Stone Age moderns of sub-Saharan African descent
trigger an Upper Palaeolithic revolution in the Lower Nile Valley? Anthropologie 42, 215e226.
Vermeersch, P.M., 1992. The Upper and Late Palaeolithic of northern and
eastern Africa. In: Klees, F., Kuper, R. (Eds.), New Light on the Northeast
281
African Past: Current Prehistoric Research. Heinrich-Barth-Institut, Köln,
pp. 100e153.
Vermeersch, P.M., 2010. Middle and Upper Palaeolithic in the Egyptian Nile valley.
In: Garcea, E.E.A. (Ed.), South-Eastern Mediterranean Peoples between 130,000
and 10,000 Years Ago. Oxbow Books, Oxford, pp. 66e88.
Villa, P., Soriano, S., Tsanova, T., Deganof, I., Higham, T.F.G., d’Errico, F., Backwell, L.,
Lucejko, J.J., Colombini, M.P., Beaumont, P.B., 2012. Border Cave and the beginning
of the Later Stone Age in South Africa. Proc. Natl. Acad. Sci. 109, 13208e13213.
Wadley, L., 1997. Rose Cottage Cave: archaeological work 1987 to 1997. S. Afr. J.
Archaeol. Sci. 93, 439e444.
Ward, S., 2007. Reconstructing Late Pleistocene Vegetation Dynamics from
Archaeological Cave Sites in the Western Mediterranean: Links with Climate
and Cultural Changes. Ph.D. Dissertation. University of Oxford.
Weisrock, A., Wengler, L., Mathieu, J., Ouammou, A., Fontugne, M., Mercier, N.,
Reyss, J.-L., Valladas, H., Guery, P., 2006. Upper Pleistocene comparative OSL, U/
Th and 14C datings of sedimentary sequences and correlative morphodynamical implications in the South-Western Anti-Atlas (Oued Noun, 29 N, Morocco).
Quaternaire 17, 45e59.
Wengler, L., 2010. A new relationship between Mousterian and Aterian in North
Africa. In: Conard, N.J., Delagnes, A. (Eds.), 2010. Settlement Dynamics of the
Middle Palaeolithic and Middle Stone Age, vol. III. Kerns Verlag, Tübingen,
pp. 67e80.
Wrinn, P.J., Rink, W.J., 2003. ESR dating of tooth enamel from Aterian levels at
Mugharet el ‘Aliya (Tangier, Morocco)’. J. Archaeol. Sci. 30, 123e133.