lable at ScienceDirect

Journal of Human Evolution 184 (2023) 103438
Contents lists avai
Journal of Human Evolution

journal homepage: www.elsevier .com/locate/ jhevol
New Blombos Cave evidence supports a multistep evolutionary
scenario for the culturalization of the human body

Francesco d'Errico a, b, *, Karen Loise van Niekerk b, c, Lila Geis a,
Christopher Stuart Henshilwood b, d

a Univ. Bordeaux, UMR CNRS 5199, Bâtiment B2, All�ee Geoffroy Saint Hilaire, CS 50023, F-33615, Pessac Cedex, France
b SFF Centre for Early Sapiens Behaviour (SapienCE), University of Bergen, Bergen, Norway
c School of Geography, Archaeology and Environmental Studies, University of Witwatersrand, Johannesburg, South Africa
d Evolutionary Studies Institute, University of the Witwatersrand, Private Bag 3, Wits, 2050, South Africa
a r t i c l e i n f o

Article history:
Received 22 December 2022
Accepted 3 September 2023
Available online xxx

Keywords::
Middle Stone Age
Still Bay
Symbolism
Beads
Microscopy
Taphonomy
* Corresponding author.
E-mail address: francesco.derrico@u-bordeaux.fr (

https://doi.org/10.1016/j.jhevol.2023.103438
0047-2484/© 2023 Published by Elsevier Ltd.
a b s t r a c t

The emergence of technologies to culturally modify the appearance of the human body is a debated
issue, with earliest evidence consisting of perforated marine shells dated between 140 and 60 ka at
archaeological sites from Africa and western Asia. In this study, we submit unpublished marine and
estuarine gastropods from Blombos Cave Middle Stone Age layers to taxonomic, taphonomic, techno-
logical, and use-wear analyses. We show that unperforated and naturally perforated eye-catching shells
belonging to the species Semicassis zeylanica, Conus tinianus, and another Conus species, possibly Conus
algoensis, were brought to the cave between 100 and 73 ka. At ca. 70 ka, a previously unrecorded marine
gastropod, belonging to the species Tritia ovulata, was perforated by pecking and was worn as an
ornamental object, isolated or in association with numerous intentionally perforated shells of the species
Nassarius kraussianus. Fluctuations in sea level and consequent variations in the site-to-shoreline dis-
tances and landscape modifications during the Middle Stone Age have affected the availability of marine
shells involved in symbolic practices. During the M3 and M2 Lower phases, with a sea level 50 m lower,
the site was approximately 3.5 km away from the coast. In the later M2 Upper and M1 phases, with a sea
level at �60 m, the distance increased to about 5.7 km. By the end of the M1 phase, when the site was
abandoned, Blombos Cave was situated 18e30 km from the shoreline. We use the new Blombos evidence
and a review of the latest findings from Africa and Eurasia to propose a testable ten-step evolutionary
scenario for the culturalization of the human body with roots in the deep past.

© 2023 Published by Elsevier Ltd.
1. Introduction

1.1. Background

The culturalization of the human body refers to the process by
which cultural beliefs, norms, and practices shape and influence the
physical body. It involves the ways in which societies and cultures
impose specific meanings, values, and behaviors onto the body,
thereby transforming it into a symbol and a site of cultural
expression. This process encompasses various aspects, such as body
modifications, adornments, grooming practices, clothing choices,
and bodily rituals, which serve to communicate social identities,
F. d'Errico).
affiliations, and cultural meanings (see Joyce, 2005; Nowell and
Cooke, 2021 for comparable definitions). The culturalization of
the human body highlights the dynamic relationship between
culture and the body, illustrating how cultural forces shape and
mold individuals' physical appearance, behavior, and perception of
self. There is a large body of theory on the cultural construction of
bodies in anthropology, but it is only in the last few decades that
the body has become one of the central theoretical topics of current
sociological, philosophical, and anthropological debate (see Lock,
1993; Joyce, 2005; Bori�c and Robb, 2014 for seminal ideas and
syntheses). Paleolithic archeology, partly because only some of the
behaviors involved in the culturalization of the body leave traces in
the archaeological record, has only incorporated some of the per-
spectives opened up by these theoretical reflections. The question
of the origins of the phenomenon has played a prominent, albeit
not exclusive, role as ochre, and beads have been traditionally seen

mailto:francesco.derrico@u-bordeaux.fr
http://crossmark.crossref.org/dialog/?doi=10.1016/j.jhevol.2023.103438&domain=pdf
www.sciencedirect.com/science/journal/00472484
http://www.elsevier.com/locate/jhevol
https://doi.org/10.1016/j.jhevol.2023.103438
https://doi.org/10.1016/j.jhevol.2023.103438


F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
as signaling, in one way or another, the emergence of symbolically
mediated cultures and typically human cognition (e.g., McBrearty
and Brooks, 2000; d'Errico et al., 2003; Wadley, 2003; Kuhn and
Stiner, 2007; Henshilwood and Dubreuil, 2011; Stiner, 2014; Bar-
Yosef Mayer et al., 2017; Bar-Yosef Mayer and Bosch, 2019; Steele
et al., 2019). Our understanding of the origin and evolution of
technologies for culturalizing our body has been remarkably
refined over the past three decades. However, key questions, that
we attempt to answer here, remain unresolved. Evidence suggest-
ing ancient use of ochre in Africa and Europe indicates that body
painting, clothing decoration, and tattooing may date back to more
than 300 ka (De Lumley et al., 2016; Roebroeks et al., 2012; Watts
et al., 2016; Brooks et al., 2018), possibly 500 ka (Dapschauskas
et al., 2022). However, due to its attested or supposed use in utili-
tarian activities (Wadley et al., 2004; Rifkin, 2011; Soriano et al.,
2015), ochre is seen by many authors as an ambiguous element of
the archaeological record, which cannot be interpreted by, and in
itself, without additional compelling evidence as proof that a cul-
turalization of the body had occurred at times when ochre is first
recorded in the archaeological record. The recent synthesis by
Dapschauskas et al. (2022) identifies three phases of ochre use in
the African Middle Stone Age (MSA), an initial phase from 500 to
330 ka, an ‘emergent’ phase from 330 to 160 ka, and an ‘habitual’
phase from 160 to 40 ka. These authors interpret the latter phase,
when a third of archaeological sites contain ochre as the material
manifestation of intensifying ritual activity in early populations of
Homo sapiens.

Personal ornaments are generally seen as a more robust proxy
for the emergence of the culturalized body and, by extension, a
reflection of cultures comparable to those known ethnographically
(e.g., Ambrose, 1998; Kuhn et al., 2001; Henshilwood et al., 2004;
Vanhaeren and d'Errico, 2006; Bouzouggar et al., 2007; Bar-Yosef
Mayer et al., 2009; d'Errico et al., 2009; d'Errico and Backwell,
2016; Conard, 2010). Their production and use entail the choice
of distinct raw materials, the elaboration of technologies enabling
them to be displayed on the physical body, the modification or
coating of them to change their appearance and color, and the
production of detectable use-wear traces attesting to their use for
ornamental purposes. These suites of choices allow us to better
characterize the degree of technological and cognitive complexity
involved in the implementation of this practice.

This view, however, is not universally shared, and some authors
do not consider personal ornaments as reliable indicator of com-
plex human cognition (Coolidge and Wynn, 2005; Garofoli, 2015).

The discovery, almost two decades ago, of marine shells used as
beads at the southern African MSA site of Blombos Cave, in layers
dated to ca. 73 ka (Henshilwood et al., 2004; d'Errico et al., 2005;
Vanhaeren et al., 2013) was the first to challenge the longstanding
view that ornaments were first displayed on the human body in
Europe, and not earlier than 40 ka. Additional marine shell beads
dated to between ca 142 and 50 ka were identified in subsequent
years at many sites from North Africa, South Africa, East Africa, and
the Mediterranean Levant (Supplementary Online Material [SOM]
Table S1).

In North Africa, the earliest known beads come from Moroccan
sites. The oldest are thirty-two perforated Tritia gibbosula and a
gastropod not figured in the publication, reported as a Columbella
rustica, recently found at Bizmoune Cave, in 142 ka old Aterian
layers (Sehasseh et al., 2021). Twenty perforated T. gibbosula and
three perforated possible Nassarius circumcinctus shells come from
the Grotte des Pigeons in layers attributed to the Aterian and dated
to ~82.5 ka (Bouzouggar et al., 2007; d'Errico et al., 2009). Exca-
vations at Contrebandiers Cave have identified 151 T. gibbosula in
Aterian and Mousterian layers spanning from 115 to 95 ka (d'Errico
et al., 2009; Jacobs et al., 2011; Dibble et al., 2012). At El Mnasra
2

Cave, 221 Tritia sp. and 15 possible N. circumcinctus shells, 177 of
them perforated, were retrieved from three Aterian layers spanning
from ~115 to ~44 ka (El Hajraoui et al., 2012; Jacobs et al., 2012;
Janati-Idrissi et al., 2012; Stoetzel et al., 2014; Campmas et al., 2015,
2016; Ben Arous et al., 2022). Single perforated specimens of
T. gibbosula and C. rusticawere found at Ifri n'Ammar, in an Aterian
layer dated to 83 ka (d'Errico et al., 2009; Richter et al., 2010;
Eiwanger et al., 2012). Five perforated Tritia sp. were recovered in
atypical Aterian layers of El Harhoura 2, dated between 116 and 106
ka (El Hajraoui et al., 2012; Janati-Idrissi et al., 2012; Stoetzel et al.,
2014). Three perforated T. gibbosula, one N. circumcinctus, and one
C. rustica were collected from Aterian layers of Rhafas, probably
dated to between 85 and 76 ka (d'Errico et al., 2009; Doerschner
et al., 2016). The single perforated T. gibbosula shell found at
Oued Djebbana, Bir-el-Ater, the eponymous site of the Aterian,
located in Algeria close to the Tunisian border, comes from a layer
for which we only have a minimum chronological estimate of <30
ka (Morel, 1974a, 1974b; Vanhaeren et al., 2006).

Several sites from the Levant have yielded shell ornaments. Two
perforated T. gibbosula beads were recovered at the Skhul Cave, in
layers dating to 100 or 135 ka (Vanhaeren et al., 2006). At Qafzeh
Cave, layers dating to ca. 100 ka yielded Glycymeris insubrica shells
with natural perforations (Bar-Yosef Mayer et al., 2009; Bar-Yosef
Mayer et al., 2020). Notches on the natural perforations present
on the umbos of these specimens, initially interpreted as due to use
wear produced by suspension of the shells as ornaments, have been
reinterpreted as resulting from natural wear (Cabral and Martins,
2016). Naturally perforated but still undescribed Glycymeris sp.
shells in Mousterian contexts are also present at Ras el Kelb Cave on
the Lebanese coast (Reece, 1998) and in Layer C of Sefunim rock
shelter in Israel (Lamdan, 1984).

In southern Africa, Blombos Cave has yielded 68 perforated
Nassarius kraussianus in Still Bay layers dated at ca. 73 ka
(Henshilwood et al., 2004; d'Errico et al., 2005; Vanhaeren et al.,
2013). Sibudu rock shelter, Kwa-Zulu Natal, located 18 km from
the coast, has yielded twenty-three marine gastropods, nine of
which are perforated. A cluster of perforated Afrolittorina africana
shells, a perforated Mancinella capensis, and some unperforated
shells of both A. africana and M. capensis are found in 70.5 ± 2.0 ka
Still Bay layers. In a Howiesons Poort layer, 64.7 ± 1.9 ka, there is a
single perforated A. africana shell interpreted as dislodged by rock
fall from the underlying Still Bay layers. Two N. kraussianus shells,
one of them perforated and bearing traces of utilization as a bead,
were found in a lateMSA layer dated at 46.6 ka (d'Errico et al., 2008;
Vanhaeren et al., 2019). At Border Cave, Kwa-Zulu Natal, a whole
perforated Conus ebraeuswas found in a pit inwhich the body of an
infant was deposited (d'Errico and Backwell, 2016). Stratigraphic
correlations indicate the pit was excavated in layer 1RGBS, recently
dated to 80e67 ka (Tribolo et al., 2022). A larger perforated shell
belonging to the same species and bearing a similar perforation,
found on a stratigraphic section, is also tentatively attributed to the
same layer (d'Errico and Backwell, 2016). Two perforated
N. kraussianus shells were found in more recent layers, dated at ca.
43 ka.

In East Africa, the recently published record of Panga Ya Saidi,
Kenya (Shipton et al., 2018; d'Errico et al., 2020) identifies the use of
naturally perforated and worn Conus shell spires as beads, possibly
as early as ca. 67 ka, and certainly after 48 ka. At Porc Epic Cave, Dire
Dawa, Ethiopia, 419 naturally perforated land snail opercula of the
species Revoilia guillainopsis, recovered in MSA layers dated at ca.
44 ka, have been interpreted as possible beads (Assefa et al., 2008).

The MSA-to-Later-Stone-Age (LSA) transition in Africa is
apparently marked by a decreased interest in marine shell orna-
ments and the appearance of disk beads, mostly made from ostrich
eggshells (OESs). Eastern Africa has yielded the earliest occurrences



F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
of these beads, with specimens directly dated to ca. 52 ka at Mumba
Rock shelter in Tanzania (Gliganic et al., 2012), >50 ka at Magubike
Rock shelter in southern Tanzania (Miller and Willoughby, 2014),
40.6 ka at Kisese II in central Tanzania (Tryon et al., 2018), 39.9 ka at
Enkapune yaMuto in southern Kenya (Ambrose, 1998), and 33 ka at
Panga Ya Saidi on the Kenyan coast (d'Errico et al., 2020). Border
Cave has yielded amongst the earliest evidence of OES bead use in
southern Africa, directly dated to 43e42 ka (d'Errico et al., 2012). It
has been recently proposed that commonalities in OES bead di-
mensions between eastern and southern Africa broke down
approximately 33 ka, with populations developing separate OES
traditions until herders entered southern Africa after 2 ka (Miller
and Wang, 2022).

Specialized studies of the earliest marine shell beads have
provided information on how they were acquired, modified, and
used. Taphonomic analysis and experimental reproduction of the
modifications recorded on the shells have shown that
N. kraussianus from Blombos Cave were collected alive and
perforated through the aperture with the help of pointed tools,
probably bone awls found in the same layers. T. gibbosula and
Glycymeris sp., found in North Africa and the Levant, were instead
gathered dead on the shores, already bearing natural perforations
that were rarely enlarged with stone tools (d'Errico et al., 2009).
Ochre residues are almost ubiquitous at North African and South
African sites, and their presence on the whole shell surface on
some specimens from Taforalt is interpreted as evidence that at
least in some instances the shells were coated with ochre while
being worn (d'Errico et al., 2009). Experimental data and chemical
analyses indicate that N. kraussianus shell beads from Blombos
Cave were heated in a reductive environment, probably to change
their color (d'Errico et al., 2015). T. gibbosula that appear to have
been blackened by heating also occur at North African sites and on
one specimen from Grotte des Pigeons, traces of use-wear develop
on cracks produced by heating, suggesting the latter took place
before the use of the shell as bead (d'Errico et al., 2009). Use-wear
is recorded on most of the earliest shell beads from Africa and the
Levant (Steele et al., 2019). Experimental reproduction of use-wear
recorded on N. kraussianus shell beads from Blombos Cave has
suggested that a change in the way beads were strung and the
visual appearance of the resulting beadwork occurred at this site
between lower and upper Still Bay layers (Vanhaeren et al., 2013).
The study also shows that these changes were coincident with
environmental shift and the way the space inside the cave was
used. A comparable approach was used to infer the attachment
mode of shells belonging to the same species but from Holocene
layers at Bushman rock shelter (Dayet et al., 2017). A peculiar use
wear pattern, not recorded at Blombos Cave, has been identified
on the N. kraussianus found at Border Cave in 43e42 ka layers
(d'Errico et al., 2012). A new study of the Blombos Cave use-wear
pattern on N. kraussianus, based on data published by Vanhaeren
et al. (2013) but using a quantitative approach to compare
experimental and archaeological wear, confirms that changes in
use-wear patterns correspond to a shift in stringing methods but
proposes a different link between wear and the way beads were
attached (Hatton et al., 2020). This proposition, however, has not
been tested experimentally.

The choice of targeted species, the transport to sites far from the
coast, the prolonged use of the objects, their probable deliberate
coloring with ochre or through heat treatment of shells in reductive
environments (d'Errico et al., 2015), and the similarity of these
practices with those observed in traditional societies are among the
reasons given to argue in favor of a symbolic dimension of these
ancient ornamental objects (e.g., Ambrose, 1998; Bouzouggar et al.,
2007; Bar-Yosef Mayer et al., 2009; Kuhn et al., 2001; Henshilwood
3

et al., 2004; Vanhaeren and d'Errico, 2006; d'Errico et al., 2009;
d'Errico and Backwell, 2016; Conard, 2010).

The use of marine shells as ornaments between ca.140 and 60 ka
at sites in North Africa, South Africa and the Mediterranean Levant,
and the production of OES beads from around 55 ka in different
regions of Africa appear now as established facts. This is especially
true considering that the oldest known ornaments are apparently
the result of practices that were already relatively codified, as
demonstrated by the almost exclusive use of a single species over
vast territories, a constant mode of acquisition, linked to the
collection of either dead or live shells, the almost systematic pres-
ence of ochre on the objects, and thepractice of controlled heatingof
certain pieces to change their color. But this framework still leaves
essential questions unanswered. How did we move from a world
withoutdurable ornaments to one inwhichornamentswerealready
fairly standardized? A recent work proposes, based on evidence
fromMisliya Cave,Mediterranean Levant, that the emergence of the
first ornaments would have been preceded by targeted collection
and transport to sites of shells that were not perforated (Bar-Yosef
Mayer et al., 2020). Previous work has also noted the presence of
unmodified eye-catching marine shells at Pinnacle Point, South Af-
rica, in ca.110 ka and more recent layers, of water-worn, unperfo-
rated Glycymeris connollyi and Phalium labiatum, interpreted as
deliberately collected for esthetic reasons, possibly foreshadowing
the emergence of shell ornaments (Jerardino and Marean, 2010).
Naturally perforated Acanthocardia and Glycemeris, together with
other unperforated marine shells, one of which bearing traces of
ochre, come from the 115 ka site of Los Aviones, and one unperfo-
rated, ochered Pecten from ca. 50 ka levels of Cueva Anton, Spain. In
addition, at the Grotte des Pigeons,Morocco, a site that has yielded a
large collection of naturally perforated shells used as ornaments,
found in layers dated to around 80 ka, several specimens are
unperforated (d'Errico et al., 2009). This indicates, considering the
distance of the site from the sea, that in periods inwhich shells with
natural perforations were used as personal ornaments, non-
perforated shells of the same species were also collected and
transported over distances of more than 50 km. Unperforated shells
are however present in these assemblages in small proportions,
which opens up the possibility that they were collected by accident
or with the intention of perforating them at a later stage.

The evidence from the Levant, North Africa and South Africa
raises the question of the motivations behind a collection of shells
not intended for visual display on the physical body. Are these
practices those from which the earliest bead traditions stemmed?
Another unanswered question is when the production of orna-
ments incorporated technologies such as deliberate coloring with
ochre or controlled heating that allowed a change of their natural
appearance. These techniques, which increase the complexity of
the information that can be conveyed by the ornaments, seem to
have been incorporated into the North and South African shell or-
naments traditions since they first appeared.

One way to complexify or vary the information conveyed by
objects of adornment, while using the same types of objects, is to
change thenumberondisplay, attach them to each other in different
ways, or vary their location on the body. The limited number of
studies conducted so far in this field and the small number of shells
on which the wear is well preserved are an obstacle to recon-
structing the attachment modes of these ancient ornaments and to
understand how they evolved. Another way to complexify infor-
mation conveyed by ornaments is to incorporate different types of
objects in a beadwork. When did the first instances of this cultural
practice occur? Did this change occur at the beginning of the Upper
Paleolithic of Eurasia as is currently assumed, or are the prole-
gomena of this change to be found in earlier periods?



F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
In this article, we build on the analysis of an extended collection
of marine shells from Blombos Cave to address, and contribute to
answer, the questions raised earlier.

We show that eye-catching unperforated marine shells with no
nutritional value were brought to the site between 100 and 77 ka,
well before Blombos inhabitants engaged in the production of
standardized shell beads and that species other than N. kraussianus
were used as beads at the site, at ca. 73 ka, and possibly already at
100 ka. In addition, the discovery of new N. kraussianus shell beads
from Still Bay horizons reveal previously unrecorded use wear
patterns, possibly reflecting additional unidentified modes of
stringing. We propose, based on these and other findings, a
multistep evolutionary scenario for the culturalization of the hu-
man body with roots in the deep past.

1.2. Archaeological context

Blombos Cave (Fig. 1) is situated approximately 300 km east of
Cape Town,west of the village of Still Bay (34�250S, 21�130E). Bounded
by the Duiwenhoks River to the west and the Goukou River to the
east, the site lies within the Blombosfontein Nature Reserve, ca.
100 m from the Indian Ocean rocky shoreline and 34.5 m above the
sea level. The cave opening faces south and was formed by wave
action cutting into the calcified sediments of the Mio-Pliocene
Wankoe Formation (Henshilwood et al., 2001a, 2001b). Initial
investigation of the sitewas undertaken by one of us (C.S.H.) between
1991 and 1992 and is ongoing since 1997. Excavation has, from the
beginning, followed the natural stratigraphy of the deposits and is
done quadrant by quadrant with stratigraphic units identified by
color and texture, defined by their thickness and content, and sub-
sequently labeled as units or subunits. Layer thickness varies be-
tween 2 and 30 cm. Identified layers are given alphabetically ordered
names (CA, CB, CD, CC) starting from the top. Layers with contextual
similarities are attributed a third letter (e.g., CFA, CFB, CFD). All
cultural remains are three-dimensionally plotted with a total station,
and sediment is dry sieved with meshes of 3 mm and 1.5 mm.

The site records a relatively short LSA occupation, radiocarbon
dated to 2000e290 BP, and a longMSA sequence with good organic
preservation. The site's cultural succession is already published
Figure 1. Map of Southern Africa with the location of Blombos Cave (modified from www.m
this figure legend, the reader is referred to the Web version of this article.)

4

extensively elsewhere (d'Errico and Henshilwood, 2007; Villa et al.,
2009; Mourre et al., 2010; Vanhaeren et al., 2013; Douze et al.,
2015; Soriano et al., 2015; Henshilwood et al., 2018; Haaland
et al., 2021). A culturally sterile layer of undisturbed sand (DUN)
overlies the MSA sequence and separates it from the LSA deposits.
TheMSA sequence, fromwhich the artefacts described in this paper
come, consists of four phases. The upper two, BBC M1 (layers
CAeCD) and BBC M2 Upper (layers CDB-CFD), are associated with
the Still Bay techno-complex. These phases contain bifacial foliate
points, a number of which were heated prior to the final shaping by
pressure flaking (Mourre et al., 2010). The Still Bay assemblages
from these layers also include bone awls and spear points
(Henshilwood et al., 2001a; d'Errico and Henshilwood, 2007),
possible engravings of parallel and joining lines on bone (d'Errico
et al., 2001), 68 beads made from N. kraussianus shells
(Henshilwood et al., 2004; d'Errico et al., 2005; Vanhaeren et al.,
2013), eight pieces of ochre engraved with geometric patterns,
two of which display distinct cross-hatched designs (Henshilwood
et al., 2002, 2009), and a criss-cross pattern drawn with an ochre
pencil on a small silcrete flake (Henshilwood et al., 2018).

The M2 Lower phase (CGAA-GCAC) is a low-intensity occupa-
tional horizon with no Still Bayetype artefacts or engraved ochre
(Haaland et al., 2021). The M3 phase (layers CHeCPA) has yielded
abundant unmodified and utilized ochre (Henshilwood et al., 2009;
Watts, 2009), as well as a minimum of sixteen pieces of ochre
engraved with geometric designs (Henshilwood et al., 2009). Two
toolkits consisting of abalone shells containing a pigmented com-
pound, found in close spatial association with modified ochre, seal
bone, and ochre-processing tools, were recovered from this phase
(Henshilwood et al., 2011). All Blombos Cave layers are primarily
composed of sand,with lenses rich in shells, organicmatter, ash, and
recognizable combustion features. The last investigations on the
chronology of Blombos Cave deposits (Jacobs et al., 2020) show that
the sand at the base of the current excavation, conducted inwhat is
provisionally called the pre-M3 phase, started to accumulate
107.8 þ 10.3/�12.1 ka ago. Transition to the first human occupation
of the site occurred at 97.7 þ 4.9/�7.6 ka, followed by M2 lower at
82.5 þ 3.8/�6.2 ka, M2 upper at 76.0 þ 3.3/�5.6 ka, and M1 at
73.7 þ 2.6/�5.2 ka. The MSA deposits ceased to accumulate at
aps-for-free.com by Hans Braxmeier). (For interpretation of the references to color in

http://www.maps-for-free.com


F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
71.0 þ 3.8/�5.7 ka, after which a dune (DUN) sealed the cave until
the late Holocene. This updated chronology indicates that sediment
deposition occurred from MIS 5e through to MIS 4, with human
occupation restricted toMIS 5de5a and the very beginning ofMIS 4.

The numerous paleoenvironmental studies carried out at
Blombos Cave (Henshilwood et al., 2001b; Hillestad Nel, 2007;
Thompson and Henshilwood, 2011, 2014; Langejans et al., 2012a,
2012b, 2013; Badenhorst et al., 2014, 2016; Discamps and
Henshilwood, 2015; Roberts et al., 2016; Nel and Henshilwood,
2016, 2021; Reynard and Henshilwood, 2017, 2019) have pro-
duced an integrated vision of the evolution of the coastal ecosys-
tems to which the MSA populations have adapted. Such a
multiproxy approach highlights relatively stable environmental
conditions during phase M3, contemporaneous with the end of MIS
5c, MIS 5b, and the beginning of MIS 5a. Gradual changes are
recorded toward the end of M3 and through M2, the latter corre-
sponding to the end of MIS 5a. Phase M1, contemporaneous with
the beginning of the MIS 4, represents a period of marked change:
large ungulates become more dominant in this phase, suggesting a
shift to more open grassland (Henshilwood et al., 2001b;
Thompson and Henshilwood, 2011; Discamps and Henshilwood,
2015). Shifts in the micromammal communities support the same
interpretation (Nel and Henshilwood, 2016).

1.3. Availability of estuarine resources through time

Fluctuations in sea level and consequent variations in the site-to-
shoreline distances and landscape modifications during MSA have
affected the availability of habitable land and marine resources,
including marine shells involved in symbolic practices. Over this
period, relative sea-level fluctuated between about �18 m
and�76mbelow thepresent day sea level (Waelbroeck et al., 2002),
exposing approximately 2.3e32.5 km of the now-submerged
Agulhas Bank (Fisher et al., 2010). Sea floor mapping, seismic tran-
sects, and geophysical surveys of the coast in front of BBC (Cawthra
et al., 2020; Jacobs et al., 2020) have allowed a fairly precise recon-
struction of the evolution of the coastline and associated landscapes
during the main phases of Blombos Cave occupation (Fig. 2).

With a sea level 50 m lower than the present (Fig. 2a), which is
representative of the lowest sea level during the M3 and M2 Lower,
dated from 96 to 76 ka, Blombos Cave is, with some fluctuation,
3.5 km away from the coast. The site is separated from the closest
shore byca 2.5 kmof Lower Cretaceous bedrock substrate composed
of siltymudstone and siltstone and1kmof calcareous coastal dunes.
To the east, the estuary of the Goukou River keeps the same distance
from Blombos Cave. To the west, the Duivenhoks River bends to the
east, runs almost parallel to the present coastline, and flows into the
ocean only 10 km from the site, leaving a large band of floodplain
sediment on its east bank.When the sea levelwas at�60m (Fig. 2b),
the situation occurring during the M2 Upper and M1 phase dated
from c 76e71 ka, the coast lay ~5.7 km fromBlombos Cave. The only
substantial change in the geomorphological setting of the site
occurring at this time is the even closer location of the Duivenhoks
River estuary and its muddy riverbanks. With a sea level at �72 m
(Fig. 2c), which occurred at the very end of theM1 phase, at the time
the sitewas abandoned, Blombos Cavewas situated 18e30 km from
the shoreline. At this moment, the advance of the Duivenhoks and
Goukou rivers into the plain created a funnel of land in front of the
site, closed off at its end by a narrow strip of dunes and pebble
beaches. The closest access to the coast as the crowflies, situated in a
south-westerly direction, was made difficult by the streams of the
Breede and Duivenhoks rivers. On the eastern side, access to the
seashore was prevented by the estuary of the Goukou River
advancing southwards. The distance of the site from the estuaries
during the accumulation of M1 phase, the one in which the shell
5

beadswere found, aswell as the very negligible amount offlesh they
contain, their anthropogenic perforation and distinct use wear
patterns (d'Errico et al., 2005), are among the arguments demon-
strating that the shells were not collected for dietary reasons.

2. Materials and methods

Blombosmarine gastropodswere identified at genus and, where
possible, species level using criteria proposed by Branch and Branch
(1981), Kilburn and Rippey (1982), Marais and Seccombe (2010),
Herbert et al. (2018). The taxonomic status of each species was
checked using the WoRMS Editorial Board (Horton et al., 2023).

Identification of traces of manufacture, use-wear, and heating
relied on experimental criteria established by d'Errico et al. (2005,
2015, 2008), Fernandez (2006), Vanhaeren et al. (2013), Gurova
et al. (2013), T�at�a et al. (2014), d'Errico and Backwell (2016),
Langley et al. (2016), Bar-Yosef Mayer et al. (2020) and Gravel-
Miguel et al. (2022). Some of these criteria were specifically
developed for N. kraussianus (d'Errico et al., 2005; d'Errico et al.,
2015; Vanhaeren et al., 2013), the more commonly used species
to make personal ornaments at Blombos Cave during the MSA. We
also paid close attention to recently developed criteria to identify
use wear on shell bead based on the examination of ethnographic
collections of personal ornaments (Falci et al., 2019) and proposed
newmethods to detect residues on shell beads (Collins et al., 2020).
The use-wear categorization proposed by Hatton et al. (2020),
based on Vanhaeren et al. (2013), was also considered when
examining use-wear patterns on N. kraussianus. Identification of
taphonomic damage took into account criteria proposed by
Claassen (1998), Stiner (1999), d'Errico et al. (2005, 2015, 2008),
and Vanhaeren et al. (2019). When necessary for the analysis,
archaeological specimens were cleaned under the microscope with
a wet wooden toothpick and a soft brush. The apertural, abaper-
tural and, when relevant, apical view of each shell was photo-
graphed with a Canon PowerShot 67X. Small specimens were
photographed with a motorized Leica M125 equipped with PLA-
NAPO objective, a DFC295 digital camera, and a Leica Application
Suite v. 4.5 software (Leica, Wetzlar), including the Multifocus
module. The Multifocus module combines multiple digital images
taken at different Z-positions into one single sharp, fully on-focus
composite image. Photos taken with the camera and the micro-
scope served as a base to produce drawings of the various aspects of
each shell with Adobe Illustrator v. 16.0.0 (Adobe Inc., San Jose).
Surface modifications on the shells and perforation edges (perfo-
rations made by predators, damage inflicted by bioeroders, marine
encrustations, use-wear, tool marks, changes of shade and cracks
due to heating, postdepositional damage), and pigment residues
were photographed andmarked on the drawings.Wemeasured the
height and width of the shells and of the perforations and recorded,
when pertinent, the perforation type following the typology pro-
posed by Vanhaeren et al. (2013). Measurements were takenwith a
digital caliper with jaws covered with thin transparent tape to
prevent damage to the shells. The stratigraphic and spatial prove-
nance of the shells were also checked to establish whether they
could be included in groups of beads found in the same or neigh-
boring quadrants and interpreted in previous work as possibly
belonging to the same beadwork (Vanhaeren et al., 2013).

3. Results

3.1. Taxonomic and age class identification

Eighteen new marine gastropods that were used or had the
potential to be used for symbolic purposes were recovered in the
Pleistocene layers of Blombos Cave (Table 1; Fig. 3). They belong to



Figure 2. Evolution of the coastline in front of Blombos Cave. (a) �50 m reconstructed shoreline; (b) �60 m reconstructed shoreline; (c) �72 m reconstructed shoreline. Modified
after Jacobs et al. (2020). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
four species (N. kraussianus, Triviella ovulata, Semicassis zeylanica,
Conus tinianus). An additional Conus, which could be tentatively
attributed to the species Conus algoensis, is too worn for species
identification. N. kraussianus is a scavenging gastropod adapted to
estuarine environments (Branch et al., 1994). It can be found at
present all along the southern African and Mozambican coast, from
6

the estuary of the Orange River up to that of the Ruvama River, at
the boundary between Mozambique and Tanzania. The size and
morphology of the thirteen new complete specimens indicate they
are all adults. The incomplete specimen can also be attributed to an
adult considering its width and thick parietal shield (d'Errico et al.,
2005).



Ta
b
le

1
Ta

xo
n
om

ic
,c
on

te
xt
u
al
,a

n
d
d
es
cr
ip
ti
ve

d
at
a
on

th
e
n
ew

m
ar
in
e
ga

st
ro
p
od

sh
el
ls

fo
u
n
d
in

th
e
M
id
d
le

St
on

e
A
ge

la
ye

rs
of

B
lo
m
bo

s
C
av

e.

N
u
m
be

r
in Fi
gu

re
4

D
at
e
of

ex
ca
va

ti
on

M
SA

p
h
as
e

Le
ve

l
Sq

u
ar
e

qu
ad

ra
n
t

Sp
ec
ie
s

H
ei
gh

t
(m

m
)

W
id
th

(m
m
)

Pe
rf

h
ei
gh

t
(m

m
)

Pe
rf

w
id
th

(m
m
)

Pe
rf

ty
p
ea

U
se
-w

ea
r

lo
ca
ti
on

b
N
ew

u
se
-w

ea
r

O
ch

re
H
ea

ti
n
g

69
14

/1
1/
20

11
M
1

C
A

H
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
17

7.
14

2.
85

2.
31

6
P,

L,
W

,N
ye

s
e

e

70
08

/1
2/
20

11
M
1

C
B

F6
c

N
as
sa
ri
us

kr
au

ss
ia
nu

s
8.
66

7.
04

6.
74

6.
30

8
P,

L,
W

,C
n
o

e
e

71
04

/0
2/
20

19
M
1

C
B

H
7d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
10

.1
8

8.
40

3.
74

5.
70

8
P,

L,
W

,N
n
o

e
e

72
31

/0
3/
20

10
M
1

C
B

G
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
24

7.
22

3.
64

5.
47

8
P,

L,
W

,N
n
o

e
e

73
07

/1
2/
20

11
M
1

C
B

F7
b

Tr
iv
ie
lla

ov
ul
at
a

15
.5
7

12
.2
0

5.
78

5.
77

e
P,

N
,N

,N
e

e
e

74
30

/1
1/
20

11
M
1

C
C

G
7b

N
as
sa
ri
us

kr
au

ss
ia
nu

s
8.
41

7.
84

4.
82

6.
66

8
P,

L,
W

,N
n
o

e
e

75
06

/0
2/
20

19
M
1

C
C

G
7d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
8.
27

7.
39

3.
38

5.
51

8
P,

L,
W

,N
ye

s
e

e

76
c

15
/1
1/
20

11
M
1

C
C

H
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
48

8.
20

3.
69

7.
48

8
P,

L,
W

,N
n
o

e
d

77
30

/1
1/
20

11
M
1

C
C

H
7a

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
50

7.
03

5.
11

6.
75

8
P,

n
a,
W

,N
n
o

e
d

78
07

/1
2/
20

11
M
1

C
C

H
7b

N
as
sa
ri
us

kr
au

ss
ia
nu

s
10

.3
7

8.
07

5.
49

7.
08

8
P,

L,
W

,N
n
o

e
e

79
06

/0
4/
20

10
M
1

C
C
C

G
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
4.
07

7.
87

3.
81

8.
08

8
P,

L,
W

,N
n
o

d
e

80
06

/0
4/
20

10
M
1

C
C
C

G
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
52

7.
80

4.
16

6.
94

8
P,

L,
W

,N
n
o

e
e

81
06

/0
4/
20

10
M
1

C
C
C

G
6d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
9.
68

8.
13

3.
92

7.
33

8
P,

L,
W

,N
n
o

d
e

82
06

/0
4/
20

10
M
1

C
C
C

H
6c

N
as
sa
ri
us

kr
au

ss
ia
nu

s
8.
00

6.
28

3.
52

2.
90

6
P,

L,
W

,N
n
o

e
e

83
15

/0
4/
20

10
M
1

C
D

F6
d

N
as
sa
ri
us

kr
au

ss
ia
nu

s
8.
79

5.
88

3.
24

6.
30

8
P,

L,
W

,C
n
o

e
e

84
17

/1
1/
20

11
M
2

C
FA

H
6d

Se
m
ic
as
si
s
ze
yl
an

ic
a

43
.0
0

39
.6
2

24
.6
4

26
.2
2

e
N

e
e

e

85
22

/1
1/
20

11
M
3

C
IB
h
2

F6
d

Co
nu

s
ti
ni
an

us
24

.8
9

12
.0
4

e
e

e
N

e
e

e

86
30

/1
1/
20

11
M
3

C
N
/

C
O

F6
d

Co
nu

s
sp

.
28

.5
2

16
.0
5

1.
82

1.
69

e
N
,L

,N
,N

e
e

e

A
bb

re
vi
at
io
n
s:

Pe
rf
¼

p
er
fo
ra
ti
on

;M
SA

¼
M
id
d
le

St
on

e
A
ge

;P
¼

p
er
fo
ra
ti
on

ed
ge

;L
¼

lip
;W

¼
p
ar
ie
ta
lw

al
l;
C
¼

co
lu
m
el
la
;n

o
¼

n
on

e;
n
a
¼

n
ot

ap
p
lic

ab
le
;N

ew
u
se

w
ea

r
¼

u
se

w
ea

r
ty
p
e
u
n
re
co

rd
ed

in
p
re
vi
ou

s
p
u
bl
ic
at
io
n
s.

a
Pe

rf
or
at
io
n
ty
p
es

co
rr
es
p
on

d
to

th
os
e
d
efi

n
ed

in
d
'E
rr
ic
o
et

al
.(
20

05
):
Ty

p
e
6
¼

u
n
iq
u
e
m
ed

iu
m

si
ze

p
er
fo
ra
ti
on

lo
ca
te
d
n
ea

r
th
e
lip

;T
yp

e
7
¼
Ty

p
e
6
p
lu
s
sm

al
lp

er
fo
ra
ti
on

s
d
u
e
to

p
os
t-
d
ep

os
it
io
n
al

d
am

ag
e;

Ty
p
e
8
¼
la
rg
e

p
er
fo
ra
ti
on

ex
te
n
d
in
g
fr
om

th
e
lip

to
w
ar
d
s
th
e
ri
gh

t
ed

ge
;
Ty

p
e
9
¼

Ty
p
e
8
p
lu
s
sm

al
l
p
er
fo
ra
ti
on

s
d
u
e
to

p
os
t-
d
ep

os
it
io
n
al

d
am

ag
e.

b
Lo

ca
ti
on

of
u
se
-w

ea
r
on

P,
L,

W
,a

n
d
C
re
sp

ec
ti
ve

ly
.

c
Th

is
sp

ec
im

en
be

lo
n
gs

to
gr
ou

p
7.

d

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
Pr
es
en

ce
of

oc
h
re

or
h
ea

ti
n
g.
7

Triviella ovulata, commonly called smooth pearl cowries or ‘baby's
toes’, is known around the South African coast from the Cape
Peninsula to the Mozambique border in depths of 8e35 m. Triviella
ovulata has a rounded, white to pink shell. When alive, the shell is
covered by a white mantle with black spots. The species lives in
colonies of marine invertebrates on which it feeds. Adults have a
mean height of 20 mm and only exceptionally reach 40 mm. The
size of the single specimen found at Blombos Cave (15.57 mm)
indicates a subadult stage of development.

S. zeylanica, also known as helmet shell, is as the other members
of this family a large gastropod-shaped as a bonnet, which occurs
from the intertidal zone to a depth of 100m. Its present distribution
extends from the Cape of Good Hope to the Mozambique border.
Members of this large family prey on sea urchins by making a hole
through the combined action of an acid secretion and by rasping
the urchin wall with their radula. The smooth form found at
Blombos Cave occurs in rocky areas in the intertidal zone. This
species can reach a height of ca. 80 mm. The width of the incom-
plete specimen found at Blombos Cave points to a young adult.

Conus tinianus (see revision of the taxon in Puillandre et al.,
2015), or variable conus, is a morphologically elongated member
of this vast genus. It occurs at present off the east coast of South
Africa, from Cape Agulhas to the Mozambique border. The size of
the shell varies between 20 mm and 60 mm. The small size of the
specimen identified at Blombos Cave is characteristic of a juvenile
to subadult individual.

3.2. Stratigraphic and spatial distribution

The specimens were recovered in M1, M2 Upper, and M3 phases
(Table 1). The fourteen N. kraussianus and the single Tr. ovulata come
from the M1 phase and the S. zeylanica from the M2 Upper phase.
The two Conus, one identified as Co. tinianus and the other as Conus
sp. or, possibly, Co. algoensis, come from theM3 phase. The Tr. ovulata
was found in layer CB, a thin stratigraphic unit situated at the top of
phase M1, separated from the LSA layers by layer CA and the thick
sterile sand layer called DUN. The attribution of the Tr. ovulata to the
MSA and particularly to the Still Bay technocomplex is however
stratigraphically certain due to the absence of overlying LSA layers in
the area of the cave in which this object was found. It is also sup-
ported by the fact that shell beads made of this species and, more in
general, of Triviidae and Cypraeidae are absent among the personal
ornaments recovered in the LSA layers of the site (Vibe, 2007). Apart
from 290 OES beads and a few bone rings, the LSA layers have
yielded N. kraussianus beads, a few perforated G. connollyi, Bullia
digitalis, Conus sp., and pendants made from Turbo sarmaticus.

The discovery of N. kraussianus in virtually all layers composing
the M1 phase (CA, CB, CC, CCC, CD) confirms the use and accidental
loss of these shell beads by MSA inhabitants throughout the time
span covering the accumulation of this phase (Vanhaeren et al.,
2013), dated at present between ca. 74 and 71 ka (Jacobs et al.,
2020). The reasons why the N. kraussianus discovered in the MSA
levels of Blombos Cave cannot be interpreted as the result of
stratigraphic mixing with the LSA layers overlying the DUN aeolian
sand deposit have already been explained elsewhere (d'Errico et al.,
2005; Henshilwood, 2005; Vanhaeren et al., 2013; Haaland et al.,
2021). They relate to differences in the color and size of the
shells, the type of perforation and use-wear, the shells' spatial
distribution, and relationship with other remains identified in the
M1 phase. The new N. kraussianus exhibit the same features
observed in the shells of the same species from the M1 phase
described in previous studies (see below). Fifty-six of the 68
N. kraussianus shells from the Blombos CaveMSA layers were found
in seven groups of two to 24 beads, each recovered in different
layers and in a single square (1 � 1 m) or in two adjacent



Figure 3. Modern specimens of the novel marine shell species identified in the Middle Stone Age layers of Blombos Cave. (a) Triviella ovulata (height ¼ 18.4 mm); (b) Semicassis
zeylanica (height: 39.3 mm); (c) Conus tinianus (height: 35.7 mm). Photos courtesy of Marcus Lussi. (For interpretation of the references to color in this figure legend, the reader is
referred to the Web version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
subsquares (50 � 50 cm) during the same excavation day. The
finding that these clusters are associated with changes through
time in the arrangement of the living space within the cave and
differences in the size of the perforations and the location of use
wear traces led the authors to propose that shells from each cluster
groupmay have originally belonged to a single beadwork item, lost,
or disposed during a single event. Three N. kraussianus from the
enlarged collection described here may represent a new group
Figure 4. Marine shells from Blombos Cave Middle Stone Age layers with indication of th
73 ¼ Triviella ovulata; 84 ¼ Semicassis zeylanica; 85 ¼ Conus tinianus; 86 ¼ Conus sp. Scal
interpretation of the references to color in this figure legend, the reader is referred to the

8

(specimens 79e81; Table 1 and Fig. 4). They all originate from layer
CCC, from the same subsquare and were recovered during the same
excavation day. The S. zeylanica comes from a layer (CFA) situated at
the top of the M2 phase, dated at ca. 74 ka. The Co. tinianus and
Conus sp. were recovered from layers CIBh2 and CN/CO, located at
the middle and the bottom of the M3 phase. Although these layers
were not directly dated by optically stimulated luminescence (OSL),
the age of layer CIBh2 can accurately be gauged as lying between
e phase and layer of provenance. Specimens: 69e72, 74e83 ¼ Nassarius kraussianus;
e bar ¼ 1 cm. See Table 1 for additional contextual and descriptive information. (For
Web version of this article.)



F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
the age of the overlying layer CIBh1 (83 ka), and the underlying
layer CJ (93.2 ka). Layer CN/COwould be close to ca. 94.9 ka, the age
of layer CO.
3.3. Taphonomic, technological, and use-wear analysis

The state of preservationof thenewN. kraussianus fromM1(Fig. 4)
does not differ substantially from that observed on the other
N. kraussianus recovered from the same phase. All the shells are
perforated on the dorsal side. Most of them have a lightedark brown
color, typical of theN. kraussianus beads from the Blombos CaveMSA
layers and different from the bright yellowish-white color of those
found in the LSA layers. Two are dark brown, and one has micro-
fractures andflakingof thenacre on the ventral side (Fig. 5) indicating
prolonged heating in a reductive environment (d'Errico et al., 2015).
The unburnt specimens have well-preserved surfaces with only two
showing a small, elongated perforation on the spire (specimens 70
and 81; Table 1; Fig. 4). None display predator perforations on their
ventral sides, a feature common in estuarine thanatocoenoses of this
species, or porous surface textures produced by gradual degradation
of the shell in these environments (d'Errico et al., 2005). Dead
N.kraussianus fromestuarieswithperforationson thedorsal side tend
tobeheavilyetchedand fragile,which isnot thecasewith those found
atBlombosCave. Thearchaeological specimensalsodonotdisplay the
features recently identified on N. kraussianus shells washed up on
beachesandmodifiedbywaveaction,whichareheavily roundedover
all surfaces and display microimpacts (Vanhaeren et al., 2019).
Naturally perforated N. kraussianus found on beaches display an
advanced stage of roundingobliterating all outer anatomical features.
Theyoften alsohave small gravel pieces stuck in the shell aperture. All
these features are absent from the Blombos CaveMSAN. kraussianus.
The aforementioned evidence confirms, as previously demonstrated,
that atBlombosCave, theN. kraussianuswere collected alive andwere
later deliberately perforated (d'Errico et al., 2005).

The Tr. ovulata from the M1 phase is well preserved apart from a
large perforation on the dorsal side and a spot of decalcification
producing a 200-mm elongated hole next to the perforation
Figure 5. Dark brown Nassarius kraussianus from Blombos Cave M1 phase, layer CC, disp
prolonged heating in a reductive environment. (For interpretation of the references to colo

9

(Fig. 6a). It was almost certainly collected dead on or close to the
shore, considering the depth at which this species lives (8e35 m;
Branch and Branch, 1981).

The S. zeylanica from the M2 Upper phase is missing the spire
(specimen 84; Fig. 4). Its surface has a matt appearance reminiscent
of shells washed on the shore. The presence of perforations caused
by bioeroders inside the shell confirms it was collected dead.

The Co. tinianus from the M3 phase shows differential preser-
vation from the apex to the anterior canal (Fig. 7). The spire is
heavily eroded and is covered by micro perforations produced by
bio-eroders, while the remainder of the shell surface is well pre-
served. Although the heavy erosion of the spire may suggest
collection of a dead animal, living Conus may occasionally have
eroded spires. The irregular outline of the lip, due to numerous
microflake scars, is likely due to the lip of this species being thin
and fragile, thus prone to natural breakage fromwave action on the
shore or trampling at the cave.

The Conus sp (which might be Co. algoensis) from the M3 phase
is homogenously smoothed by wave action (specimen 86; Fig. 4).
The apex is missing, leaving a large circular perforation at the
center of the spire cap (Fig. 8a, b). A second, smaller perforation on
the body whorl is likely due to drilling by a predator and subse-
quent erosion (Fig. 8c). A round piece of beach gravel is stuck in the
shell aperture (specimen 86; Fig. 4). Overall, these features support
the collection of a dead shell.
3.4. Modification

The perforations on the dorsal side of the new N. kraussianus
shells fall in types 6 and 8 perforation types identified in natural
modern and archaeological collections (Figs. 4 and 9), i.e., a small
perforation close to the lip and a large elliptical perforation
removing a large portion of the dorsal side, respectively (d'Errico
et al., 2005). It has been shown that these types are rare in estua-
rine and shore thanatocoenoses and when present are associated
with advanced alteration and decoloration of the shell's dorsal side.
In contrast, types 6 and 8 account for almost 90% of the perforations
laying on its ventral face cracks and flaking of the nacre interpreted as evidence of
r in this figure legend, the reader is referred to the Web version of this article.)



Figure 6. Close-up views of the Triviella ovalata from M1 phase, layer CB. (a) perfo-
ration displaying smoothed crenulated edge on the right side and sinuous smooth edge
with a notch on the left side. Notice a spot of decalcification close to the top left edge of
the perforation associated with a small hole; (b) top right edge of the perforation
showing numerous pits produced by pecking the shell surface. Arrows indicate pits of
same morphology and orientation produced by the same tool; (c) striations close to the
bottom of the perforation edge produced by a lithic tool, possibly to regularize the
edge after its perforation and enlargement by pecking. (For interpretation of the ref-
erences to color in this figure legend, the reader is referred to the Web version of this
article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
on the N. kraussianus from the Blombos Cave MSA layers
(Vanhaeren et al., 2013), which leads to the interpretation of this
modification as resulting from human agency, a hypothesis sup-
ported by experimental piercing of the shells with a pointed bone
tool through the aperture, the otherwise good state of preservation
10
of the archaeological shells and the presence of use wear patterns
being absent on shells from natural thanatocoenoses (see below).
As in some specimens described in previous studies, ochre residues
were detected in two of the new N. kraussianus beads analyzed in
this study (Table 1).

The Tr. ovulata bears on its dorsal aspect a large irregular
perforation located in the middle of the body whorl. Diagnostic
traces of modification occur around the top right and the lower
margin of the perforation. In the first area, they take the form of
numerous micro pits produced by repeatedly pecking the shell
surface with a pointed lithic tool (Fig. 6b). Several pits show the
same morphology and orientation, indicating that the pecking was
executed with rapidly repeated gestures using the same tool
(Fig. 6b). Pecking was probably the technique used to perforate the
shell. The crenulated morphology of the perforation's right edge
and the presence of pecking pits relatively far from the edge indi-
cate that this technique was certainly used to enlarge the perfo-
ration. The reason for perforating the body whorl by pecking
instead of by puncturing it through the aperture, the technique
used for perforating the N. kraussianus, is certainly related to the
aperture narrowness and the robustness of the body whorl of this
species, whichmakes puncturing through the aperture challenging.
In the second area, traces of modifications consist of several recti-
linear or slightly curved striations parallel to the perforation
outline, left by a lithic point (Fig. 6c). These striations probably
result from the attempt to regularize to some extent the irregular
perforation edge left by pecking.

No apparent traces of anthropogenic modifications were recor-
ded on the S. zeylanica and Co. tinianus apart from the presence of
two microscopic spots of ochre on the spire of the latter, which can
be due to accidental contact with ochre pieces or ochre containing
sediment in a layer rich in this material (Fig. 7b). The lip of the
Conus sp. bears an elongated facet that removes the very edge of it,
exposing the inner aragonitic layers. This facet may have been
produced by longitudinally scraping the lip with a robust lithic
edge. Superficial striations left by this action are barely visible on
the inner lip surface (Fig. 8d, e).

3.5. Use-wear

Use-wear on the N. kraussianus shells from Blombos Cave is
recorded around the perforation on the dorsal aspect in the form of
smoothing and polish, and on the lip, parietal whorl and the
columella in the form of facets (Table 1; Figs. 9 and 10). This use-
wear is not always present at all these locations and does not al-
ways extend over their entire surface. Most of the wear patterns
identified on the new N. kraussianus are the same recorded on
Blombos Cave MSA shell beads previously and confirm recorded
differences between layers such as the noticeable tendency of
almost all N. kraussianus from lower levels CC, CCC, and CD to bear
larger perforations and more invasive facets on the lip and parietal
whorl (Fig. 9). However, two specimens (specimens 69 and 75;
Figs. 9 and 11; Table 1) from layers CA and CC show previously
unrecorded wear patterns from those layers.

The first displays two adjacent facets on the parietal whorl,
not found on any of the shells from CA. The second displays an
elongated wear facet on the left side of the body whorl, on the
shell's ventral side. The perforation edge of the Tr. ovulata spec-
imen is heavily smoothed by use-wear, certainly produced by a
thread, except for a portion of the top right edge, displaying a
fracture that has removed a thin band of the worn edge (Fig. 6a).
The bottom right edge displays use wear that is slightly less
developed than that present on the left edge, suggesting the
latter was the outline portion on which friction from a stringing
thread was exerting more intense attrition. The edge of the



Figure 7. a) Conus tinianus from the Blombos M3 phase showing differential preservation from the apex to the anterior canal. (b) Close-up view showing erosion and micro
perforations produced by bio-eroders as well as a spot of ochre (arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version
of this article.)

Figure 8. Conus sp. from Blombos M3 phase. (a), (b) Natural hole with smoothed edge at the centre of the spire cap produced by erosion of the apex; (c) Small perforation on the
body whorl interpreted as due to a predator and subsequent erosion; (d), (e) Facet flattening the lip and exposing inner aragonitic layers, possibly produced by scraping the lip. (For
interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
perforation on the apex of the Conus sp. is heavily worn and is
possibly smoother than the surrounding surface of the spire
(Fig. 8b). It is difficult to establish whether this use wear should
11
be attributed to beach abrasion, transport, and curation, wear
produced by a thread, post depositional processes or a combi-
nation of these causes.



Figure 9. Drawings of marine shells from Blombos Cave Middle Stone Age layers with indication of the phase and layer of provenance. Use-wear areas are highlighted in gray. Dots
and subparallel horizontal lines in specimen 73 identify the location of pecking and incising respectively; criss-cross red pattern in specimen 85 identifies red staining and spots of
ochre; parallel striations on the outer lip of specimen 86 circumscribes possible scraping marks. See Table 1 for additional contextual and descriptive information.

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
4. Discussion

4.1. Expanding the Blombos Cave personal ornament record

Although the sample size from which we have derived our in-
ferences is still small, the analysis of the new marine shells
discovered at Blombos Cave sheds new light on the origins of
practices aimed at the culturalization of the human body. Our study
reveals that between 100 and 77 ka, visually striking marine gas-
tropods belonging to the genus Conus, one of which bears a natural
perforation, were collected on the coast, which at that time was
between 2 and 3 km away, and brought to the site. The perforated
Conus with the potentially modified lip may have been hung and
used as an ornamental object. Between 77 and 73 ka when the site
was ca. 10 km from the coast, a Semicassis shell was brought to the
site. The state of preservation of this object makes it difficult to
know whether the spire was naturally perforated when the object
was collected or was intentionally modified for use as an orna-
mental object. It was not until approximately 73 ka, when the site
was between 20 and 30 km from the sea, and access to the coast
was made difficult by several rivers and estuaries, that the in-
habitants of Blombos Cave developed ornamental items using
numerous specimens, collected live, of one species, N. kraussianus,
which was available in marshy estuaries between 20 and 40 km
from the site. These shells were defleshed, systematically perfo-
rated, deliberately or accidently colored with ochre, sometimes
heated to change their color, and attached to one another creating
designs that changed over time. New wear patterns identified in
the framework of this study and work conducted at other sites
(Dayet et al., 2017) indicate that further experimentation would
contribute to a better understanding of the evolution of these
designs.

Contrary to what was previously considered as definitively
established, i.e., that the Still Bay inhabitants of Blombos Cave used
12
only one species of estuarine shell to assemble their ornaments, our
study shows that toward the end of the Still Bay occupation at least
one other, but in this case a marine gastropod, Tr. ovulata, was also
used as an ornamental object. To collect this dead shell from a
beach, the Blombos Cave inhabitants had towalk ca. 40 km, or cross
two now-submerged rivers, which would have shortened their trip
by about 10 km. In either case, it is a species that was more difficult
to acquire than N. kraussianus. We cannot gaugewhether the object
wasworn alone, as a pendant, or was integrated into an ornamental
device in which it was associated with N. kraussianus. In the latter
case, it is likely, considering its rarity in relation to the number of
N. kraussianus found so far in the upper layers of the M1 phase, that
this object would have had a prominent and visible position in the
adornment.

4.2. A multistep evolutionary scenario

The new data presented here, and the wealth of information
accumulated over the last decade on the first practices aimed at the
culturalization of the body clearly indicate that, contrary towhatwas
believed until a couple of decades ago, these practices did not sud-
denly manifest themselves with the complexity that we recognize
today inmany human cultures. Instead, they seem to have developed
in stages whose significance and implications we are only now
beginning to grasp.

Two new types of evidence from Blombos Cave contribute to the
identification of these stages. The discovery of unmodified and
naturally perforated shells from 100 to 73 ka confirm previous
scant evidence that marine shells were collected, taken to the site
and, in some cases, perhaps worn as personal ornaments before a
stage in which shells belonging to selected species were system-
atically, and intentionally perforated with suitable techniques and
associated to create composite beadworks. The new evidence from
this site also highlights that at the end of a stage during which a



Figure 10. Nassarius kraussianus beads from layers CC (a) and CD (b) of Blombos M1 Phase showing enlargement and intense smoothing of the perforation on the dorsal side and
invasive use-wear facets on the parietal whorl (arrows). (a) Specimen 74; (b) specimen 83. See also Figures 4 and 9 and Table 1. (For interpretation of the references to color in this
figure legend, the reader is referred to the Web version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
single species is perforated and worn as ornaments, a new species
was introduced, arguably leading to the complexification of the
beadworks and the symbolic code associated with them. This new
information is precious considering the relatively low number of
well-dated sites that record variation in shell use between 140 and
50 ka. These practices tell us about the technological evolution of
populations in the past but, above all, about the growing role they
must have assumed in increasingly complex social interactions
(Joyce, 2005; Moro Abadía and Nowell, 2015; Nowell and Cooke,
2021). Any major change in these practices is likely to reflect
changes in these interactions, both within and across human
groups. Currently, we observe at least ten stages in the evolution of
13
these practices (Fig. 12). These stages should not be seen as a uni-
linear and inevitable process. Independent invention and cultural
transfer have undoubtedly been mechanisms for skipping some of
these stages, and in some cases, we can even envisage a reversal of
this process. In proposing these stages, we have confined ourselves
to considering technologies for the culturalization of the body that
leave archaeological traces of themselves. This choice enables us to
rely on robust and dated data, but it is probable that these practices
were complemented, probably very early on, by a panoply of other
practices expressed through gestures, dance, mutilation, scarifica-
tion, tattooing, and clothing, which have left until recently only
circumstantial, ambiguous, or episodic evidence of themselves



Figure 11. Nassarius kraussianus beads from layer CA (a) and CB (b) of Blombos M1 Phase showing (arrows) a wear pattern previously unrecorded in layer CA (a) and in the M1
phase (b). (a) Specimen 69; (b) specimen 75. See also Figures 2 and 7 and Table 1. (For interpretation of the references to color in this figure legend, the reader is referred to the Web
version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
(d'Errico et al., 2018; Garfinkel, 2018; Nowell and Cooke, 2021;
Doyon et al., 2023). Also, these ten steps should be seen as an
interpretive framework to be tested against future discoveries.

The first stage probably involved the use of mineral pigments for
body decoration. The exponential increase in the use of ochreduring
the MSA of southern Africa indicates that while ochre may have
served primarily utilitarian needs during its first uses at the begin-
ning of theMSA, around 500e300 ka, symbolic functions, including
its use for body decoration, must have become predominant after
160 ka, as convincingly argued recently by Dapschauskas and
colleagues (2022). This first stage is, for the moment, mainly attes-
ted in sub-Saharan Africa and to a lesser extent in Europe (Rosso,
2022). The transport and use of high-quality red ochre from
14
distant sources is consistent with this view. Pigment use certainly
continued in later periods, and it cannot be ruled out that objects,
including adornments, made of perishable materials accompanied
these early instances of body culturalization.

A second stage, so far attested in southern Africa, involved the
collection and curation of marine shells and perhaps other natural
objects without modifying them (Jerardino and Marean, 2010; Bar-
Yosef Mayer et al., 2020; Hoffmann et al., 2018; this study). The fact
that these are unique specimens and that some of them do not bear
natural or anthropic perforations suggests that these objects may
have been used initially as amulets or similar devices, not neces-
sarily intended to be worn visibly. This phase may have coincided
with or been followed by the use of shells or other objects with



Figure 12. A tentative multistep evolutionary scenario for the culturalization of the human body. The scenario includes evidence from both Africa and Eurasia. Thicker lines indicate
continuous and widespread use of the relevant body culturalization technology in large regions. The tapering from thick to thin lines indicates a gradual reduction in the practice.
Thinner lines indicate continuous, less intensive use. Dashed lines indicate periods during which we may reasonably assume that the practice was implemented although the record
has not been investigated in detail. See text and SOM Table S1 for supporting evidence. (For interpretation of the references to color in this figure legend, the reader is referred to the
Web version of this article.)

F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
natural perforations or features allowing them to be attached. This
practice was apparently used at sites such as Blombos Cave in
southern Africa (this study), Qafzeh in the western Asia (Cabral and
Martins, 2016; Bar-Yosef Mayer et al., 2020), or Cueva de los Aviones
in Europe (Zilh~ao et al., 2010; Hoffmann et al., 2018) between 150
and 80 ka. These studies provide evidence that already in this
phase, ochre was applied deliberately to the shells or was trans-
ferred accidently between ochered skin and the objects.

The third stage consisted of assembling several homologous
objects with natural or slightly modified perforations, for example
shells belonging to the same species or to species that are
morphologically very close, into an ornamental device. According
to the latest findings, this practice seems to have been present in
north-western Africa since about 140 ka and to have continued in
this region until 80 ka (Bouzouggar et al., 2007; d'Errico et al., 2009;
Steele et al., 2019; Sehasseh et al., 2021). This practice appears to be
associated with a more systematic use of ochre and dedicated
heating techniques to modify the color of the shells. The exclusive
or almost exclusive use of natural perforations limits the possibil-
ities of varying the arrangement of the shells.

Anext step entailed the implementationofperforation techniques
that allowed for variations in the arrangement of homologous orna-
mentsandmoreeasilyproducingdifferentornamental itemswith the
same shells. This practice is attested at Sibudu rock shelter from80ka
and at Blombos Cave between 73 and 70 ka (Vanhaeren et al., 2013,
2019).

The fifth step entailed integrating a larger shell of another
species into an ornamental item mostly composed of smaller shells
or to present the two types of objects separately. This practice is
attested at Blombos Cave at the end of Phase M1, dated to around
72 ka (this study), and at approximately the same time at Sibudu
(Vanhaeren et al., 2019). A possible explanation for the scarcity of
instances of this step in the archaeological record could be linked to
the significantly lower occurrence of large shells in these orna-
mental devices than smaller shells. This discrepancy diminishes the
likelihood of their discovery during excavation efforts.

A further step was the use of fragments of shells heavily
modified and perforated by natural agents as ornaments. This stage
is so far only known from Panga Ya Saidi, Kenya, where naturally
perforated Conus spires were used as beads (d'Errico et al., 2020).
The naturally perforated opercula from Porc Epic may represent an
additional instance of this stage. This practice opened the way to
15
the following, seventh, step in the evolution of body culturaliza-
tion: the production from around 50 ka, probably first in East Africa,
of fully manufactured circular beads made of OESs (d'Errico et al.,
2012; Miller and Willoughby, 2014; Tryon et al., 2018). The latest
research on these objects shows a regionalization, expressed by a
difference in the size of the beads from 33 ka (Miller and Wang,
2022). A much greater diversification marks the next, eighth,
evolutionary step from around 45 ka, which took the form, in large
regions of Eurasia, of an explosion of ornament types in different
raw materials, shapes, modes of production, and, no doubt,
differentmodes of attachment to the body (Vanhaeren and d'Errico,
2006; Fern�andez and Joris, 2008; Langley et al., 2019; d'Errico et al.,
2021; Rybin, 2015; Wolf, 2015). This wealth of ornament types is
fully comparable in terms of types richness to the most elaborate
ornament traditions known in more recent prehistoric and histor-
ical societies (Dubin, 2009; Rigaud et al., 2015; McAdam and
Davidson, 2018). The geographic distribution of these highly
diversified Paleolithic ornaments suggests their use as ethnic
markers (Vanhaeren and d'Errico, 2006; d'Errico and Vanhaeren,
2015; Wolf, 2015). The use of exotic objects, obtained through
long-distance trade, and their presence in quantity in some
Paleolithic burials and not in others, suggests the use of ornaments,
already in these societies, as markers of social inequality
(Vanhaeren and d'Errico, 2005).

The use of ornaments, and other practices of body culturaliza-
tion for this purpose, was exacerbated by the ninth step: the pro-
duction by specialized craftsmen of ornaments intended for
members of other social groups following the advent of production
economies and accumulation of wealth (Tite et al., 2002; Dubin,
2009; Nicholson, 2012).

The final tenth step corresponds to the industrial production of
ornaments, which allows to a large extent all members of a society
to wear any kind of adornament, including beads to which special
values were attributed in previous steps either because they were
made of rare raw materials or because their production required
special skills (Razazadeh,1997; Dubin, 2009; Tomalin, 2016). This is
achieved by the industrial production of beads made of synthetic
gemstones, pearls, amber, and the like (Shigley, 2000).

The logic that seems to have governed this step-by-step process
from the beginning is that of a gradual increase in the complexity of
the technologies used to culturalize the body and in the time and
effort invested in the acquisition and transformation of raw



F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
materials. In themost recent phases, this effort led to the use of rare
or exotic materials and the rise of artisans with specialized
knowledge. This process led, on the one hand, to the development
of increasingly complex visual codes together with the diversifi-
cation of the ornament's functions and, on the other hand, to the
transformation of ornaments into valuable goods to be kept,
exchanged, or inherited. This required the creation, in each culture,
of dedicated cognitive functions and neural circuits allowing the
increasing complexity of the messages conveyed by ornaments to
be deciphered and novel values to be attributed to these objects
and their association on the body (Salagnon et al., 2023). It also
required the creation of means to precisely quantify this wealth.
This step-by-step process is also reflected by the increasing artifi-
cialization of ornament shapes: from unmodified natural items,
humans moved to natural objects transformed by perforating,
coloring, and arranging them in different ways. This was followed
by a shift to entirely artificial geometric shapes and colors in the
following periods and to a multiplication and standardization of
these shapes and colors, without necessarily abandoning the nat-
ural objects used in previous stages. It is likely that the process we
have just identified, although limited to personal ornaments and
the use of ochre, also informs us about the evolution of other
practices for culturalizing the body. We can anticipate that these
practices, too, have progressed toward increased complexity in
terms of techniques, symbolic codes, and the involvement of
specialized individuals to ensure their continuity and, if needed,
further elaborate development.

It is also clear that these practices have certainly co-evolved and
been influenced in their development by other cultural in-
novations, not originally designed to be applied to the body.
Although it is difficult to identify a single cause behind this process,
it is tempting to interpret these evolutionary steps as reflecting
both a long-term human niche construction (Kendal et al., 2011;
Laland et al., 2016) and a form of self-domestication (Wrangham,
2019), which, by relying on increasingly sophisticated means of
communication and strategies of cultural transmission, has led
human societies to create environments in which the human body
becomes the means of displaying oneself in the society, of con-
structing oneself as a person, and of reading these multiple iden-
tities in others.

5. Conclusion

In summary, our unique ability to culturalize our body has deep
roots. The available archaeological evidence and the new data
provided by the Blombos Cave material indicate that humans
gradually complexified these practices and transformed them into
tools for communication, storage of information, and creation of
wealth that gradually but radically changed the way we look at
ourselves and others, and the nature of our societies. We propose,
for the first time, an evolutionary scenario for the culturalization of
the human body that, although only relying on technologies that
have left traces on the archaeological record, identify trends that
probably shed light on the evolution of ephemeral practices. This
scenario starts 500 ka and identifies major tipping points at 140 ka,
70 ka, and 45 ka, followed by an acceleration corresponding to the
emergence of production economies. Future research should focus
on better characterizing the early phases of this evolutionary pro-
cess, the mechanisms that may have triggered tipping points, and
their technological, social and cognitive implications for the soci-
eties involved.

Declaration of competing interest

None.
16
Acknowledgments

We thank two anonymous reviewers and Manuel Will for
constructive comments on a first version of themanuscript, Markus
Lussi for his advice on the taxonomic assignment of some speci-
mens, and Samantha Mienies for her curatorial assistance. All au-
thors acknowledge support from the Research Council of Norway
through its Centre's of Excellence funding scheme (SFF Centre for
Early Sapiens Behavioure SapienCEe project number 262618). F.D.
research was also supported by the Talents programme of the
University of Bordeaux IdEx, (grant number: 191022_001), the
Grand Programme de Recherche ‘Human Past’ of the Initiative
d'Excellence (IdEx) of the University of Bordeaux, and the European
Research Council under the Horizon 2020 program (QUANTA
project, contract no. 951388). C.S.H. acknowledges support from
National Research Foundation/Department of Science and Tech-
nology funded SARChI Chair at the University of theWitwatersrand,
South Africa.
Appendix A. Supplementary Online Material

Supplementary Online Material to this article can be found
online at https://doi.org/10.1016/j.jhevol.2023.103438.
References

Ambrose, S.H., 1998. Chronology of the Later Stone Age and food production in East
Africa. J. Archaeol. Sci. 25, 377e392. https://doi.org/10.1006/jasc.1997.0277.

Assefa, Z., Lam, Y., Mienis, H.K., 2008. Symbolic use of terrestrial gastropod opercula
during the Middle Stone Age at Porc-Epic Cave, Ethiopia. Curr. Anthropol. 49,
746e756. https://doi.org/10.1086/589509.

Badenhorst, S., Van Niekerk, K.L., Henshilwood, C., 2014. Rock hyraxes (Procavia
capensis) from Middle Stone Age levels at Blombos Cave, South Africa. Afr.
Archaeol. Rev. 31, 25e43. https://doi.org/10.1007/s10437-014-9154-7.

Badenhorst, S., Van Niekerk, K.L., Henshilwood, C.S., 2016. Large mammal remains
from the 100 ka Middle Stone Age layers of Blombos Cave, South Africa. S. Afr.
Archaeol. Bull. 71, 46e52.

Bar-Yosef Mayer, D.E., Vandermeersch, B., Bar-Yosef, O., 2009. Shells and ochre in
Middle Paleolithic Qafzeh Cave, Israel: Indications for modern behavior. J. Hum.
Evol. 56, 307e314. https://doi.org/10.1016/j.jhevol.2008.10.005.

Bar-Yosef Mayer, D., Bonsall, C., Choyke, A.M., 2017. Not Just for Show: The
Archaeology of Beads, Beadwork, and Personal Ornaments. Oxbow Books
Limited, Oxford.

Bar-Yosef Mayer, D.E., Bosch, M., 2019. Personal ornaments in early Prehistory.
Humans' earliest personal ornaments: An Introduction. PaleoAnthropology
2019, 19e23. https://doi.org/10.4207/PA.2019.ART121.

Bar-Yosef Mayer, D.E., Groman-Yaroslavski, I., Bar-Yosef, O., Hershkovitz, I., Kampen-
Hasday, A., Vandermeersch, B., Zaidner, Y., Weinstein-Evron, M., 2020. On holes
and strings: Earliest displays of human adornment in the Middle Palaeolithic.
PLoS One 15, e0234924. https://doi.org/10.1371/journal.pone.0234924.

Ben Arous, E., Philippe, A., Shao, Q., Richter, D., Lenoble, A., Mercier, N., Richard, M.,
Stoetzel, E., Tombret, O., El Hajraoui, M.A., Nespoulet, R., Falgu�eres, C., 2022. An
improved chronology for the Middle Stone Age at El Mnasra cave, Morocco.
PLoS One 17, e0261282. https://doi.org/10.1371/journal.pone.0261282.

Bori�c, D., Robb, J., 2014. Body theory in archaeology. In: Boric, D., Robb, J. (Eds.), Past
Bodies: Body-centered Research in Archaeology. Oxbow Books, Oxford, pp. 1e7.

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. USA 104. https://doi.org/10.1073/pnas.0703877104, 9964e9869.

Branch, G., Branch, M., 1981. The Living Shores of Southern Africa. Struik, Cape
Town.

Branch, G., Griffiths, C., Branch, M., Beckley, L., 1994. Two Oceans: A guide to the
Marine Life of Southern Africa. David Philip Publishers, Cape Town.

Brooks, A.S., Yellen, J.E., Potts, R., Behrensmeyer, A.K., Deino, A.L., Leslie, D.E.,
Ambrose, S.H., Ferguson, J.R., d'Errico, F., Zipkin, A.M., Whittaker, S., Post, J.,
Veatch, E.G., Foecke, K., Clark, J.B., 2018. Long-distance stone transport and
pigment use in the earliest Middle Stone Age. Science 360, 90e94. https://
doi.org/10.1126/science.aao2646.

Cabral, J.P., Martins, J.M.S., 2016. Archaeological Glycymeris glycymeris shells
perforated at the umbo: Natural or man-made holes? J. Archaeol. Sci. Rep. 10,
474e482. https://doi.org/10.1016/j.jasrep.2016.11.008.

Campmas, E., Michel, P., Costamagno, S., Amani, F., Stoetzel, E., Nespoulet, R., El
Hajraoui, M.A., 2015. Were Upper Pleistocene human/non-human predator
occupations at the T�emara caves (El Harhoura 2 and El Mnasra, Morocco)

https://doi.org/10.1016/j.jhevol.2023.103438
https://doi.org/10.1006/jasc.1997.0277
https://doi.org/10.1086/589509
https://doi.org/10.1007/s10437-014-9154-7
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref4
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref4
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref4
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref4
https://doi.org/10.1016/j.jhevol.2008.10.005
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref6
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref6
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref6
https://doi.org/10.4207/PA.2019.ART121
https://doi.org/10.1371/journal.pone.0234924
https://doi.org/10.1371/journal.pone.0261282
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref10
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref10
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref10
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref10
https://doi.org/10.1073/pnas.0703877104
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref12
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref12
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref13
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref13
https://doi.org/10.1126/science.aao2646
https://doi.org/10.1126/science.aao2646
https://doi.org/10.1016/j.jasrep.2016.11.008


F. d'Errico, K.L. van Niekerk, L. Geis et al. Journal of Human Evolution 184 (2023) 103438
influenced by climate change? J. Hum. Evol. 78, 122e143. https://doi.org/
10.1016/j.jhevol.2014.08.008.

Campmas, E., Amani, F., Morala, A., Deb�enath, A., El Hajraoui, M.A., Nespoulet, R.,
2016. Initial insights into Aterian hunter-gatherer settlements on coastal
landscapes: The example of Unit 8 of El Mnasra Cave (T�emara, Morocco). Quat.
Int. 413, 5e20. https://doi.org/10.1016/j.quaint.2015.11.136.

Cawthra, H.C., Cowling, R.M., And�o, S., Marean, C.W., 2020. Geological and soil maps
of the Palaeo-Agulhas Plain for the Last Glacial Maximum. Quat. Sci. Rev. 235,
105858. https://doi.org/10.1016/j.quascirev.2019.07.040.

Claassen, C., 1998. Shells. Cambridge University Press, Cambridge.
Collins, B., Wojcieszak, M., Nowell, A., Hodgskiss, T., Ames, C.J., 2020. Beads and

bead residues as windows to past behaviours and taphonomy: A case study
from Grassridge Rockshelter, Eastern Cape, South Africa. Archaeol. Anthropol.
Sci. 12, 1e20. https://doi.org/10.1007/s12520-020-01164-5.

Conard, N.J., 2010. Cultural modernity: Consensus or conundrum? Proc. Natl. Acad.
Sci. USA 107, 7621e7622. https://doi.org/10.1073/pnas.1001458107.

Coolidge, F.L., Wynn, T., 2005. Working memory, its executive functions, and the
emergence of modern thinking. Camb. Archaeol. J. 15, 5e26. https://doi.org/
10.1017/S0959774305000016.

d'Errico, F., Henshilwood, C.S., Nilssen, P., 2001. An engraved bone fragment from c.
70,000-year-old Middle Stone Age levels at Blombos Cave, South Africa: Im-
plications for the origin of symbolism and language. Antiquity 75, 309e318.
https://doi.org/10.1017/S0003598X00060968.

d'Errico, F., Henshilwood, C.S., Lawson, G., Vanhaeren, M., Tillier, A.M., Soressi, M.,
Bresson, F., Maureille, B., Nowell, A., Lakarra, J., Backwell, L., Julien, M., 2003.
Archaeological evidence for the emergence of language, symbolism, and music
e An alternative multidiciplinary perspective. J. World Prehist. 17, 1e70.

d'Errico, F., Henshilwood, C., Vanhaeren, M., Van Niekerk, K.L., 2005. Nassarius
kraussianus shell beads from Blombos Cave: Evidence for symbolic behaviour in
the Middle Stone Age. J. Hum. Evol. 48, 3e24. https://doi.org/10.1016/
j.jhevol.2004.09.002.

d'Errico, F., Henshilwood, C.S., 2007. Additional evidence for bone technology in the
southern African Middle Stone Age. J. Hum. Evol. 52, 142e163. https://doi.org/
10.1016/j.jhevol.2006.08.003.

d'Errico, F., Vanhaeren, M., Wadley, L., 2008. Possible shell beads from the Middle
Stone Age layers of Sibudu Cave, South Africa. J. Archaeol. Sci. 35, 2675e2685.
https://doi.org/10.1016/j.jas.2008.04.023.

d'Errico, F., Vanhaeren, M., Barton, N., Bouzouggar, A., Mienis, H., Richter, D.,
Hublin, J.J., McPherron, S.P., Lozouet, P., 2009. Additional evidence on the use of
personal ornaments in the Middle Paleolithic of North Africa. Proc. Natl. Acad.
Sci. USA 106, 16051e16056. https://doi.org/10.1073/pnas.0903532106.

d'Errico, F., Backwell, L., Villa, P., Degano, I., Lucejko, J.J., Bamford, M.K., Higham, T.F.,
Colombini, M.P., Beaumont, P.B., 2012. Early evidence of San material culture
represented by organic artifacts from Border Cave, South Africa. Proc. Natl.
Acad. Sci. USA 109, 13214e13219. https://doi.org/10.1073/pnas.1204213109.

d'Errico, F., Vanhaeren, M., Van Niekerk, K., Henshilwood, C.S., Erasmus, R.M.,
2015. Assessing the accidental versus deliberate colour modification of shell
beads: A case study on perforated Nassarius kraussianus from Blombos Cave
Middle Stone Age levels. Archaeometry 57, 51e76. https://doi.org/10.1111/
arcm.12072.

d'Errico, F., Vanhaeren, M., 2015. Upper Palaeolithic mortuary practices: Reflection
of ethnic affiliation, social complexity, and cultural turnover. In: Renfrew, C.,
Boyd, M., Morley, I. (Eds.), Death Rituals, Social Order and the Archaeology of
Immortality in the Ancient World: Death Shall Have No Dominion. Cambridge
University Press, Cambridge, pp. 45e61.

d'Errico, F., Backwell, L., 2016. Earliest evidence of personal ornaments associated
with burial: The Conus shells from Border Cave. J. Hum. Evol. 93, 91e108.
https://doi.org/10.1016/j.jhevol.2016.01.002.

d'Errico, F., Doyon, L., Zhang, S., Baumann, M., L�azni�ckov�a-Galetov�a, M., Gao, X.,
Chen, F., Zhang, Y., 2018. The origin and evolution of sewing technologies in
Eurasia and North America. J. Hum. Evol. 125, 71e86. https://doi.org/10.1016/
j.jhevol.2018.10.004.

d'Errico, F., Martí, A.P., Shipton, C., Le Vraux, E., Ndiema, E., Goldstein, S.,
Petraglia, M.D., Boivin, N., 2020. Trajectories of cultural innovation from the
Middle to Later Stone Age in Eastern Africa: Personal ornaments, bone artifacts,
and ocher from Panga ya Saidi, Kenya. J. Hum. Evol. 141, 102737. https://doi.org/
10.1016/j.jhevol.2019.102737.

d'Errico, F., Martí, A.P., Wei, Y., Gao, X., Vanhaeren, M., Doyon, L., 2021. Zhoukoudian
Upper Cave personal ornaments and ochre: Rediscovery and reevaluation.
J. Hum. Evol. 161, 103088. https://doi.org/10.1016/j.jhevol.2021.103088.

Dapschauskas, R., G€oden, M.B., Sommer, C., Kandel, A.W., 2022. The emergence of
habitual ochre use in Africa and its significance for the development of ritual
behavior during The Middle Stone Age. J. World Prehist. 35, 233e319. https://
doi.org/10.1007/s10963-022-09170-2.

Dayet, L., Erasmus, R., Val, A., Feyfant, L., Porraz, G., 2017. Beads, pigments and early
Holocene ornamental traditions at Bushman Rock Shelter, South Africa.
J. Archaeol. Sci. Rep. 13, 635e651. https://doi.org/10.1016/j.jasrep.2017.05.015.

De Lumley, H., Audubert, F., Khatib, S., Perrenoud, C., Roussel, B., Saos, T., Szelewa, A.,
2016. Les crayons d'ocre du site archeul�een de Terra Amata. In: De Lumley, H.
(Ed.), Terra Amata, Nice, Alpes-Maritimes, France Tome V. CNRS Editions, Paris,
pp. 233e277.

Dibble, H.L., Aldeias, V., Alvarez-Fernandez, E., Blackwell, B.A., Hallett-Desguez, E.,
Jacobs, Z., Goldberg, P., Lin, S.C., Morala, A., Meyer, M.C., 2012. New excavations
at the site of Contrebandiers Cave, Morocco. PaleoAnthropology 2012, 145e201.
https://doi.org/10.4207/PA.2012.ART74.
17
Discamps, E., Henshilwood, C.S., 2015. Intra-site variability in the Still Bay fauna at
Blombos Cave: Implications for explanatory models of the Middle Stone Age
cultural and technological evolution. PLoS One 10, e0144866. https://doi.org/
10.1371/journal.pone.0144866.

Doerschner, N., Fitzsimmons, K.E., Ditchfield, P., McLaren, S.J., Steele, T.E.,
Zielhofer, C., McPherron, S.P., Bouzouggar, A., Hublin, J.-J., 2016. A new chro-
nology for Rhafas, northeast Morocco, spanning the North African Middle Stone
Age through to the Neolithic. PLoS One 11, e0162280. https://doi.org/10.1371/
journal.pone.0162280.

Douze, K., Wurz, S., Henshilwood, C.S., 2015. Techno-cultural characterization of the
MIS 5 (c. 105e90 ka) Lithic industries at Blombos Cave, Southern Cape, South
Africa. PLoS One 10, e0142151. https://doi.org/10.1371/journal.pone.0142151.

Doyon, L., Faure, T., Sanz, M., Daura, J., Cassard, L., d'Errico, F., 2023. A 39,600-year-
old leather punch board from Canyars, Gav�a, Spain. Sci. Adv. 9, eadg0834.
https://doi.org/10.1126/sciadv.adg0834.

Dubin, L.S., 2009. The Worldwide History of Beads. Ancient, Ethnic, Contemporary.
Thames & Hudson, London.

Eiwanger, J., Mikdad, A., Moser, J., Nami, M., 2012. D�ecouverte de coquilles perfor�ees
de type Nassarius au site Ifri n'Ammar (Rif Oriental, Maroc). Bull. Arch. Maroc.
22, 9e15.

El Hajraoui, M.A., Oudouche, H., Nespoulet, R., 2012. �Etude des coquilles perfor�ees
d�ecouvertes �a T�emara. In: El Hajraoui, M.A., Nespoulet, R., Deb�enath, A.,
Dibble, H.L. (Eds.), Pr�ehistoire de la R�egion de Rabat-T�emara. Institut National
des Sciences de l'Arch�eologie et du Patrimoine, Rabat, pp. 191e199.

Falci, C.G., Cuisin, J., Delpuech, A., Van Gijn, A., Hofman, C.L., 2019. New insights into
use-wear development in bodily ornaments through the study of ethnographic
collections. J. Archaeol. Method Theory 26, 755e805. https://doi.org/10.1007/
s10816-018-9389-8.

Fern�andez, E., 2006. Los objetos de adorno-colgantes del Paleolítico superior y del
Mesolítico en la Cornisa Cant�abrica y en el Valle del Ebro: una visi�on europea.
Ph.D. Dissertation, Universidad de Salamanca.

Fern�andez, E., Joris, O., 2008. Personal ornaments in the Early Upper Paleolithic of
Western Eurasia: An evaluation of the record. Eurasian Prehistory 5, 29e42.

Fisher, E.C., Bar-Matthews, M., Jerardino, A., Marean, C.W., 2010. Middle and Late
Pleistocene paleoscape modeling along the southern coast of South Africa. Quat.
Sci. Rev. 29, 1382e1398. https://doi.org/10.1016/j.quascirev.2010.01.015.

Garfinkel, Y., 2018. The evolution of human dance: Courtship, rites of passage, trance,
calendrical ceremonies and the professional dancer. Camb. Arch. J. 28, 283e298.

Garofoli, D., 2015. Do early body ornaments prove cognitive modernity? A critical
analysis from situated cognition. Phenomenol. Cogn. Sci. 14, 803e825. https://
doi.org/10.1007/s11097-014-9356-0.

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. J. Hum. Evol. 62, 533e547. https://doi.org/10.1016/j.jhevol.2012.02.004.

Gravel-Miguel, C., Cristiani, E., Hodgkins, J., Orr, C.M., Strait, D.S., Peresani, M.,
Benazzi, S., Pothier-Bouchard, G., Keller, H.M., Meyer, D., Drohobytsky, D.,
Talamo, S., Panetta, D., Zupancich, A., Miller, C.E., Negrino, F., Riel-Salvatore, J.,
2022. The ornaments of the Arma Veirana early Mesolithic infant burial.
J. Archaeol. Method Theory 1e48. https://doi.org/10.1007/s10816-022-09573-7.

Gurova, M., Bonsall, C., Bradley, B., Anastassova, E., 2013. Approaching prehistoric
skills: Experimental drilling in the context of bead manufacturing. Bulg. e-J.
Archaeol. 3, 201e221.

Haaland, M.M., Miller, C.E., Unhammer, O.F., Reynard, J.P., van Niekerk, K.L.,
Ligouis, B., Mentzer, S.M., Henshilwood, C.S., 2021. Geoarchaeological investi-
gation of occupation deposits in Blombos Cave in South Africa indicate changes
in site use and settlement dynamics in the southern Cape during MIS 5b-4.
Quat. Res. 100, 170e223. https://doi.org/10.1017/qua.2020.75.

Hatton, A., Schoville, B.J., Wilkins, J., 2020. A quantitative analysis of wear distri-
butions on Middle Stone Age marine shell beads from Blombos Cave, South
Africa. J. Archaeol. Sci. Rep. 29, 102137. https://doi.org/10.1016/
j.jasrep.2019.102137.

Henshilwood, C.S., d'Errico, F., Marean, C.W., Milo, R.G., Yates, R., 2001a. An early
bone tool industry from the Middle Stone Age at Blombos Cave, South Africa:
Implications for the origins of modern human behaviour, symbolism and lan-
guage. J. Hum. Evol. 41, 631e678. https://doi.org/10.1006/jhev.2001.0515.

Henshilwood, C.S., Sealy, J.C., Yates, R., Cruz-Uribe, K., Goldberg, P., Grine, F.E.,
Klein, R.G., Poggenpoel, C., van Niekerk, K.L., Watts, I., 2001b. Blombos Cave,
Southern Cape, South Africa: Preliminary report on the 1992e1999 excavations
of the Middle Stone Age levels. J. Archaeol. Sci. 28, 421e448. https://doi.org/
10.1006/jasc.2000.0638.

Henshilwood, C.S., d'Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G.A.T.,
Mercier, N., Sealy, J.C., Valladas, H., Watts, I., 2002. Emergence of modern human
behavior: Middle Stone Age engravings from South Africa. Science 295,
1278e1280. https://doi.org/10.1126/science.1067575.

Henshilwood, C.S., d'Errico, F., Vanhaeren, M., van Niekerk, K., Jacobs, Z., 2004.
Middle Stone Age shell beads from South Africa. Science 304. https://doi.org/
10.1126/science.1095905, 404e404.

Henshilwood, C.S., 2005. Stratigraphic integrity of the middle stone age levels at
Blombos cave. In: d'Errico, F., Backwell, L. (Eds.), From Tools to Symbols: From
Early Hominids to Modern Humans. Witwatersrand University Press, Johan-
nesburg, pp. 441e458.

Henshilwood, C.S., d'Errico, F., Watts, I., 2009. Engraved ochres from the Middle
Stone Age levels at Blombos Cave, South Africa. J. Hum. Evol. 57, 27e47. https://
doi.org/10.1016/j.jhevol.2009.01.005.

https://doi.org/10.1016/j.jhevol.2014.08.008
https://doi.org/10.1016/j.jhevol.2014.08.008
https://doi.org/10.1016/j.quaint.2015.11.136
https://doi.org/10.1016/j.quascirev.2019.07.040
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref19
https://doi.org/10.1007/s12520-020-01164-5
https://doi.org/10.1073/pnas.1001458107
https://doi.org/10.1017/S0959774305000016
https://doi.org/10.1017/S0959774305000016
https://doi.org/10.1017/S0003598X00060968
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref24
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref24
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref24
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref24
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref24
https://doi.org/10.1016/j.jhevol.2004.09.002
https://doi.org/10.1016/j.jhevol.2004.09.002
https://doi.org/10.1016/j.jhevol.2006.08.003
https://doi.org/10.1016/j.jhevol.2006.08.003
https://doi.org/10.1016/j.jas.2008.04.023
https://doi.org/10.1073/pnas.0903532106
https://doi.org/10.1073/pnas.1204213109
https://doi.org/10.1111/arcm.12072
https://doi.org/10.1111/arcm.12072
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref31
https://doi.org/10.1016/j.jhevol.2016.01.002
https://doi.org/10.1016/j.jhevol.2018.10.004
https://doi.org/10.1016/j.jhevol.2018.10.004
https://doi.org/10.1016/j.jhevol.2019.102737
https://doi.org/10.1016/j.jhevol.2019.102737
https://doi.org/10.1016/j.jhevol.2021.103088
https://doi.org/10.1007/s10963-022-09170-2
https://doi.org/10.1007/s10963-022-09170-2
https://doi.org/10.1016/j.jasrep.2017.05.015
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref38
https://doi.org/10.4207/PA.2012.ART74
https://doi.org/10.1371/journal.pone.0144866
https://doi.org/10.1371/journal.pone.0144866
https://doi.org/10.1371/journal.pone.0162280
https://doi.org/10.1371/journal.pone.0162280
https://doi.org/10.1371/journal.pone.0142151
https://doi.org/10.1126/sciadv.adg0834
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref45
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref45
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref45
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref46
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref47
https://doi.org/10.1007/s10816-018-9389-8
https://doi.org/10.1007/s10816-018-9389-8
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref49
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref50
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref50
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref50
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref50
https://doi.org/10.1016/j.quascirev.2010.01.015
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref52
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref52
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref52
https://doi.org/10.1007/s11097-014-9356-0
https://doi.org/10.1007/s11097-014-9356-0
https://doi.org/10.1016/j.jhevol.2012.02.004
https://doi.org/10.1007/s10816-022-09573-7
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref56
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref56
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref56
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref56
https://doi.org/10.1017/qua.2020.75
https://doi.org/10.1016/j.jasrep.2019.102137
https://doi.org/10.1016/j.jasrep.2019.102137
https://doi.org/10.1006/jhev.2001.0515
https://doi.org/10.1006/jasc.2000.0638
https://doi.org/10.1006/jasc.2000.0638
https://doi.org/10.1126/science.1067575
https://doi.org/10.1126/science.1095905
https://doi.org/10.1126/science.1095905
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref63
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref63
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref63
http://refhub.elsevier.com/S0047-2484(23)00117-3/sref63
http://refhub.elsevie