Australian rock art shows some of the oldest-known
artistic images by modern humans. However, there are considerable
technical difficulties and uncertainties in dating rock art
which make it difficult to determine the age of Australia's earliest
rock art.
Australian rock art, while extensive and in
places of great age, is nevertheless not the oldest in the world.
Both rock art and portable palaeoart were made long before Australia
was apparently first settled. The oldest currently known rock
art is in India, at such sites as Auditorium Cave and Daraki-Chattan,
but similar Acheulian rock art is believed to exist in the Kalahari
Desert of South Africa.
Arrival of humans in Australia
Archaeological evidence suggests that humans first arrived
in Australia between 65,000 and 60,000 years ago. Northern Australia
is the most likely place for people to have travelled from south east
Asia across the land bridges then sailed across the ocean gaps to
northern Australia. Archaeologists have now discovered early occupation
sites at the three most probable entry areas - the Kimberley, Arnhem
Land and Cape York Peninsula.
In northern Australia there are numerous sandstone rock-shelters.
Many of these have been used for camping and their floors are layered
with charcoal and ash from camp fires, the remains of food such as
shells and animal bones, stone tools and, very often, pieces of ochre.
Ochre comes from soft varieties of iron oxide minerals (such as haematite
- a fine-grained iron oxide which produces a strong red colour with
a purple tint) and from rocks containing ferric oxide.

Nauwalabila shelter in Kakadu National
park, Arnhem Land
Stone tools and ochre are the toughest of this camping
debris. Their appearance in the layers of material on the floor of
the shelter is usually interpreted as the beginning of occupation
at the shelter. Charcoal may or may not have survived in the lower
layers of a site, depending on local preservation conditions, and
other organic material tends to survive in only the youngest part
of the deposit, spanning at the most a few thousand years.
How do we estimate the age of Australian rock art?
The best way to establish the age of rock art is to
date the art directly (such as by dating a sample of the paint
or pigment used) or indirectly (for example to obtain a minimum
age for the art work by dating something that lies on top of
the art - say a mud wasp's nest or a natural chemical coating
- or lies in a layer of material with objects or matter that
can be dated). In the case of rock painting in Australia, dates
have been obtained for pigment directly on the walls and for
painted fragments buried in deposits of campsite material. For
an interesting discussion of issues to do with dating Arnhem
Land rock art, see the article
by Chippindale and Tacon.
Techniques for dating have usually involved radio-carbon
dating of material associated with the art, but there are also
newer techniques now available including optically stimulated
luminescence (OSL) and accelerator mass spectrometry (AMS).
These are described in the page on Dating
Rock Art by Robert Bednarik. Radiocarbon dating is limited
to a maximum age of around 40 000 years, and the newer techniques
are required for dating of older materials. AMS is a new radiocarbon
dating method enabling the dating of much smaller samples of
carbon than the traditional radiocarbon (C-14) method.
What are the Earliest Dates for Australian
Rock Art?
Ochre is the main pigment used in rock art and is plentiful
across most of Australia. Pieces of ochre, including some showing
signs of wear through use, have been found in almost all of Australia's
ice-age sites. Most have been radiocarbon dated and the dates range
from 10 000 to 40 000 years.
Does the use of ochre necessarily imply painting? As
well as rock art, ochre has many other uses in modern Aboriginal ceremony,
and is repeatedly found in association with burial not only in Australia
but also in other parts of the world. In Arnhem Land, there is no
certainty either that ochre was used for painting from the beginning;
or that painting with ochre was on rock surfaces (rather than on perishable
subjects); or that the first paintings on rock are amongst the ones
that survive. However, the hardness of much of the ochre found in
deposits strongly suggests that it was used on rock or other hard
surfaces and the pattern of wear is totally consistent with use of
the ochre in art.
The oldest dates so far found by direct dating of art
were obtained by geologist Alan Watchman for layers of pigment in
two rock-shelters on Cape York in north Queensland, one of 25 000
years and one of almost 30000 years.
There is, however, indirect evidence going back
a lot further, leading some archaeologists to argue that the
rock art galleries in northern Australia are some of the oldest
in the world by modern humans. This is, of course, a contentious
area, with recent claims for dates in southern France and northern
Italy going back as far as 35 000 years.
Archaeologist Sue O'Connor at the Australian National
University has found a buried fragment of rock painting preserved
in the limestone rock-shelter of Carpenter's Gap in the Kimberley
(near Windjana Gorge National Park) in a layer dated to 40 000
years old. The red pigment seems to be the remains of paint
on a rock art fragment fallen from the ceiling above.
The layer containing the painted fragment yielded
ochre, burnt bone, stone artefacts and charcoal with an accelerator
mass spectrometry (AMS) radiocarbon date of 39,700 ± 1,000 BP
(BP means Before the Present, which in this context is 1950,
when the radiocarbon dating technique was developed). Another
AMS date on charcoal from 20 centimetres below this gave statistically
the same date. These dates give a minimum age for the fragment
and for the occupation of the shelter. However convincing arguments
that this fragment is evidence of pigment application have yet
to be presented.
Probably Australia's earliest known artistic system
is the rock engravings in the Olary region
of South Australia. These geometric engravings of circles, tracks,
cupules and other designs are thickly coated with desert varnish.
The style of these older engravings is remarkably uniform across
Australia and other parts of the world. Initial claims that
these carvings were been radiocarbon dated to around 40 000
years ago have not been supported subsequently.
There has also been a recent report (June 2001) on the
dating of rock carvings in the East Pilbara of Western Australia.
The carvings, located in the Woodstock region of the Pilbara,
were dated using a new micro-erosion technique by rock art expert
Robert Bednarik. The technique dated some of the carvings at
more than 26 000 years old, and there is evidence to suggest
that much of the rock engravings in the area are of similar
age.
The Pilbara carvings are the same age as the cave
engravings at Malangine in South Australia, and have the same
distinctive style showing multiple circles with edges that come
close but never intersect. Mr Bednarik believes they are part
of an ancient continent-wide tradition of rock engraving. With
more than one million individual petroglyphs, the Pilbara has
what appears to be the largest concentration of Ice Age art
in the world.
Earlier Suggestions of Artistic Activity
The earliest evidence of human occupation yet found
in Australia is in two rock-shelters in Arnhem Land. In the lowest
layer of material at these sites there are used pieces of ochre -
evidence for paint used by artists 60,000 years ago! These shelters
lie at the foot of the western Arnhem Land escarpment in the Kakadu
region of the Northern Territory.
Malakunanja II is a shallow rock-shelter about 50 kilometres
inland from the present coast, while the rock-shelter of Nauwalabila
I (see picture above)
is 70 kilometres further south. Both have faded paintings on their
back walls, but these cannot be dated and are likely to be much younger
than the oldest stone tools in the occupational deposits.

Pieces of ground haematite and red and yellow
ochre were found, along with some stone tools, throughout the
deposits, including the lowest layers, in both rock-shelters.
The most common form of haematite found in these archaeological
sites is fairly small pieces with flat facets on their sides.
They are often triangular or elongated, resembling crayons or
pieces of chalk, and were possibly used as such to make drawings
on rock.
In contrast, some large lumps of haematite have
been found, far too big and heavy to have been used as crayons,
but frequently showing ground patches and fine scratched lines.
Such lumps were possibly ground or scraped with another artefact
to produce 'powder paint'.
The problem with using haematite as an indicator
of artistic activity is that it is not used uniquely for this
purpose: it was also used by many prehistoric societies for
body painting, for decoration of artefacts, for ceremonies,
as a preservative, for medical purposes and as an insect repellent.
Nevertheless, some archaeologists have pointed to its presence
at the two Arnhem Land sites and its worn shape as evidence
of artistic use.
The oldest find was a large piece of haematite weighing
one kilogram, in the lowest layer of Nauwalabila. It showed
clear signs of use and must have been brought from some considerable
distance away to be used as a source of red pigment.
The upper layers of the human occupation deposits
in both rock-shelters were dated by the radiocarbon method using
organic material such as charcoal from ancient cooking fires.
In the lower (older) layers accumulated before 20,000 years
ago, all organic debris had disappeared. In these layers the
quartz sands containing stone tools and other artefacts were
dated by the luminescence method (OSL). Archaeologist Rhys Jones
obtained dates for the earliest occupational layers of about
53,000 to 60,000 BP, more probably the latter.
There is thus evidence for artistic activity for
the entire span of human occupation of Australia, with the oldest
painted and engraved surfaces dated back perhaps as far as 40
000 BP, and signs of artists working with ochre paint possibly
as far back as 60 000 years ago.
Note: the above article is based on
information given in the book "Rock Art of the Dreamtime" by
Josephine Flood, Angus and Roberston 1997; also from "Prehistory
of Australia" by John Mulvaney and Johan Kamminga, Allen and
Unwin 1999; and updated by research results published in late
2000.
Thanks are due to Robert Bednarik for correcting
several errors in November 2004 - any remaining errors are of
course the responsibility of Aboriginal Art Online.
For more information, see the excellent Web site
of the Australian
Rock Art Research Association (AURANET).