See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236175791

The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang,

Arnhem Land (Australia) and Global Implications for the Evolution of Fully
Modern Humans

Article in Cambridge Archaeological Journal · February 2012

DOI: 10.1017/S0959774312000017

CITATIONS READS
40 1,589

6 authors, including:

Jean-Michel Geneste Bruno David

Ministère de la culture et de la communication Monash University (Australia)
82 PUBLICATIONS 2,867 CITATIONS 201 PUBLICATIONS 3,360 CITATIONS

SEE PROFILE SEE PROFILE

Hugues Plisson Chris Clarkson

University of Bordeaux The University of Queensland
46 PUBLICATIONS 1,166 CITATIONS 123 PUBLICATIONS 3,976 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Lascaux scientific council View project

CAVE, ARCHIVE, MEMORY, INDICATOR, CHRONOLOGY, DATING, MODEL, SYSTEM, GEOMORPHOLOGY, ARCHEOLOGY, ROCK-ART, View project

All content following this page was uploaded by Fiona Petchey on 26 February 2014.

The user has requested enhancement of the downloaded file.

Cambridge Archaeological Journal
http://journals.cambridge.org/CAJ

Additional services for Cambridge Archaeological Journal:

Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

The Origins of Ground-edge Axes: New Findings from Nawarla

Gabarnmang, Arnhem Land (Australia) and Global Implications for the
Evolution of Fully Modern Humans

Jean-Michel Geneste, Bruno David, Hugues Plisson, Jean-Jacques Delannoy and Fiona Petchey

Cambridge Archaeological Journal / Volume 22 / Issue 01 / February 2012, pp 1 - 17

DOI: 10.1017/S0959774312000017, Published online: 28 February 2012

Link to this article: http://journals.cambridge.org/abstract_S0959774312000017

How to cite this article:

Jean-Michel Geneste, Bruno David, Hugues Plisson, Jean-Jacques Delannoy and Fiona Petchey (2012). The Origins of
Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land (Australia) and Global Implications for the
Evolution of Fully Modern Humans. Cambridge Archaeological Journal, 22, pp 1-17 doi:10.1017/S0959774312000017

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/CAJ, IP address: 153.122.19.69 on 11 Dec 2013

 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land

The Origins of Ground-edge Axes:

New Findings from Nawarla Gabarnmang, Arnhem Land
(Australia) and Global Implications for the
Evolution of Fully Modern Humans

Jean-Michel Geneste, Bruno David, Hugues Plisson,

Jean-Jacques Delannoy & Fiona Petchey

The grinding of stone to make sharp cutting edges did not evolve with the emergence of
biologically modern humans in Africa, but late in the Pleistocene at the completion or near-
completion of the Out-of-Africa 2 migration. Here we discuss the earliest securely-dated
fragment of ground-edge axe from Australia, dated at 35,500 cal. bp, an age slightly older or
comparable to the earliest ages for edge-grinding from the Pacific Ocean’s western seaboard.
In this region ground-edge axes did not evolve with the emergence of agriculture, nor for
the clearance of forests, but, rather, as socially mediated technology, part of the development
of symbolic storage that is the hallmark of the evolution of cognitively modern humans at
the geographical end, during, or following, Out-of-Africa 2.

The world’s earliest stone tools date to 2.52–2.60 mil-
lion years ago and predate the earliest known Homo
species along the Gona River of the Afar region of
Ethiopia, eastern Africa (Semaw et al. 1997). Even
older knapped stone tools are posited to exist, with
the recent discovery of cut marks on 3.4 million-year-
old bones from Dikika attributed to Australopithecus
afarensis, also in Ethiopia (McPherron et al. 2010). Such
early core and flake tools were all manufactured by
freehand percussion, although bipolar debitage is
also found in Oldowan times (Braun et al. 2009; Clark
& Kleindienst 2001; de la Torre et al. 2003; de la Torre
& Mora 2009).
With the evolution of Homo sapiens between
200,000 and 100,000 years ago emerged a broad range
of technological and cognitive innovations (e.g. Dea-
con & Deacon 1999) that over time came to incorporate
new forms of symbolic behaviour including rock art
(Henshilwood et al. 2002; Conard & Bolus 2003; 2008;
Conard et al. 2003; d’Errico et al. 2010; Parkington et
al. 2005), complex multi-component tools (Mellars
et al. 2007; Zilhão & d’Errico 2003), environmental
manipulations (de Beaune 2000; 2003), long-distance
trade (Bon 2009, 241–63; Hovers & Braun 2009) and
fully linguistically mediated group behaviour (Bicker-
ton 2003; d’Errico et al. 2003; Gamble & Coward 2009;
Vanhaeren & d’Errico 2006). Each of these components
of social life, from complex multi-component tools
to environmental manipulations, is permeated with
symbolic storage because it incorporates very particu-
lar ways of doing things subject to human cognisance.
These ‘ways of doing’ result in particular material
signatures to all kinds of products, from objects to
lived landscapes. While such products may be seen
as ‘economic production’, they also involve a socio-
economy of symbolic meaning and reproduction.
However, such technological and cognitive innova-
tions did not suddenly come in a single package with
the evolution of biologically fully modern humans
(for a review of the evidence, see Habgood & Franklin
2008). Rather, over the tens of thousands of years
following the first emergence of biologically modern
humans during the Middle Stone Age by 100,000
years ago in Africa (Anton & Swisher 2004; Field et

Cambridge Archaeological Journal 22:1, 1–17 © 2012 McDonald Institute for Archaeological Research
doi:10.1017/S0959774312000017 Received 13 Apr 2011; Revised 10 Aug 2011;1Accepted 2 Sep 2011
 Jean-Michel Geneste et al.
al. 2007), new technologies and symbolic behaviours
were gradually and cumulatively added out of Africa.
Sahul, that region on the extreme eastern edge of the
Out-of-Africa 2 migration route, has a crucial part to
play in understanding those behavioural traits that
emerged following, versus prior to, the out of Africa
migration, for innovations identified in Sahul will
indicate post-migration inventions that emerged sub-
sequent to Homo sapiens’ early behavioural develop-
ments. Identifying such post-migration developments
enables a better understanding of the extent to which
biologically modern humans became behaviourally
modern in a process of ongoing developments.
One of these traits of modernity concerns ‘sym-
bolic storage’, which many authors have identified as
originating in Africa but continuing to unfold during
and subsequent to the Out-of-Africa 2 migration (e.g.
Habgood & Franklin 2008; Wadley 2001). Symbolic
storage cannot in itself be reduced to short-term ergo-
nomic efficiency, although this does not mean that it
works counter to adaptive principles. Rather, symbolic
storage in material culture concerns the way in which
objects store information about meaning, meanings
that relate to one’s physical and socio-cultural loca-
tion. Symbolic storage imbues things, physical and
non-physical, with a presence enhanced by social
value. By adding social meaning, symbolic storage
also socially structures use-value, for meaning comes
into being in social and cultural codes, in culturally
engaged social interaction.
Here we report on the discovery of the oldest
fragment of ground-edge axe in the world, and the
origins of ground-edge axes, as one such innovation
indicative of symbolic storage that took place follow-
ing Out-of-Africa 2 along the Southern Arc dispersal
route as humans crossed from Africa to southern Asia
into northeast Asia and Australia between 70,000
and 50,000 years ago (cf. Balme et al. 2009). We argue
that this innovation cannot be understood simply in
abstract technological terms, but must also consider
the evolution of symbolic behaviour as socially medi-
ated technology.
The origins of grinding
Despite the manufacture and use of flaked stone tools
by hominins in Ethiopia more than two million years
ago (McPherron et al. 2010; Semaw et al. 1997), the
shaping of stone tools by grinding does not appear
in human evolution until about 35,500 years ago
anywhere in the world, coincident in time with the
emergence of naturalistic rock imagery (Clottes 2007;
Parkington et al. 2005; Valladas et al. 2001). Here we
distinguish between tools intentionally shaped by
2
grinding with an abrading tool, such as ground-edge
axes and ground flakes, from the abrading tools used
to manufacture various materials such as objects of
adornment, plant produce, stone tools, and miner-
als for the production of pigments, such as files
and grinding stones. Globally, and concerning raw
materials other than stone, the oldest-known cases
of grinding are found on bone tools from Swartkrans
and Drimolen (Backwell & d’Errico in press; d’Errico
& Backwell 2009).
In Europe, fine, intentionally abraded shallow
bowls in sedimentary rocks and calcite have been
found in residential sites attributed to the Mousterian.
The discovery of stone receptacles for the preparation
of ochre at Cioarei Cave is testimony to the intentional
transformation of blocks of stone by pounding and
abrasion (Carciumaru et al. 2010).
Generally, scraped and abraded pigments are
most notably found at habitation sites dating from
the end of the Middle Palaeolithic. In the southwest
of France, more than 500 fragments of red (iron
oxides, haematite, goethite) and black (manganese
dioxide) pigments were ground for decoration and
thus symbolic expression by Homo neanderthalensis
during the Mousterian. There are now more than 70
sites that have yielded such pigments; at Pech de l’Azé
rockshelter these go as far back as 50–60,000 bp (Sor-
essi & d’Errico 2007; d’Errico et al. 2009). In the cave
of Blombos in South Africa, red-ochre crayons were
striated by abrasion during use in the Middle Stone
Age, 70,000 years ago and presumably by Homo sapiens
(Henshilwood et al. 2002; Henshilwood & d’Errico
2005; d’Errico et al. 2010). Grinding tools and abraded
objects appear sporadically during the European
Upper Palaeolithic after 30,000 bp and include pestles,
handstones, cupules and shallow stone bowls (de
Beaune 2000; 2004). The earliest grinding tools used for
plant-food processing in Europe come from Bilancino
II in Italy (dated to 28,298±301 cal. bc), Kostenki 16 in
Russia (31,904±698 cal. bc) and Pavlov VI in the Czech
Republic (29,482±288 cal. bc) (Revedin et al. 2010). At
the Upper Palaeolithic occupation site of Kostenki IV
(Aleksandrovskaya) on the Russian plains, mortars
and pestles have also been reported from levels dated
to 23,000–21,000 bp (Semenov 1964 [1957], 137, fig. 67).
In Europe there is also firm evidence for the
grinding of stone in the manufacture of portable and
fixed rock art during the Upper Palaeolithic. In France
shallow bowls ground in stone used for mixing pig-
ments and lamps for the burning of animal fat are
symmetrically shaped and decorated, such as those
of Lascaux and of Solvieux (Sackett 1999, figs. 4.7, 4.8;
de Beaune 2000; 2004). Further to the east at Denisova
Cave in Siberia a finely polished bracelet dated to
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land
30,000 bp is made of semi-precious ground chloritolite
(Derevianko et al. 2008), although the indirect dat-
ing, in a complex stratigraphy with superior Bronze
age levels, is questionable. From France to Siberia,
female (‘Venus’) figurines from the early and middle
phases of the Upper Palaeolithic evidence the use of
grinding and polishing stone technology to produce
figurative objects with presumably high social value.
Examples dating to the Aurignacian and Gravettian
between 30,000 and 25,000 bp include the famous stone
and ivory figurines from Galgenberg, Willendorf,
Grimaldi, Sireuil, Tursac and Kostenki for example
(Conard 2009; Leroi-Gourhan 1995; Neugebauer-
Maresch 1989; Svoboda 2008).
In the Near East, tools to grind grain and vege-
table fibre are well known from Natufian levels dated
to 15,000 bp (Bar-Yosef 1998, fig. 11; 2002). In southern
Africa, grinding stones and the abrasion of stone dates
to the Late Stone Age from 8500 bp onwards (Ambrose
1998). Ground shell artefacts are found in the Moluccas
12,000 bp (Bellwood 1997, 187, cited in Anderson &
Summerhayes 2008, 51).
In Australia, stones used for grinding have
been discovered at Malakunanja II bracketed by
61,000±13,000 and 45,000±9000 year-old TL dates
(Roberts et al. 1990, 153), although little has been pub-
lished about these stones. Pieces of use-striated hae-
matite occur in the same date-context at Malakunanja
II (Roberts et al. 1990, 153), and also at Nauwalabila
1 in close association with a 53,400±5400 year-old
TL date (Jones & Johnson 1985; Roberts et al. 1993).
Use-striated ochre is also found at Sandy Creek 1
on Cape York Peninsula dated at 31,900+700/–600 bp
(Cole et al. 1995, 154). While there are claims for
grinding stones dating back to 33,300±530 bp from
Cuddie Springs (Fullagar & Field 1997), the chrono-
stratigraphic integrity of this site has been seriously
questioned by a number of authors (e.g. Brook et al.
2007; David 2002, 169–76; Gillespie & David 2001;
Gillespie & Brook 2006), and cannot at this stage
be taken as reliable evidence (see also Habgood &
Franklin 2008 for an excellent review of the antiquity
of grinding stones and ochre use in Australia).
These findings indicate that the appearance of
grinding tools for the production of objects of art, with
implications for the production of visual symbolism,
have a long antiquity around the world, with most
examples coming from late Pleistocene levels onwards
in direct association with modern Homo sapiens.
The earliest ground-stone axes
Grinding as a technology to create functional stone
tools has a similar antiquity in some parts of the world
3
but not in others. In Western Europe, ground axes first
appear during the Mesolithic in the United Kingdom,
Ireland, Scandinavia and Russia and increase signifi-
cantly in frequency during the Neolithic (Hernek 2005;
Pailler & Sheridan 2009).
The first ground-edge axes are found at the
beginning of the Mesolithic in Ireland, such as at
Lough Boora (Co. Offaly), in habitation levels dated
to 7160–6260 bc. In Wales, Scotland and Ireland, the
numerous ground-edge axes reliably dated to the
Mesolithic indicate that edge-grinding had appeared
by the eighth millennium bc, well before the com-
mencement of Neolithic lifeways (Pailler & Sheridan
2009, 31–2).
In Siberia the oldest ground axes date to 20,000 bp
in the valley of Yenisei (Oda & Keally 1973, 19, cited
by Anderson & Summerhayes 2008, 49). On the Rus-
sian plain and in northern Russia ground axes make
their appearance around 8500 bp during the Meso-
lithic; at Zamostje the surfaces of about 50 per cent
of axes are entirely ground, and at Nijnee Veretje 1
they are nearly completely ground (Lozovski 1996).
In the Near East, ground-stone objects make their
appearance at the end of the Natufian as containers
and tools for the processing of plant foods rather than
as axe-heads. In Sultanian tool kits, between 10,300 bp
and 9300 bp, limestone celts are present (Bar-Yosef
1998, fig. 11). In northwestern and Saharan Africa,
ground-edge axes begin to appear after the Capsian
which developed between 8500 and 6500 bp (Philipp-
son 2005; Smith 2005). Elsewhere in Africa it is in
the Late Stone Age and with agro-pastoralists that
ground-edge axes become common, often in associa-
tion with other ground objects. At Border Cave in
southern Africa, as early as 38,000 bp the transition
from the Middle Stone Age to the Late Stone Age saw
the development of a microlithic industry associated
with more ‘modern’ assemblages such as bone points
and bored stones shaped and perforated by abrasion
(Beaumont et al. 1978; Sadr et al. 2006, 405).
Unlike the contiguous region of Europe–Africa–
west Asia–north Asia, eastern Asia and the New
Guinea–Australian region of Sahul present a different
scenario.
In southern China a ground-edge stone axe
from layer 4 at Bailiandong Cave in Guangxi Pro-
vince is dated by two radiocarbon determinations
of 19,350±180 bp and 20,960±150 bp (Zhao et al. 2004,
132). At the site of Liyuzui also in Guangxi Province a
ground-edge axe is bracketed by radiocarbon dates of
21,025±450 bp and 11,450±150 bp. Terminal Pleistocene
ages have also been assigned to ground-edge tools
from Dushizai Cave and Huangyandong Cave (both
of Guangdong Province). There are also less certain
 Jean-Michel Geneste et al.
terminal Pleistocene ‘polished’ stones from Xianren-
dong Cave and Diatonghua Cave (Zhao et al. 2004).
In Japan the oldest ground-edge axes date to
c. 30,000–23,000 bp, the earliest from Stratum Xb of
the Musashidai site in Tokyo (Oda & Keally 1992,
25). Approximately 60 per cent of the more than 200
ground-edge or waisted axe-like implements that
have been found below the AT marker tephra on
Honshu and Kyushu are ground. While poorly dated,
these tools are older than the AT marker tephra dated
to sometime between 21,000 and 22,000 bp (Machida
& Arai 1983, cited in Oda & Keally 1992, 25). Until
recently these artefacts represented the oldest evi-
dence of edge grinding in the world, with Australia’s
Pleistocene axes dating to a broadly similar age.
Pleistocene ground-edge axes have been exca-
vated from northern Australian sites since the 1960s
(White 1967). Most of these tools come from Arnhem
Land in the centre-north of the country: examples
made on porphyritic dolerite, quartz dolerite and
hornfels are known from the lowermost Levels IIIa
and IIIb at Malangangerr, dated from 18,000±400 bp
to 22,900±1000 bp respectively (Schrire 1982, 84, 107).
At Nawamoyn, porphyritic dolerite and hornfels
ground-edge axes have also been found in Lev-
els IIIa and IIIb, the latter dated to 21,450±380 bp
(Schrire 1982, 118, 143). At Nauwalabila 1, fragments
of ground-edge axes were recovered as far down as
Unit 45 (149 cm below ground), originally estimated
from the extrapolated depth-age curve to date to
15,000–16,000 bp (Jones & Johnson 1985, 216) and
subsequently shown to date to sometime between
12,330±120 and 18,330±280 bp by an extensive AMS
radiocarbon dating programme using ABOX pre-
treatment (Bird et al. 2002, 1065) and 13,500±900
and 30,000±2400 years old by OSL dating (Roberts
et al. 1993, 58). Chemically-weathered imported
porphyritic dolerite flakes likely to have come from
ground-edge axes were also found as far down as
Unit 70 at depths of 2.13–2.43 cm below ground and
thereby dating to sometime between 30,000±2400 and
53,400±5400 years ago by OSL (Roberts et al. 1993, 58),
but these artefacts are too weathered and decom-
posed for surface features including traces of grind-
ing to have preserved (Jones & Johnson 1985, 196–7,
216–18, 297). To the west at Widgingarri Shelter 1 in
the Kimberley, fragments of ground-edge axes have
been reported from spit 17 dated to 27,610±600 bp
(O’Connor 1990, 194, 246–51). To the east of Arn-
hem Land, a local amateur dug a hole at the site of
Sandy Creek 1 in Cape York Peninsula in the 1960s,
retrieving an edge-ground axe supposedly from 3 m
(9' 10") below ground (Morwood & Trezise 1989).
In 1989 archaeologist Mike Morwood excavated a
4
trench parallel and adjacent to Trezise’s original pit
and obtained charcoal from the nearby stratigraphy
to reveal an age of 31,900+700/–600 bp for the basal
rubble that is said to have contained the axe. As the
axe was supposedly found at the base of the rubble,
Morwood and Trezise (1989, 81) argued that the
radiocarbon determination ‘provides a minimum age
for the edge-ground axe’. However, the original axe
had long been lost and its supposed basal chrono-
stratigraphic positioning was compromised by the
amateur enterprise and lack of rigorous scientific
procedures, prompting serious doubts about the
reliability of the Sandy Creek 1 claims.
Nawarla Gabarnmang
Nawarla Gabarnmang is a newly-rediscovered site
in Jawoyn Aboriginal country in southwestern Arn-
hem Land (Fig. 1). Located on top of the escarpment
in the Kombolgie Formation, it is a highly unusual
horizontally-bedded but intercalated stable hard
orthoquartzite and softer and less stable quartz sand-
stone formation which has over a geologically-long
period been subjected to in situ weathering of the bed-
rock through hydration and solution processes along
bedding planes and fissure lines. The end-product is
a double-ended cave with sub-horizontal ceiling rang-
ing from 1.75 m to 2.45 m above the floor level. Across
the more open and thus easily accessed part of the site
is a natural grid-shaped alignment of 36 stone pillars
(Fig. 2) each separated by fissure lines along which the
bedrock has weathered away from ground to ceiling,
leaving behind the pillars as relicts of weathering
processes. The archaeological implication of these
geomorphological features is highly significant: a spa-
tious cave, 19 m long by 19 m wide shelters expansive
culturally-bearing sediments entirely protected from
rainfall. Despite the low pH (3.71–5.93) and thus high
acidity values of its sediments (as is the case across
all of Arnhem Land generally), the implication is
that ancient cultural remains have a better chance of
being preserved in this site than in any other known
archaeological site in Arnhem Land.
In May 2010, two 50 cm × 60 cm archaeological
excavations (Squares A and B) were undertaken in dif-
ferent parts of the site to test its archaeological poten-
tial. Here we report on Square A only (Fig. 3). Excava-
tion was undertaken in mean 1.8 cm-thick arbitrary
Excavation Units (XU) following Stratigraphic Units
(SU) where visible in situ until bedrock was reached
66.2 cm below surface. Cultural materials ≥2.0 cm
observed in situ were plotted in three dimensions
(3-D) and bagged separately. Charcoal samples
were also plotted in 3-D and individually bagged
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land

Table 1. Radiocarbon data from Nawarla Gabarnmang, Square A. * = δ13C value measured, but not reported. See text for calibration references.
Conventional
Depth below δ13C‰
XU SU radiocarbon age Calibrated age Lab no.
ground (cm) (±0.2)
(bp)
320–280, 170–150 bp (68%)
1 1 0.3 24.7 250±30 Wk-28123
430–360, 330–270, 190–140, 20–10 bp (95%)
280–250, 230–170, 160–130, 120–70, 40–10 bp (68%)
3 2 6.1 26.0 142±30 Wk-28614
290–0 bp (95%)
1520–1400 bp (68%)
5 3 8.3 23.0 1560±30 Wk-28615
1530–1380 bp (95%)
2350–2300, 2240–2180 bp (68%)
6 3 6.7 24.6 2264±30 Wk-28124
2350–2290, 2260–2150 bp (95%)
2310–2240, 2180–2170, 2160–2120 bp (68%)
8 3 13.0 26.2 2165±30 Wk-28612
2310–2060 bp (95%)
9600–9530 bp (68%)
11 3–4 interface 17.4 25.1 8616±34 Wk-28611
9670–9520 bp (95%)
10,180–10,130, 10,070–10,010, 10,000–9940 bp (68%)
12 3–4 interface 19.9 26.4 8924±31 Wk-28125
10,200–10,110, 10,090–9910 bp (95%)
10,560–10,480, 10,470–10,420 bp (68%)
14 3–4 interface 21.8 25.3 9287±30 Wk-28613
10,580–10,380, 10,320–10,300 bp (95%)
11,190–11,090, 10,920–10,900 bp (68%)
16 4 23.0 na* 9670±34 Wk-28126
11,200–11,070, 10,960–10,860, 10,850–10,800 bp (95%)
13,840–13,710 bp (68%)
18 4 25.2 27.1 11,907±40 Wk-28619
13,910–13,610 bp (95%)
14,220–14,000 bp (68%)
20 4 28.9 25.4 12,261±43 Wk-28622
14,800–14,760, 14,550–13,920 bp (95%)
15,080–14,790, 14,750–14,660 bp (68%)
23 4 34.4 25.2 12,599±42 Wk-28617
15,200–14,520, 14,300–14,280 bp (95%)
21,410–21,050 bp (68%)
24 4–5 interface 37.8 26.1 17,722±64 Wk-28128
21,480–20,810, 20,700–20,560 bp (95%)
15,240–14,940 bp (68%)
25 4–5 interface 38.5 25.2 12,746±43 Wk-28620
15,570–14,870, 14,830–14,770 bp (95%)
36,120–36,020, 35,340–34,730 bp (68%)
28 5 46.5 23.7 30,615±309 Wk-28616
36,260–35,820, 35,640–34,620 bp (95%)
36,420–35,200 bp (68%)
29 5 46.1 25.4 31,316±618 Wk-28130
37,110–34,690 bp (95%)
36,280–35,760, 35,740–35,070 bp (68%)
31 5 52.7 24.7 31,063±595 Wk-28131
36,710–34,650 bp (95%)
36,170–35,960, 35,460–34,850 bp (68%)
31 5 51.2 24.2 30,761±314 Wk-28621
36,290–35,760, 35,710–34,720 bp (95%)

for AMS radiocarbon dating. All other excavated
sediments were sieved in 2.1 mm mesh sieves and
all retained materials bagged for laboratory sorting
at the Monash University archaeology laboratories.
Relatively large pieces of wood charcoal were found
throughout the excavation, sometimes in extremely
large quantities measured by the hundreds of grams
per XU, thus offering unprecedented opportuni-
ties for radiocarbon dating in Arnhem Land. Here
we report the significant finding of a confirmed
fragment of ground-edge axe sandwiched midway
between four AMS radiocarbon determinations,
each on a single fragment of charcoal dated to more
5
than 30,000 bp, and thus representing the oldest
securely dated ground-edge implement in the world.
The deepest and oldest layer is SU5 with a
thickness ranging from 7 cm to 31 cm, depending
on the uneven bedrock configuration and whether
a sediment-filled narrow fissure in the bedrock is
included. The ground-edge axe fragment comes from
XU30 in SU5, at a depth between 47.7 cm and 50.4 cm
below ground. The buried stones within these XUs
lay horizontally flat, and thus parallel to the configu-
ration of the SU in which they lay, indicating a lack
of significant post-depositional turbation. No roots
or rootlets were evident at this depth; indeed plants
 Jean-Michel Geneste et al.

Coo
VAN DIEMEN GULF E a st All
pe
r

Cr
at

ee
ig

k
or R
i ve
r

S ou
r
iv e

th A

rR
lligat
lligato
W

or
ild

Nawamoyn
Ri
m

st A

ve
an

We Jimeri
r

Paribari
Riv
er

Ngarradj Warde Djobkeng Malangangerr

Malakunanja

Eas
tA
llig
ato
r Rive
ag r

M
e la
Cree
iver k
nR
dma

r
Rive
W il

r
West Alligato

Nangalawurr
Nawulandja Spirit Cave,Yiboiog
Ki’na Anbangbang
Ji
m

m
Ji

Cr
De

at
k h
ee

Ad

er Bala-uru
d

C re
ek

Nauwalabila 1
Djuwarr 1
So

Nawarla
uth

Gabarnmang
Allig
tora
R iv
er

M
ar
yR
Katherin e River

iv
er

son Rive
gus
i v er

Fer
r

wetlands
eR
n
eri
Kath

0 30
Katherine km
Figure 2. Nawarla Gabarnmang, showing stone pillars,
Figure 1. Location of Nawarla Gabarnmang, with excavation Square B in the foreground. The site is
northern Australia. extensively decorated with rock paintings.

‘EAST’ ‘SOUTH’
cm
0
SU1 xu 1 xu 1 SU1
xu 2
xu 2
SU2 xu 3 xu 3 SU2 bedrock
xu 4 xu 4
xu 5 xu 5
10 xu 6 xu 6
xu 7 xu 7
xu 8 xu 8 SU3
SU3 xu 9 xu 9
xu 10 xu 10
xu 11 xu 11
20 xu 12 xu 12 not excavated
xu 13 xu 13
xu 14 xu 14
xu 15 xu 15
xu 16 xu 16
xu 17 xu 17
xu 18 xu 18
xu 19 xu 19
xu 20 xu 20
xu 21 SU4
30 xu 21 bedrock
xu 22 xu 22
pedestal
xu 23 SU4 xu 23
xu 24 (not excavated)
xu 25 xu 24
xu 26 xu 25
40 xu 26
xu 27 xu 27
xu 28 SU5
xu 28
xu 29 SU5
bedrock xu 29
xu 30 xu 30
50
bedrock ? bedrock xu 31
xu 31
xu 32 xu 32
rock xu 33
xu 34 xu 33

60 xu 35
xu 34
charcoal xu 36 xu 35
not excavated
xu 36
Figure 3. East and south
sections of Square A,
root
70 not excavated

rootlets (indicative) Nawarla Gabarnmang.

6
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land

Figure 4. A) Side, dorsal and ventral views of the Pleistocene Nawarla Gabarnmang ground-edge axe fragment from
Square A. B) Close-up view of the ground surface of the Pleistocene ground-edge axe fragment. C) Close-up view of the
ground surface (away from the cutting edge) of the late Holocene ground-edge axe from Square B. D) Close-up view of
the ground surface (at the cutting edge) of the late Holocene ground-edge axe from Square B. E) Late Holocene ground-
edge axe from Square B, Nawarla Gabarnmang.
7
 Jean-Michel Geneste et al.
tend not to grow today anywhere within the driplines
at the site. Sediments in each of the five Stratigraphic
Units consist of compact and well-consolidated loamy
sands produced by chemical alteration of the sur-
rounding sandstone bedrock and accumulated aeolian
sediments. The terminal Pleistocene sediments of
SU4, incorporating the last glacial maximum into the
Pleistocene–Holocene boundary, consist of a layer of
exfoliated sandstone roof-fall rocks infilled with com-
pact loamy sands sealing the underlying SU5 (which
contains the fragment of ground-edge axe reported
here) from overlying sediments. While a proportion
of these sediments arguably originated directly from
the bedrock, in particular from minor fallen roof or
pillar material, another, more important proportion
originated externally, transported a short distance
through aeolian processes to be deposited into the site
and thus forming significant proportions of the soft
deposit. At the very base of Square A and upwards for
a few centimetres only above bedrock (but well below
the lowermost radiocarbon determinations), a concen-
tration of rounded gravel-sized sediments indicates
transport along narrow and shallow channels etched
into the bedrock by water action during a period prior
to the deposition of the present soft sediments.
Throughout much of the excavated sequence
above the thin concentration of very basal gravels,
stone artefacts mainly made of quartzite and chert
number in the thousands, but preliminary analysis
of the entire excavated deposit indicates that the
fragment of ground-edge axe reported here is a rare
example of volcanic rock found. We note that high-
quality quartzite for the manufacture of long cutting
edges is available in large quantities within the site
itself, whose ceiling was also a quartzite quarry.
Four charcoal samples were taken for radio-
carbon determination from the lower excavation units
(XU27 to XU31 in SU5) surrounding the fragment of
ground-edge axe. These four near-basal determina-
tions continue a sequence of an additional 14 AMS
radiocarbon ages from the surface down, in good
chrono-stratigraphic order. These samples were
prepared at the Waikato Radiocarbon Dating Labo-
ratory in New Zealand using a standard acid-base-
acid treatment (Olsson 1986). Graphite targets were
prepared by the reduction of cryogenically-purified
CO2 to graphite, using hydrogen over pre-cleaned
iron powder at 550°C. The radiocarbon results were
calibrated using IntCal09 (Reimer et al. 2009) in OxCal
v4.1.7 (Bronk Ramsey 2010) and produced a combined
statistically indistinguishable age of 36,230–35,880
and 35,580–34,850 cal. bp, equivalent to a mean age of
35,400±410 cal. bp for the four SU5 samples (95.4 per
cent prob.; χ23:0.05 = 1.232 < 7.815).
8
Technical description of the fragment of
ground-edge axe
The stone raw material is crystalline, of a basaltic type
with reasonably fine crystals. The surface is beige in
colour and chemically altered. It has been affected by
contact with clay deposits, and residual red staining
in spaces between grains suggests the presence of
iron oxide. Thin-sectioning for petrographic analysis
cannot be undertaken because of the high historical
and cultural value of the piece.
The object is a thin flake detached by percussion
from a straight edge. It measures 2.55 cm × 3.93 cm
and 0.61 cm in thickness (Fig. 4). The ventral surface
reveals the crystalline structure of the rock. The point
of impact from which the flake was detached can be
clearly distinguished. The flake exhibits no evidence
of further modification following detachment. Its mor-
phology shows that it was removed from the upper
surface of the original tool. The percussion angle was
close to 90° with the adjacent side, and the precision
of the percussion angle and force combined with the
thinness of the flake indicates that it was removed
intentionally.
The dorsal surface of the flake is smooth, slightly
convex in width and slightly less convex in length.
Under acute-angled light it exhibits grinding stria-
tions along its entire length, created by a material
necessarily of greater hardness and abrasiveness
than the flake itself. The flake was ground smooth
with a relatively fine-grained raw material. Likely
candidates are the quartz sandstone or ortho-
quartzite bedrock of the Kombolgie Formation, found
across much of Arnhem Land and of which Nawarla
Gabarnmang is a part.
The distribution and structure of the abraded
surface indicates that grinding was intentional. The
abraded surface contains a homogeneous central pla-
teau, ground smooth along its upper reaches, below
which along its slope the relief becomes coarser with
an increasing density of shallow surface depressions,
vestiges of prior hammer dressing or of the original
surface of the rock.
On this plateau can be distinguished under
binocular microscope (10× magnification) sets of
superimposed and criss-crossing striations notable
for their short length (0.5 cm on average). Some sets
are more pronounced, in particular among the longer
striations. This suggests either a disparity in the size
of the grains on the grinding stone used (if this was
the case this disparity should also be seen in their
distribution across the flake, which it is not) or, more
likely, variations in the down-force employed during
grinding, further re-enforcing an intentional cause.
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land
The ends of the abraded striations along the edge
of the flake are flush with the detachment edge. There
is no evidence of subsequent modification in the form
of retouch or subsequent grinding. We are therefore
dealing with a fragment of an originally larger ground
object that can be further investigated microscopically.
The cutting edge of ground-edge stone axes is
generally characterized by their unidirectional final
abrasion, either longitudinal or transverse (Fig. 4D).
The ground flake from Nawarla Gabarnmang cannot
therefore come from the cutting edge itself of the axe,
given the positioning of its abrasions and the multi-
directional distribution of its striations. Ground-edge
stone axes are usually only partially ground, and in
Arnhem Land no examples of fully-ground tools are
known. The axe is usually manufactured from a large
flake or core blank initially shaped by percussion. On
some axes only the cutting edge is ground, but grind-
ing can also be found on the adjacent sides. For an axe
to work efficiently, the zone of contact with the worked
material needs to be smooth. Residual surface protru-
sions thus need to be flattened and ground smooth to
eliminate friction points that would limit the capacity
of the cutting edge to penetrate. Any coarseness along
the cutting edge would counter the work of the axe as
a tool of penetration. Consideration of these functional
characteristics can be seen on the known complete
ground-edge axes from Arnhem Land, one example
of which was also excavated at Nawarla Gabarnmang,
in late Holocene levels in XU24 of Square B (Fig. 4E).
This late Holocene axe enables better under-
standing of the Pleistocene ground flake, as the most
salient aspect of the former’s cutting edge is that it
has been reduced by grinding against a fine-grained
abrasive material, with very short and highly localized
criss-crossing movements resulting in sets of intersect-
ing striations (Fig. 4C) reminiscent in structure and
distribution of those on the Pleistocene ground flake
(Fig. 4B). The left side of the late Holocene axe’s cutting
edge has a morphology that is entirely comparable to
the detached Pleistocene flake, wider than long, from
a removal surface partially ground and detached by
percussion from an unground rectilinear edge.
Although the Pleistocene flake does not intersect
the cutting chord of the original tool, the sum of its
observed technological traits excludes the possibi-
lity that we are dealing with an artefact that was
ground after the production of the flake, or a portion
of grinding stone or percussion tool, and no other
similar Aboriginal object is known from ancient or
more recent, ethnographic times other than ground-
edge axes. Therefore, we can safely conclude that the
Pleistocene ground flake from Nawarla Gabarnmang,
wider than it is long, with large straight butt and a
9
dorsal surface evenly ground to a slight convexity
prior to its detachment, exhibits all the characteris-
tics of a lateral fragment of the cutting surface of a
ground-edge axe. The percussion angle is close to 90°
with the dorsal surface, and the precision of applica-
tion of percussion force combined with the thinness
of the removal indicates an intentional action. The
removal of the flake was aimed at thinning a side
of the axe, near its lateral margin and fairly close
to the effective cutting edge so as to remove a small
protuberance indicated by a concentration of short,
multidirectional striations.
Archaeological implications
The advent of ground-edge axes around the world has
been widely heralded as a technological achievement
of ecological, environmental and adaptive innova-
tion and efficiency that enabled the cutting of trees
for clearing, the working of timber, branches or bark,
extraction of prey from natural hollows, hollowing
of dugout canoes, working of wood towards the
construction of tools and complex installations and
the like. In the Northern Hemisphere, explanations
for the onset of edge grinding typically evoke the
exploitation of forest landscapes where axes are seen
as highly efficient, adaptive tools that emerged as a
result of increasing post-glacial demographic pres-
sures necessitating deforestation and the clearing of
land in preparation for agriculture and a sedentary
lifestyle (Lewin & Foley 2003; Toth & Schick 2009).
Other explanations would see ground-edge axes in
Neolithic Europe used in wood-working and for cut-
ting thinner items than tree-trunks. There is a very
rich organic material culture in northern Europe that
used wood for anything from houses to snow-shoes
or baskets, and these small but sharp axes would have
been useful for those.
In Australia, explanations revolving around the
clearing of early Holocene forests for agriculture do
not hold, firstly because ground-edge axes precede
by many millennia the last glacial maximum, and
secondly because they were not typically used for
deforestation in a move towards agriculture and
sedentism. On the other hand, ground-edge axes were
here produced into the late nineteenth century ad, and
continue to be made today in the neighbouring island
of New Guinea (Pétrequin & Pétrequin 1993).
Ethnographic sources indicate that the exchange
of ground-edge axes was integrated with regional
circulation networks of prestige goods, kinship rela-
tions and ceremonial activities informed by local
and regional cosmologies (see Paton 1994; Sharp
1952; Thomson 1949). Ground-edge axes played an
 Jean-Michel Geneste et al.
important role in the dissemination of cosmologi-
cal understandings of the world and their symbolic
representations. The universal lesson here is that
the construction, use and geographical circulation
of objects such as ground-edge axes are mediated by
the way that people understand the world to operate
cosmologically, and in doing so the object becomes a
vehicle of socially-mediated symbolic storage. Under-
standing the manufacture, trade, function and use of
ground-edge axes in Australia, and by implication
elsewhere, requires us to go beyond their utilitarian
roles as chopping tools to see them as objects of high
social repute accessed between groups through the
way their raw materials and originary landscapes
were imbued with meaning and cosmological powers
that connect people to places and to the ancestors (cf.
Brumm 2010). Owing to their highly-localized position
in the landscape, stone quarries, and in this regard
ground-edge axe quarries in particular owing to their
intensified and therefore embedded work effort as
highly curated quarried objects, also externalize ter-
ritorial understandings and their culturally-specific
cosmological dimensions (Thomson 1949). It is this
spatio-temporal landscape inscription of objects that
takes place when stone-axe quarries are accessed,
axes are manufactured, and exchanged, conferring
extended notions of value and utilitarianism onto their
handlers that go beyond their immediate economic
worth as chopping tools. In those exchanges came
the passing of values, meanings and symbolic under-
standings not only of the objects themselves, but of
the lived and meaningful landscapes which connected
groups with their resources as cosmological-territorial
inscriptions.
In relating the ancient, 35,500-year-old fragment
of ground-edge axe from Nawarla Gabarnmang to
ethnographic observations of the cosmological sig-
nificance of axes in Aboriginal Australia, we are not
arguing for continuity in meaning across this vast
period of time or from Australia to Europe, but for
the universal point that objects are always socially
meaningful — and thus always have social value and
thus social use-value over and above their more conven-
tionally considered ‘economic efficiency’ — through
the way that places and things are cosmologically
mediated in social interaction. Here we consider
this issue through the example of stone axes, as a
tool conventionally thought of only for its extractive
potential. The above principle of the social value of
objects applies just as much to things made and held
by peoples of Europe, the Americas or Australia, as it
does to cognitively fully-modern peoples of the past
and of the now (for greater exposition of this topic, see
e.g. Appadurai 1986; Dant 1999).
10
The question remains as to why, if at all, ground-
edge axes deserve special consideration greater than
that applied to flaked artefacts as objects of symbolic
storage. We suggest that in themselves, flaked and
ground tools are no different in this sense. What is
different in the case of ground-edge axes is two-fold:
1. The social aesthetics of symmetry. Ground-edge axes
are produced as symmetrical or near-symmetrical
objects, and this aesthetic dimension gives them
added symbolic weight (we would extend this
same argument to other symmetrical objects
including Acheulian ‘hand-axes’, as others have
done before us). Grinding is a short-cut to sym-
metry; here it is not so much grinding itself that is
at issue, for symmetry can also be created by flak-
ing, but that grinding facilitates symmetry beyond
flaking for most stone workers. Their relatively
large size enhances their potential to be seen at a
distance, and in this their aesthetic value crosses
from the hidden to the public, affording ground-
edge axes a heightened sense of social prestige.
2. The symbolism of efficiency. The success of an axe as
a cutting or chopping tool resides in the angular
smoothness of its cutting edge. Other than melting
metal (and prior to the advent of recent synthetic
products), stone grinding is the most efficient
technological process known for creating a smooth
cutting edge on an axe. This edge-efficiency
renders ground-edge axes not only the most effi-
cient chopping tools, but also gives them a high
symbolic value as ‘state of the art’.
Ground-edge axes are thus social investments in
aesthetism and symbols of efficacy; what is at stake
here is not only the productive efficiency of the tool,
but its socially-reproductive value (symbolic value
enhances the maintenance of knowledge). Symbolic
storage operates in all objects, not only ground-edge
axes, but the latters’ social symbolism is enhanced
by the foregoing qualities. Because the extraction
of stone from localized quarries and the prolonged
working of the stone into ground edges is geared to
the manufacture of a non-expedient tool that can be
used anytime in the near or distant future, a ground-
edge axe’s immediate work value transforms the tool’s
delayed-use value through its prolonged social value.
The temporal space in this delay of value from the
immediate work of manufacture to a potential future
use allows for its social exchange, and in doing so
for its radical re-orientation from a functional tool
to a symbol of prestige where the material attracts
and embodies social meaning replete with flexible,
re-definable use-values. Simply put, a ground-edge
axe’s functional value remains dormant until the axe
is put into use, its on-going (i.e. non-dormant) mate-
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land
rial existence as possession not only allowing but also
requiring, by virtue of its retention, the object to accrue
meaning in its social embeddedness. This social reten-
tion of the object, used or unused, directs its social
value; and the knowledge of its social presence and
material possibility from then-on causes a sociality of
its material presence even in the event of its particular
absence (i.e. through memory + meaningfulness, the
damage, destruction or disappearance of individual
axes attain social meaning as a momentum towards
curation, renewal and access of ground-edge axes into
the future). In these senses a ground-edge axe is an
example of heightened ‘symbolic storage’.
Edge-grinding enhanced symbolic storage in
stone in a new way evident only after the arrival
of Homo sapiens along the western rim of the Pacific,
indicative of the continued cognitive evolution of, or
at least the application of symbolic cognitive abilities
to a new material realm by, modern humans following
the Out-of-Africa 2 migration.
Conclusion
Because of the high investment in effort and time
required for their manufacture as an efficient, large,
socially-visible, portable symmetrical tool, as well
as the limited distribution of raw material sources,
ground-edge axes attain high regional symbolic
value steeped in the cosmologies of their makers and
receivers, rendering them worthy of exchange as
socially-valued objects, often over very long distances
(e.g. Brumm 2010). This social layer of heightened
non-utilitarian value, we argue, represents a level of
human cognition and symbolism that grew from the
Out-of-Africa 2 migration as early biologically fully-
modern Homo sapiens populations adapted to new
landscapes and their social configurations.
The discovery of the Nawarla Gabarnmang
ground-edge axe fragment confirms that the oldest
ground-edge axes appeared in northern Australia
around 35,500 years ago, many millennia before their
independent invention in the contiguous region of
Africa, Europe and western Asia. The implication of
its timing and of the geographical distribution of the
world’s oldest known ground-edge axes suggests a
continued process of symbolic evolution subsequent
to the Out-of-Africa 2 migration and associated with
the development of behavioural modernity. That the
earliest, Pleistocene ground-edge axes emerged in a
utilitarian context untethered to post-glacial reforesta-
tion, population packing, anthropogenic deforestation,
the advent of agriculture or sedentism indicates that
other kinds of explanations are required. The Austra-
lian ethnographic evidence is salutary in this regard.
11
With the Nawarla Gabarnmang fragment of
ground-edge axe, now the oldest in the world, the
excavated sites of Arnhem Land have arguably pro-
duced one of the most significant set of ground-edge
axes of the Pleistocene. This poses questions as to how
and why it is here along the western Pacific rim that
stone edge-grinding appears to have first evolved,
without denying the possibility of subsequent inde-
pendent inventions elsewhere. We conclude with
three observations that enable a better appraisal of
this question.
1. Hunter-gatherer populations of the Southern
Hemisphere adopted the grinding of rock for the
manufacture of ground-edge axes well before their
counterparts in Mesolithic and Neolithic Europe.
This discovery has archaeological implications
very different from those that have prevailed in
the standard explanations of the Northern Hemi-
sphere.
In the Northern Hemisphere excluding east
Asia, and in Europe in particular, the grinding of
stone occurs only during the Upper Palaeolithic
after 30,000 bp, as evidenced by examples of
mobiliary and fixed (rock) art. Yet the application
of grinding on the edges of axes appears much later,
between 8500 and 5000 bp and in association with
changing environmental, ecological and economic
conditions. Thus late Mesolithic–Neolithic tech-
nological innovations such as the onset of edge-
grinding have been directly attributed to the func-
tional practicalities of intensified food production,
sedentism, and restructured subsistence impera-
tives (Lewin & Foley 2003; Toth & Schick 2009).
These are economic and adaptive approaches to
technological innovation that, while themselves
relevant components of change, have usually (but
not always) come as exclusive explanations at the
cost of symbolic interpretations of a kind seen
to operate ethnographically in other parts of the
globe. The problem with such explanations is that,
by themselves, they silence meaning outside of
economic rationalism. At Nawarla Gabarnmang,
and Arnhem Land generally, we argue that the
onset of the world’s earliest ground-edge axe
technology some 35,500 years ago was at its core
embedded in symbolic dimensions that allowed
people to interact meaningfully with each other
and with their landscapes and their products, in
the process enhancing the socially-structured and
value-laden movement of goods and the construc-
tion of socially meaningful territory, stone sources
included. We argue that here ground-edge axes
were adopted because they aided subsistence and
made certain tasks more efficient, but in doing so
 Jean-Michel Geneste et al.
they took on important new social and symbolic
roles that served to cement social and environmen-
tal (territorial) relations.
2. Irrespective of their spatial and temporal distri-
butions and of our explanatory frameworks, the
grinding of stone in the production of ground-edge
axes is embedded in the gambit of social realities.
In this sense ground-edge axes remain inseparable
from the lithic raw material sources from which
they came: their lives endure as ancestry, or histori-
cal artefacts. This is because a ground-edge axe’s
originary landscape is symbolically transported
across the landscape as the tool is moved around
by people (whether carried by an individual or
traded between individuals and groups), spatially
expanding its meaningfulness (and modifying its
specific meanings) along the way. Such objects live
only in the eyes of their owners, onlookers and the
communities in which they make their mark.
The onset of ground-edge technology some
35,500 years ago in northern Australia, following
the Southern Arc dispersal out of Africa, and the
subsequent emergence of similar technologies in
other parts of the world, indicate a common cog-
nitive and symbolic process of spatial expansion
and transference of meaning through and into the
material, a process significantly enhanced among
fully modern Homo sapiens.
3. In light of the above, subsistence-based economic
arguments are not, in themselves, sufficient to
understand the origins and adoption of ground-
edge technology towards the manufacture of
stone axes, although questions of tool efficiency
are an important component of the issue. While
ground-edge axes arguably possess more efficient
cutting or chopping edges, their high manufactur-
ing and maintenance costs, coupled with their use
as occasional chopping rather than systematic
forest-clearance tools in Australian contexts, high-
light their importance as items of social value that,
while requiring a consideration of their extractive
potential, are not reducible to such functional,
extractive abilities. Here a ground-edge axe’s
symbolic value is other and, in this, more than the
sum of its functional values. Ground-edge axes
are highly-curated, more-or-less symmetrical and
lasting tools usually made of volcanic raw materi-
als found in highly-localized sources subject to
territorial circumscription, unlike sandstone or
quartzite tools which, in Arnhem Land at least,
can be obtained from widespread sources across
a vast landscape. The considerable work-time
invested in the manufacture of a ground-edge axe
extends its value beyond its immediate production
12
and use; such time investment extends the period
of manufacture into a future use-time through
the tool’s social value expressed as maintenance
and care. This social value is itself encoded in the
cosmological meanings attached to sources as land-
scape features that are themselves given meaning
through origin stories and the like, while at the
same time also encapsulating the tool’s functional
utility. Social value thus becomes embedded into
an object’s exchange value through its extraction,
transformation, use-potential, curation, main-
tenance and trade in a network of inter-personal
relations. In this sense, as highly-curated and
shaped objects originating from localized sources
subject to territorial circumscription, ground-edge
axes gain a level of symbolic value unmatched by
more expedient tools typical of most percussion-
flaked artefacts. The act of owning and carrying a
ground-edge axe thus gives the owner more than
an economic advantage by which to cut down
trees or work wood, but also social value as the
extended work-value of an object originating from
a source that itself is already imbued with mean-
ing and prestige through its territorially-mediated
cosmological identity.
The implications of these findings are either that:
1. edge-grinding to make cutting or chopping edges
first developed in Sahul, past the eastern edge of
Wallacea and the Sunda shelf where a sea crossing
was required to the east of the Asian continent.
Such a scenario would see the earliest invention
of edge-grinding taking place geographically past,
and therefore after, the terrestrial completion of
the Out-of-Africa 2 migration to the eastern Asian
seaboard;
2. the global temporal pattern, currently showing that
the earliest ground-edge stone tools in northern
Australia predate similar tools in eastern Asia by
1000 to 4000 years, does not reflect the true geo-
graphical chronology, a problem of sampling limi-
tations at a global scale; in this case yet-unknown
east Asian ground-edge tools would predate the
earliest Australian evidence; or
3. an as-yet undiscovered common ancestral axe-
making tradition exists at the geographic way-
point between colonization of eastern Asia and
Sahul.
If the early Australian and east Asian ground-edge
technology are historically related (i.e. the two
regional cases are homologous and thus are not
independent inventions), which is suggested by their
spatial proximity and near-contemporaneity, Option 1
would indicate initial invention in Sahul (Australia-
New Guinea) with a back-migration subsequently
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land

introducing the technology into eastern Asia via Acknowledgements

Sunda. Option 2 would see an original invention
of edge-grinding technology in eastern Asia soon We thank the Jawoyn Association Aboriginal Corporation
followed by its dispersal into Australia by incoming Committee, and Margaret Katherine, Sybil Ranch, Preston
populations. This would indicate evidence for the Lee, Wes Miller and Ray Whear, for inviting and caring for
us during research in their country. Thanks to Ben Gunn and
arrival of people bearing ground-edge axes into an
Lance Syme for assistance in the field, Luke Weatherley for
already-occupied Australia, presumably from Sunda, sorting, Laure Dubreuil of the Department of Anthropology,
shortly before 35,000 years ago, some 15,000 years Trent University, Canada for advice concerning grinding
after the initial colonization of Sahul. Option 3 would stones of the Near East, and Chris Clarkson, Francesco
see future locations of ground-edge axe manufacture d’Errico and two anonymous referees for useful comments
located close to the way-point between a branching of on earlier drafts. We thank the Australian Research Council
dispersing populations into eastern Asia and Sahul. for QEII Fellowship DP0877782 to BD and ARC Linkage
This would also imply that early axe-making techno- Grant LP110200927 to enable this research to take place.
logy will one day be detected in Southeast Asia.
At this stage, which of these three options is Jean-Michel Geneste
correct is difficult to determine with the available Centre national de la préhistoire, ministère la culture
evidence. Insufficient archaeological sampling in et de la communication
eastern Asia and northern Australia, and insufficient Univ. Bordeaux, CNRS, PACEA, UMR 5199
precise and accurate dating, render also insufficient F-33400 Talence
the resulting spatial history. One hint as to the more France
likely option comes not from the earliest evidence Email: jean-michel.geneste@culture.gouv.fr
of edge-grinding itself, but rather of flaked waisted
(as evidence of hafting) axe technology, the earliest Bruno David
evidence also coming from Sahul. At Bobongara in School of Geography and Environmental Science
the Huon Peninsula (Groube et al. 1986) and at Kosipe Monash University
(Summerhayes et al. 2010) in the highlands of Papua Clayton, Victoria 3800
New Guinea, unground, symmetrical flaked and Australia
waisted axes have been recovered from archaeological Email: bruno.david@monash.edu
contexts securely dated to ≥40,000 and 44,000–49,000
years ago respectively. In contrast, the earliest flaked Hugues Plisson
and waisted stone axes in eastern Asia come from Univ. Bordeaux, CNRS, PACEA, UMR 5199
Hoabinhian and Middle Jomon sites, and are dated to F-33400 Talence
the early- and mid-Holocene respectively (for reviews, France
see Anderson & Summerhayes 2008; Golson 2001). On Email: hugues.plisson@u-bordeaux1.fr
this basis, the current evidence suggests that ground-
edge axes emerged in situ in northern Australia from Jean-Jacques Delannoy
locally invented waisted (hafted) axe technology some EDYTEM - UMR 5204 du CNRS
8000 to 15,000 years after the invention of hafted flaked Environnements, Dynamiques et Territoires de la Montagne
axes in Sahul rather than in eastern Asia. Golson Centre Interdisciplinaire Scientifique de la Montagne
(2001) suggested that the spread of what were then Université de Savoie F 73376
apparently very early waisted and ground-edge Le Bourget du Lac Cedex
axes across northern Australia-Papua New Guinea France
indicates the presence of an ancient and innovating Email: Jean-Jacques.Delannoy@univ-savoie.fr
sphere of cultural interaction in northern Sahul. In
any case, there is no indication at this stage that edge- Fiona Petchey
grinding technology arrived in Australia with the first Radiocarbon Dating Laboratory
Australians, nor is there indication of its invention University of Waikato
in eastern Asia around the time of the colonization Hamilton, 3240
of Australia. On this basis, and with an absence of New Zealand
earlier evidence elsewhere in the world, the invention Email: fpetchey@waikato.ac.nz
of edge-grinding technology at or subsequent to the
completion of the Out-of-Africa 2 migration seems References
secure, with implications for developing symbolic
Ambrose, S., 1998. Chronology of the Later Stone Age and
practices as outlined in this article.
13
 Jean-Michel Geneste et al.
food production in East Africa. Journal of Archaeological
Science 25, 377–92.
Anderson, A. & G. Summerhayes, 2008. Edge-ground and
waisted axes in the western Pacific islands: implica-
tions for an example from the Yaeyama Islands, south-
ernmost Japan. Asian Perspectives 47, 45–58.
Anton, S. & C.C. Swisher, 2004. Early dispersals of Homo
from Africa. Annual Review of Anthropology 33, 271–96.
Appadurai, A., 1986. The Social Life of Things: Commodities
in Cultural Perspective. Cambridge: Cambridge Uni-
versity Press.
Backwell, L.R. & F. d’Errico, in press. Drimolen: additional
evidence of early hominid bone tool use and modifi-
cation. First attempt at exploring inter-site variability.
Paper presented at the Palaeontological Society of
Southern Africa meeting, Matjiesfontein, 11–15 Sep-
tember. Palaeontologia Africana 44.
Balme, J., I. Davidson, J. McDonald, N. Stern & P. Veth, 2009.
Symbolic behaviour and the peopling of the southern
arc route to Australia. Quaternary International 202,
59–68.
Bar-Yosef, O., 1998. The Natufian culture in the Levant,
threshold to the origins of agriculture. Evolutionary
Anthropology 6(5), 159–77.
Bar-Yosef, O., 2002. The Natufian culture and the Early
Neolithic: social and economic trends in southwestern
Asia, in Examining the Farming/Language Dispersal
Hypothesis, eds. P. Bellwood & C. Renfrew. (McDonald
Institute Monographs.) Cambridge: McDonald Insti-
tute for Archaeological Research, 113–26.
Bellwood, P., 1997. Prehistory of the Indo-Malayan Archipelago.
Honolulu (HI): University of Hawai’i Press.
Beaumont, P.B., H. De Villiers & J.C. Vogel, 1978. Modern
man in sub-Saharan Africa prior to 49,000 years bp:
a review and evaluation with particular reference to
Border Cave. South African Journal of Science 74, 409–19.
de Beaune, S.A., 2000. Pour une archéologie du geste: broyer,
moudre, piler, des premiers chasseurs aux premiers agri-
culteurs. Paris: CNRS.
de Beaune, S.A., 2003. Du grain à moudre sur les Néander-
taliens. La Recherche 360, 56–9.
de Beaune, S.A., 2004. The invention of technology: prehis-
tory and cognition. Current Anthropology 45, 139–64.
Bickerton, D., 2003. Symbol and structure: a comprehen-
sive framework for language evolution, in Language
Evolution, eds. M.H. Christiansen & S. Kirby. Oxford:
Oxford University Press, 77–93.
Bird, M.I., C.S.M. Turney, L.K. Fifield et al., 2002. Radio-
carbon analysis of the early archaeological site of
Nauwalabila 1, Arnhem Land, Australia: implications
for sample suitability and stratigraphic integrity.
Quaternary Science Reviews 21, 1061–75.
Bon, F., 2009. Préhistoire: la fabrique de l’homme. Paris: Seuil.
Braun, D.R., T.W. Plummer, P.W. Ditchfield, L.C. Bishop
& J.W. Ferraro, 2009. Oldowan technology and raw
material variability at Kanjera South, in Interdisciplin-
ary Approaches to the Oldowan, eds. E. Hovers & D.R.
Braun. (Vertebrate Paleobiology and Paleoanthropo-
logy.) Dordrecht: Springer, 99–110.
Bronk Ramsey, C., 2010. OxCal Program v. 4.1.7. Oxford
14
University of Oxford, Radiocarbon Accelerator Unit.
Brook, B.W., D.M.J.S. Bowman, D.A. Burney et al., 2007.
Would the Australian megafauna have become
extinct if humans had never colonised the continent?
Comments on ‘A review of the evidence for a human
role in the extinction of Australian megafauna and
an alternative explanation’ by S. Wroe and J. Field.
Quaternary Science Reviews 26, 560–64.
Brumm, A., 2010. ‘The falling sky’: symbolic and cosmo-
logical associations of the Mt William greenstone
axe quarry, central Victoria, Australia. Cambridge
Archaeological Journal 20(2), 179–96.
Carciumaru, M., E.C. Nitu & M. Tutuianu-Carciumaru, 2010.
Témoignages symboliques au Moustérien. Unpub-
lished paper presented at the IFRAO International
Congress, Foix-Tarascon-sur-Ariège.
Clark, J.D. & M.R. Kleindienst, 2001. The Stone Age cultural
sequence: terminology, typology and raw material,
in Kalambo Falls Prehistoric Site, vol. III: The Earlier
Cultures — Middle and Earlier Stone Age, ed. J.D. Clark.
Cambridge: Cambridge University Press, 34–65.
Clottes, J., 2007. At the origins of art: new discoveries of
decorated caves, in Discovery! Unearthing the New
Treasures of Archaeology, ed. B.M. Fagan. London:
Thames and Hudson, 114–19.
Cole, N.A., A. Watchman & M.J. Morwood, 1995. Chro-
nology of Laura rock art, in Quinkan Prehistory: the
Archaeology of Aboriginal Art in S.E. Cape York Peninsula,
Australia, eds. M.J. Morwood & D.R. Hobbs. (Tempus
3.) St Lucia: The University of Queensland, Anthropol-
ogy Museum, 147–61.
Conard, N.J., 2009. A female figurine from the basal Auri-
gnacian of Hohle Fels Cave in southwestern Germany.
Nature 459, 248–52.
Conard, N.J. & M. Bolus, 2003. Radiocarbon dating the
appearance of modern humans and the timing of
cultural innovations in Europe: new results and new
challenges. Journal of Human Evolution 44, 331–71.
Conard, N.J. & M. Bolus, 2008. Radiocarbon dating the
late Middle Paleolithic and the Aurignacian of the
Swabian Jura. Journal of Human Evolution 55, 886–97.
Conard, N.J., G. Dippon & P. Goldberg, 2003. Chronostrati-
graphy and archeological context of the Aurignacian
deposits at Geißenklösterle, in The Chronology of the
Aurignacian and of the Transitional Technocomplexes: Dat-
ing, Stratigraphies, Cultural Implications, eds. J. Zilhão
& F. d’Errico. (Trabalhos de Arqueologia 33.) Lisbon:
Instituto Português de Arqueologia, 165–76.
Dant, T., 1999. Material Culture in the Social World: Values,
Activities, Lifestyles. Buckingham: Open University
Press.
David, B., 2002. Landscapes, Rock-art and the Dreaming: an
Archaeology of Preunderstanding. London: Leicester
University Press.
Deacon, H.J. & J. Deacon, 1999. Human Beginnings in South
Africa: Uncovering the Secrets of the Stone Age. Clare-
mont: David Philip Publishing.
Derevianko A.P., M.V. Shunkov & P.V. Volkov, 2008. A
Paleolithic bracelet from Denisova Cave. Archaeology,
Ethnology and Anthropology of Eurasia 34(2), 13–25.
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land
d’Errico F. & L. Backwell, 2009. Assessing the function of
early hominin bone tools. Journal of Archaeological
Science 36, 1764–73.
d’Errico, F., C. Henshilwood, G. Lawson et al., 2003. Archaeo-
logical evidence for the emergence of language, sym-
bolism, and music — an alternative multidisciplinary
perspective. Journal of World Prehistory 17(1), 1–70.
d’Errico, F., M. Vanhaeren, C. Henshilwood et al., 2009.
From the origin of language to the diversification of
languages: What can archaeology and palaeoanthro-
pology say?, in Becoming Eloquent: Advances in the
Emergence of Language, Human Cognition, and Modern
Cultures, eds. F. d’Errico & J.-M. Hombert. Amster-
dam: John Benjamins Publishing Company, 13–68.
d’Errico, F., C. Henshilwood, M. Vanhaeren, L. Backwell,
R. Garcia Moreno & R. Rifkin, 2010. The Origin of
Symbolic Material Culture: Models, Methods, Data
and Research Perspectives. Unpublished paper prese-
anted at ‘Pleistocene Art of the World Congress 2010’,
International Federation of Rock Art Organisations
(IFRAO), 6–11 September, Tarascon-sur-Ariège and
Foix, France.
Field, J.S., M.D. Petraglia & M.M. Lahr, 2007. The southern
dispersal hypothesis and the South Asian archaeologi-
cal record: examination of dispersal routes through
GIS analysis. Journal of Anthropological Archaeology
26, 88–108.
Fullagar, R. & J. Field, 1997. Pleistocene seed-grinding
implements from the Australian arid zone. Antiquity
71, 300–307.
Gamble, C. & F. Coward, 2009. Big brains, small worlds:
material culture and the evolution of mind. Philosophi-
cal Transactions of the Royal Society Series B 363, 1969–79.
Gillespie, R. & B.W. Brook, 2006. Is there a Pleistocene
archaeological site at Cuddie Springs? Archaeology in
Oceania 41, 1–11.
Gillespie, R. & B. David, 2001. The importance, or impotence,
of Cuddie Springs. Australasian Science 22(9), 42.
Golson, J., 2001. New Guinea, Australia and the Sahul con-
nection, in Histories of Old Ages: Essays in Honour of
Rhys Jones, eds. A. Anderson, I. Lilley & S. O’Connor.
Canberra: Pandanus Books, 185–210.
Groube, L., J. Chappell, J. Muke & D. Price, 1986. A 40,000
year old occupation site at Huon Peninsula, Papua
New Guinea. Nature 324, 453–5.
Habgood, P.J. & N.R. Franklin, 2008. The revolution that
didn’t arrive: a review of Pleistocene Sahul. Journal of
Human Evolution 55, 187–222.
Henshilwood, C.S. & F. d’Errico, 2005. Being modern in the
Middle Stone Age: individuals and innovation, in The
Individual Hominid in Context: Archaeological Investiga-
tions of Lower and Middle Palaeolithic Landscapes, Locales
and Artefacts, eds. C. Gamble & M. Porr. Abingdon:
Routledge, 244–64.
Henshilwood, C.S., F. d’Errico, R. Yates et al., 2002. The emer-
gence of modern human behavior: Middle Stone Age
engravings from South Africa. Science 295, 1278–80.
Hernek, R., 2005. Nytt ljus på Sandarnakulturen: om en
boplats från äldre stenåldern i Bohuslän. (Gotarc Series
B, Gothenburg Archaeological Thesis 38; Coast to
15
Coast-Books 14.) Göteborg: University of Göteborg.
Hovers, E. & D.R. Braun (eds.), 2009. Interdisciplinary
Approaches to the Oldowan. (Vertebrate Paleobiology
and Paleoanthropology.) Dordrecht: Springer.
Jones, R. & I. Johnson, 1985. A Deaf Adder Gorge: Lindner
Site, Nauwalabila 1, in Archaeological Research in
Kakadu National Park, ed. R. Jones. (Special Publication
13.) Canberra: Australian National Parks and Wildlife
Service, 165–227.
Leroi-Gourhan, A., 1995. Préhistoire de l’Art Occidental. Paris:
Citadelles et Mazenod.
Lewin, R. & R.A. Foley, 2003. Principles of Human Evolution.
Oxford: Blackwells.
Lozovski, V.M., 1996. Zamostje 2: the Last Prehistoric Hunter-
Fishers of the Russian Plain. Treignes: Éditions du
Cedarc.
Machida, H. & F. Arai, 1983. Koiki tefura to kokogaku.
Daiyonki Kenkyu 22, 133–48.
McPherron, S.P., Z. Alemseged, C.W. Marean et al., 2010.
Evidence for stone-tool-assisted consumption of
animal tissues before 3.39 million years ago at Dikika,
Ethiopia. Nature 466, 857–60.
Mellars, P., K. Boyle, O. Bar-Yosef & C. Stringer (eds.), 2007.
Rethinking the Human Revolution. (McDonald Institute
Monographs.) Cambridge: McDonald Institute for
Archaeological Research.
Morwood, M.J. & P. Trezise, 1989. Edge-ground axes in
Pleistocene Australia: new evidence. Queensland
Archaeological Research 6, 77–90.
Neugebauer-Maresch, C., 1989. Zum neufund einer
weiblichen statuette bei den rettungsgrabungen an
der Aurignacien-Station Stratzing/Krems-Rehberg,
Niederösterreich. Germania 67, 551–9.
O’Connor, S., 1990. 30,000 Years in the Kimberley: a Pre-
history of the Islands of the Buccaneer Archipelago
and Adjacent Mainland, West Kimberley, Western
Australia. Unpublished PhD thesis, University of
Western Australia, Perth.
Oda, S. & C.T. Keally, 1973. Edge-ground stone tools from
the Japanese preceramic culture. Busshitsu Bunka 22,
1–26.
Oda, S. & C.T. Keally, 1992. The origin and early develop-
ment of axe-like and edge-ground stone tools in the
Japanese Palaeolithic. Indo-Pacific Prehistory Association
Bulletin 12, 23–31.
Olsson, I., 1986. Radiometric methods, in Handbook of Holo-
cene Palaeoecology and Palaeohydrology, ed. B. Berglund.
Chichester: John Wiley and Sons, 273–312.
Pailler, Y. & A. Sheridan, 2009. Everything you always
wanted to know about … la néolithisation de la
grande Bretagne et de l’Irlande. Bulletin de la Société
Préhistorique Française 106(1), 1–32.
Parkington, J., C. Poggenpoel, J.-P. Rigaud & P.-J. Texier,
2005. From tool to symbol: the behavioural context of
intentionally marked ostrich eggshell from Diepkloof,
Western Cape, in From Tools to Symbols: From Early
Hominids to Modern Humans, eds. F. d’Errico & L.
Blackwell. Johannesburg: Witwatersrand University
Press, 475–92.
Paton, R., 1994. Speaking through stones: a study from
 Jean-Michel Geneste et al.
northern Australia. World Archaeology 26, 172–84.
Pétrequin, P. & A.-M. Pétrequin, 1993. Ecologie d’un outil: la
hache de pierre en Irian Jaya (Indonésie). Paris: CNRS-
Piaget.
Philippson, D., 2005. African Archaeology. Cambridge: Cam-
bridge University Press.
Reimer, P., M. Baillie, E. Bard et al., 2009. Intcal09 and
Marine09 radiocarbon age calibration curves, 0–50,000
years cal bp. Radiocarbon 51, 1111–50.
Revedin, A., B. Arangurenb, R. Becattinia et al., 2010. Thirty
thousand-year-old evidence of plant food process-
ing. Proceedings of the National Academy of Sciences
107(44), 18,815–19. www.pnas.org/cgi/doi/10.1073/
pnas.1006993107
Roberts, R.G., R. Jones & M.A. Smith, 1990. Thermolumines-
cence dating of a 50,000-year-old human occupation
site in northern Australia. Nature 345, 153–6.
Roberts, R.G., R. Jones & M.A. Smith, 1993. Optical dating
at Deaf Adder Gorge, Northern Territory, indicates
human occupation between 53,000 and 60,000 years
ago. Australian Archaeology 37, 58–9.
Sackett, J., 1999. The Archaeology of Solvieux: an Upper Palaeo-
lithic Open Air Site in France. (Monumenta Archaeo-
logica.) Los Angeles (CA): Institute of Archaeology,
UCLA.
Sadr, K., F. Bon, F.-X. Fauvelle-Aymar, D. Gronenborn & B.
Bosc-Zanardo, 2006. Pour un nouveau comparatisme
Afrique australe-Europe occidentale: regards croisés
de l’archéologie, de l’histoire et de l’ethnologie des
chasseurs–cueilleurs Bushmen, in Normes techniques
et pratiques sociales: de la simplicité des outillages pré-
et protohistoriques. XXVe rencontres internationales
d’archéologie et d’histoire d’Antibes, eds. L. Astruc, F.
Bon, V. Léa, P.-Y. Milcent & S. Philibert. Antibes:
APDCA, 403–16.
Schrire, C., 1982. The Alligator Rivers: Prehistory and Ecology
in Western Arnhem Land. (Terra Australis 7.) Canberra:
Australian National University Press.
Semaw, S., P. Renne, J.W.K. Harris et al., 1997. 2.5-million-
year-old stone tools from Gona, Ethiopia. Nature 385,
333–6.
Semenov, S.A., 1964 [1957]. Prehistoric Technology. London:
Cory, Adams and Mackay.
Sharp, L., 1952. Steel axes for stone-age Australians. Human
Organisation 11, 17–22.
Smith, A.B., 2005. African Herders: Emergence of Pastoral Tradi-
tions. Walnut Creek (CA): Altamira Press.
Soressi, M. & F. d’Errico, 2007. Pigments, gravures, parures:
les comportements symboliques controversés des
Néandertaliens, in Les Néandertaliens: Biologie et
Cultures, eds. B. Vandermeersch & B. Maureille.
(Documents préhistoriques 23.) Paris: Éditions du
CTHS, 297–309.
Summerhayes, G.R., M. Leavesley, A. Fairbairn et al., 2010.
Human adaptation and plant use in highland New
Guinea 49,000 to 44,000 years ago. Science 330(6000),
78–81.
Svoboda, J. (ed.), 2008. Petrkovice: on Shouldered Points and
Female Figurines. Brno: Institute of Archaeology, Acad-
16
emy of Sciences of the Czech Republic.
Thomson, D.F., 1949. Economic Structure and the Ceremonial
Exchange Cycle in Arnhem Land. Melbourne: Macmillan.
de la Torre, I. & R. Mora, 2009. Remarks on the current theo-
retical and methodological approaches to the study of
technological strategies of early humans in eastern
Africa, in Interdisciplinary Approaches to the Oldowan,
eds. E. Hovers & D.R. Braun. (Vertebrate Paleobiology
and Paleoanthropology.) Dordrecht: Springer, 15–24.
de la Torre, I., R. Mora, M. Domínguez-Rodrigo, L. De Luque
& L. Alcalá, 2003. The Oldowan industry of Peninj and
its bearing on the reconstruction of the technological
skills of Lower Pleistocene hominids. Journal of Human
Evolution 44(2), 203–24.
Toth, N.P. & K.D. Schick, 2009. The Oldowan: the tool mak-
ing of early hominids and chimpanzees compared.
Annual Review of Anthropology 38, 289–305.
Valladas, H., J. Clottes, J.-M. Geneste et al., 2001. Evolution
of prehistoric cave art. Nature 413, 479.
Vanhaeren, M. & F. d’Errico, 2006. Aurignacian ethno-
linguistic geography of Europe revealed by personal
ornaments. Journal of Archaeological Science 20, 1–24.
Wadley, L., 2001. What is cultural modernity? A general view
and a South African perspective from Rose Cottage
Cave. Cambridge Archaeological Journal 11(2), 201–21.
White, C., 1967. Early stone axes in Arnhem Land. Antiquity
41, 149–52.
Zhao, C., X. Wu, T. Wang & X. Yuan, 2004. Early polished
stone tools in South China evidence of the transition
from Palaeolithic to Neolithic. Documenta Praehistorica
331, 131–7.
Zilhão, J. & F. d’Errico, 2003. The chronology of the Aurig-
nacian and transitional technocomplexes: Where do
we stand?, in The Chronology of the Aurignacian and of
the Transitional Technocomplexes: Dating, Stratigraphies,
Cultural Implications, eds. J. Zilhão & F. d’Errico. (Tra-
balhos de Arqueologia 33.) Lisbon: Instituto Português
de Arqueologia, 313–49.
Author biographies
Professor Jean-Michel Geneste manages France’s primary
national rock-art research institution, the Centre National
de Préhistoire, and directs the Chauvet Cave International
Research Program. He has been Curator of Lascaux for
more than 20 years and is a Field Director and Research Co-
ordinator on the ‘Connecting Country Jawoyn Homeland
Archaeological Project’ (connectingcountry.arts.monash.
edu.au).
Dr Bruno David specializes in the archaeology of the Sahul
region. He works with Indigenous communities on the
archaeology of rock art and symbolic expressions, tracking
the origins and transformations of ethnographically docu-
mented cosmological and social traditions back in time by
investigating temporal and formal discontinuities in their
material signatures. He is Project Co-ordinator and Field
Director on the ‘Connecting Country Jawoyn Homeland
Archaeological Project’.
 The Origins of Ground-edge Axes: New Findings from Nawarla Gabarnmang, Arnhem Land

Dr Hugues Plisson specializes in use-wear analysis of

prehistoric tools. His methodology, based on a synthesis of
the Western and Russian schools, has influenced genera-
tions of researchers aiming to understand the function of
archaeological objects.

Professor Jean-Jacques Delannoy is a geomorphologist and

Director of the EDYTEM research laboratory. He is Professor
at the Université de Savoie where he teaches karst systems.
His work crosses geomorphological and archaeological data
to better understand site-formation processes.

Dr Fiona Petchey specializes in the reliable radiocarbon

dating of a range of objects including charcoal, shell, pollen
and bone. She oversees the pretreatment of small samples
for AMS processing.

17

View publication stats