Australian Archaeology (Number 60) June 2005

Transcription

CONTENTS
iii
Editorial
Articles
1
Kangaroo Island sealers and their descendants:
Ethnic and gender ambiguities in the archaeology
of a creolised community
Lynette Russell
6
‘BELIEF’ in the past: Dempster-Shafer theory,
GIS and archaeological predictive modelling
Shaun Canning
16
Water views: Water-based survey methods on
Cowan Creek, New South Wales
Susan Guthrie and James L. Kohen
69
24
Aboriginal occupation at Hawker Lagoon,
southern Flinders Ranges, South Australia
Keryn Walshe
2004 AAA Conference Dinner Awards
Jo Kamminga, Judith Field and Sean Ulm
71
Socializing stone artefact assemblages:
Regionalization and raw material availability
David R. Guilfoyle
Why are so few Australian archaeologists
Quaternary scientists?
Esmée Webb
Debitage
34
41
Sourcing stone from the Sydney region:
A hatchet job
Tessa Corkill
73
Thesis Abstracts
75
Corrections to AA 59, December 2004
Book Reviews
Short Reports
51
Purple haze: Evidence for a later date for
solarized amethyst glass
Samantha Bolton
76
‘The Incas’, by Terence N. D’Altroy
David Bulbeck
76
‘Colouring the past: The significance of colour
in archaeological research’, by Andrew Jones and
Gavin MacGregor
Noleen Cole
78
‘A Pacific odyssey: Archaeology and
anthropology in the western Pacific. Papers in
honour of Jim Specht’, edited by Val Attenbrow
and Richard Fullagar
Alison Crowther
52
Walter Burley Griffin and a Museum of
Archaeology at the heart of Australia’s capital
Sally Brockwell and Christopher Chippindale
54
Late Holocene occupation at Bunnengalla 1,
Musselbrook Creek, northwest Queensland
Michael Slack, Richard Fullagar, Andrew
Border, Jackson Diamond and Judith Field
58
A note on radiocarbon dates from the Paraburdoo,
Mount Brockman and Yandicoogina areas of the
Hamersley Plateau, Pilbara, Western Australia
Bruce Veitch, Fiona Hook
and Elizabeth Bradshaw
79
‘Archaeology and colonialism: Cultural contact
from 5000 BC to the present’, by Chris Gosden;
‘The archaeology of the colonixed’,
by Michael Given
Rodney Harrison
61
Indigenous archaeological sites and the Black
Swamp Fossil Bed, Rocky River Precinct,
Flinders Chase National Park, Kangaroo Island,
South Australia
Keryn Walshe
81
‘Hunter-gatherers in history, Archaeology and
anthropology’, edited by Alan Barnard
Ian McNiven
82
‘Constructing frames of reference: An analytical
method for archaeological theory building using
ethnographic and environmental data sets’,
by Lewis R. Binford
Donald Pate
Obituary
65
Bruce Veitch
Peter Veth and Sean Ulm
Backfill
67
The Australian Archaeology Electronic Archive
Michael Haslam
Australian Archaeology, Number 60, 2005
AGM Minutes
84
Minutes of the 2004 Annual General Meeting of
the Australian Archaeological Association Inc.
i
ii
Editorial
Special Volume
The next issue of Australian
Archaeology, AA 61 (December 2005),
will be a Special Volume addressing
Teaching and Learning in Australian
Archaeology. This volume will be edited
by Sarah Colley (University of Sydney),
Sean Ulm (University of Queensland),
and Donald Pate (Flinders University).
Many of the papers will be derived from
presentations made at the ‘Learning
Archaeology’ session held at the 2004
AAA Conference at the University of
New England.
Thanks to Pam Smith
and Keryn Walshe
A special thanks to Pam Smith who
acted as my Editorial Assistant in 2000 and
as Co-Editor from 2001-2004 and to
Keryn Walshe who served as Short
Reports Editor from 2000-2004. Your
long-term devotion to Australian
Archaeology is much appreciated. Pam
Smith will continue as Short Reports
Editor in 2005.
Conferences
For a detailed list of upcoming
archaeology conferences and workshops see
the Australian National University Centre
for Archaeological Research website at:
http://car.anu.edu.au/conferenceindex.html
Some upcoming conferences for 20052006 include:
World Archaeological Congress,
Inter Conference
The Repatriation of Ancestral Remains
8-10 July 2005
Canberra, ACT
This conference hosted by the Centre
for Cross-Cultural Research, Australian
National University and the National
Museum of Australia will examine
critically the successes and failures of
efforts to resolve the fate of Indigenous
ancestral remains acquired from
Australian and overseas museums and
scientific institutions. In addition it will
address the problems of identifying and
repatriating ancestral remains located in
European and other overseas collections,
especially in the light of scientific
reluctance and resistance to recognize the
rights and obligations of Indigenous
people in respect of the dead and their
possessions. Finally, the conference aims
to assess repatriation policies and practices
in the light of Indigenous community
experiences of repatriation.
Full details are available at:
http://ehlt.flinders.edu.au/wac/site/
confer_repat2005.php
Australasian Society for
Historical Archaeology
Annual Conference
29 Sept. – 2 Oct. 2005
La Trobe University
Melbourne, Victoria
The 2005 ASHA conference will be
held at La Trobe University’s city campus,
adjacent the vibrant Queen Victoria
Market precinct. The theme of the
conference is The City and the Bush. The
theme will be reflected in walking tours of
public archaeology interpretive displays on
some of Melbourne’s recent archaeological
excavations, a visit to the new Heritage
Victoria Conservation Laboratory, and a
post-conference field-trip to heritage sites
and wineries on the Mornington Peninsula.
http://www.asha.org.au/events.htm
Joint Conference of the
Australian Archaeological
Association and
Australasian Institute for
Maritime Archaeology
27-30 November 2005
Western Australian Maritime Museum
Fremantle, Western Australia
The theme of the conference is The
Archaeology of Trade and Exchange. This
theme encompasses a broad range of
archaeology of historical and prehistoric
contexts by both terrestrial and maritime
archaeologists. The conference intends to
encourage interdisciplinary links by
having several joint sessions on the theme
of trade and exchange.
The Conference will be hosted by
the Western Australian Maritime Museum
and Archaeology, the University of
Western Australia. The Western Australian
Maritime Museum is an exciting venue for
the conference. The facilities at the new
museum on Victoria Quay are modern and
the port city offers a wonderful locale to
explore ideas of exchange and trade.
Sessions will be held in the Maritime
Museum Lecture Theatre, overlooking the
mouth of the Swan River. Concurrent
sessions will use the neighbouring A-Shed
venue, Victoria Quay.
http://www.aaa-aima-2005.
conf.uwa.edu.au/
Australasian Archaeometry Conference
12-15 December 2005
Australian National University
Canberra, ACT
The Australasian Archaeometry
Conference 2005 will be hosted by the
Department of Archaeology and Natural
History, Research School of Pacific and
Asian Studies, and the Centre for
Archaeological Research at the Australian
National University. The venue will be the
Coombs Building at the ANU campus.
Sessions will cover all aspects of scientific
applications (biological, physical and
chemical sciences) in archaeology.
http://car.anu.edu.au/Archaeometry/
archaeometry_conference.html
Society for American Archaeology
71st Annual Meeting
26-30 April 2006
San Juan, Puerto Rico
For details consult the conference
website at:
http://www.saa.org/meetings/index.html
Back Issues of AA
Most of the older back issues of
Australian Archaeology have sold out due
to the success of the recent Clearance
Sales and special promotions. At present
available hard copy back issues include
Volumes 42-49 (1996-1999) and 54-59
(2002-2004). All back issues from 19742003 are now available on DVD for
A$50.00. See the advertisements in this
volume for details.
Acknowledgements
We take this opportunity to thank all
of Australian Archaeology’s contributors
and referees. We are also indebted to Ian
Murray at Graphic Print Group, Adelaide,
for assistance with the production of
Australian Archaeology.
Donald Pate
iii
iv
Book Reviews
Kangaroo Island sealers and their descendants: Ethnic and gender
ambiguities in the archaeology of a creolised community
Lynette Russell
Abstract
Our understandings of the European-Aboriginal contact
period are restricted by our limited engagement with and
interrogation of the categories used for analysis. Dividing
the past into Indigenous and non-Indigenous, Aboriginal
and invader and so on, fails to reveal the complexity and
nuances of cross-cultural, negotiated encounters and the
emergence of new social formations and identities.
Furthermore the ascription of ethnicity to historical actors
generally relies on late twentieth (early twenty-first) century
conceptions of what it means to be Aboriginal which are not
necessarily valid for the period under consideration.
Introduction
This paper was inspired by Walshe and Loy’s (2004)
request for feedback and debate on the analysis of a flaked
adze made from a late nineteenth century telegraph
insulator found on Kangaroo Island, South Australia. In the
spirit of that scholarship I offer the following discussion.
Historical studies, including historical/contact archaeology,
tend to be dominated by binaristic terminology including
male and female, men and women, native and newcomer,
Indigenous and non-Indigenous, invader and traditional
owner, victim and victor, slave and slaver, and so on. Such
dichotomies seamlessly lend themselves to the development
of interpretive paradigms such as invasion and resistance
which categorise the world into a black and white
comprehendible pseudo-reality, by simplifying and
homogenising complexity, variability and uncertainty (e.g
see Reynolds 1989; cf. Godwin 2001). However, such
dichotomies are based on epistemological categories that do
not necessarily resonate for the period being studied.
Furthermore there has been a lack of attention paid to the
two-way and often reciprocal nature of cultural interaction,
with the general perception being that European culture
(objects, materials and ideas) entered Aboriginal society,
while little influence was exerted on the culture of the
colonisers. Needless to say, I am referring here to
interactions that involved material or conceptual exchange
and the arguments can not be sustained for violent or hostile
encounters (e.g. massacres). The contact period of Australia
and in particular the complex cross-cultural history of the
sealing era (such as on Kangaroo Island) and its impact into
the latter part of the nineteenth century, presents an ideal
opportunity to explore some of these issues. One humble
adze might just reveal to us some of that complexity.
Contact and encounter
I recognise the oft-used term ‘contact’ is problematic not
least because is homogenises a range of complex and often
geographically and chronologically context-specific events
into a ‘concentrated moment of historical time’ (Torrence
Centre for Australian Indigenous Studies, Monash University, Clayton, VIC
3800, Australia. Email: [email protected].
and Clarke 2000:13; see also Colley and Bickford 1996:1617). In this paper, contact is used to refer to the very earliest
European-Indigenous interactions. Torrence and Clarke
eloquently suggest ‘negotiated outcomes’ (2000:16) as the
appropriate analytical framework. Negotiation is a
conceptual category that underpins cross-cultural and intercultural contact – negotiation explicitly implies a two-way
exchange.
Negotiated encounters leading to negotiated outcomes
provides the basis for creolisation theory, popular in
American and Caribbean archaeological studies (see
chapters in Cusick 1998; Thomas 1990; Deagan 1983).
Creolisation theory has, however, been much less
commonly explored within Australian archaeology. Its
importance should not be understated as it offers ‘an
alternative approach to understanding the cultural
interactions during colonization’ (Birmingham 2000:362).
Creolisation theory:
models circumstances in which cultural resistance
takes innovative and creative forms by adopting
much of what is relevant and useful from the
incoming, retaining many elements of what is
traditional, and then creating a new vigorous blend
or hybrid culture involving speech, technology,
music, art and religion, craft and institutions
(Birmingham 2000:362).
Within Australian contact history/archaeology,
creolisation theory has tended to be applied to mission and
other sites of interaction that post-date the contact period.
Although Birmingham’s ground breaking work has
attempted to use creolisation theory it has not so much
provided an opportunity to develop a new paradigm; but
rather a new name for the accommodation model (see
Russell 2001a, 2001b for a detailed discussion of the
resistance/accommodation paradigm). This has led to an
accommodation/creolisation model that provides an
explanatory framework for materials/ideas travelling from
the ‘newcomers’ to the ‘natives’. Although Birmingham’s
work is of immense value, I use the term and concept of
creolisation in a very different manner.
This paper is drawn from a larger project in which I am
exploring the dialectical processes of colonisation, which
considers the movement of ideas, materials, language and
even identities, back and forth between the categories of
coloniser and colonised, ultimately challenging the very
categories themselves (cf. Singleton 1998:178).
Importantly, I argue that careful definition of terms
emphasises that the nature of contact was neither one sided
nor exclusively controlled by the colonisers (see also
Godwin 2001; McNiven 2001; McBryde 2000).
Birmingham, working in an archaeological context, notes
contact in settler colonies is often characterised as a ‘mostly
unequal struggle between invaders empowered with
superior technology, ideology and language, and the
invaded who, armed only with an appropriate lifestyle and
1
Kangaroo Island sealers
familiarity with local conditions, reacted by resistance’
(2000:360). Recent archaeological research has attempted
to engage in a meaningful way with the previous monodimensional readings of contact and in particular
deconstruct resistance as it relates to settler-colonies (e.g.
Adams 1989; Cusick 1998; Deagan 1990; Colley 2000,
2002; Rogers and Wilson 1993; Rubertone 1989, 1996;
Torrence and Clarke 2000). In each of these examples
researchers have shown that cultural interaction and contact
is invariably a complex dialogic process, which neither one
side nor another entirely directs or commands.
There is no doubt that from an archaeological
perspective the impact of contact is often difficult to assess.
Aboriginal people grappled with the newcomers as they
attempted to incorporate them into their material and
spiritual worlds (McNiven and Russell 2002). Some have
argued that the application of “traditional” techniques to
new materials can be interpreted as a form of resistance, a
maintenance of traditional activities (see Birmingham
1992; Farnsworth 1992; Leone and Potter 1987; Singleton
1998). However, trade and exchange were important
mechanisms for establishing and maintaining alliances in a
process, Nicholas Thomas (1991) calls ‘entanglement’.
Each entangled object needs to be understood in terms of
the social relationships it mediates.
Kangaroo Island
Sealers from a broad range of background first settled
Kangaroo Island in the early 1800s (Cumpston 1970).
Categorising the men as European of course, simplifies and
homogenises, as some came from America, others from
Britain and elsewhere. Some of the sealers brought with
them Aboriginal (Pallawah) women from Tasmania and
shortly after arriving were joined by women from
Aboriginal tribes on the adjacent mainland (mainly
Ngarrindjerri). During the first decades of the nineteenth
century a society developed on Kangaroo Island that was an
amalgam of newcomer and Aboriginal. The women raised
children and worked on the sealskins preparing them for
sale. They also supplemented their families’ diets with
hunting and gathering. They took part in hunting the seals
using their traditional waddies, which the men also took to
using. Historical references to the islanders note that the
European men ‘smelled like foxes’(SA Govt 1836:8-9; see
also numerous references in Taylor 1996, 2002). It was
difficult differentiate from them from the “natives”; they
lived in bark huts reminiscent of Aboriginal humpies and
they wore skins for clothes as did their “wives” and children
(SA Govt 1836:8-9; Taylor 1996, 2002).
These men who had rejected ‘civilisation’ were
described as ‘worse than savages’ and one sealer, the wellknown Nat Thomas, was even attributed a creolised/hybrid
status, being described as:
A compound of sailor, sealer, farmer and wildman,
he possesses all the resources of a sailor, combined
with the instincts of the aboriginal native (The
Adelaide Observer, 15 January 1853).
Nat Thomas, the other sealers and the Pallawah and
Ngarrindjerri women who occupied Kangaroo Island defied
(and continue to defy) the essentialising (and dichotomous)
markers of race and even gender. Herbert Basedow,
recording information from the late nineteenth century,
noted that the women undertook what were perceived to be
2
traditional male activities when they formed ‘hunting
parties’ and combed the hinterlands of the Island in order to
kill their prey (1914:162). As Clarke (1996:62) observed:
‘in 1869’ government rations were supplemented with foods
obtained by three Pallawah women, using traditional
hunting and gathering techniques. Likewise these women
worked alongside the men in the sealing industry in jobs
that might well be perceived as masculine but only in a
European sense as it is well known that in traditional
Tasmanian Aboriginal society women hunted and killed
seals (Plomley 1987:996).
Kangaroo Island in the contact period enables an
exploration of the interconnected nature of the categories of
race and gender, indeed I argue that this interconnectedness
means that it is virtually impossible to tease apart Kangaroo
Island society into its constituent parts. Furthermore I
suggest that such attempts to break down the society into its
preformation elements of Aboriginal and European ignores
the reality that within a very short time of arriving on the
island this relatively new social entity would have
undergone substantial change. Like all societies, the sealing
community of Kangaroo Island was a complex amalgam of
constituent elements. As in any culture Kangaroo Island
was much more than the sum of its parts. Unlike the
dichotomous categories of resistance and accommodation,
the archaeology of this time and place resists the
essentialising categories of race and gender – Aboriginal or
European, male and female.
The nineteenth century archaeology of Kangaroo Island
consists of the remains of sealers’ camps, consisting of
stone and timber structures, glass bottles flaked into
scrapers, flaked stone tools, ceramics and some metal
objects for processing seals etc (see Campbell and Noone
1943; James 2002). If we were to use the traditional
paradigms for assessing the archaeology of Kangaroo
Island we might be moved to suggest that the stone tools
found around the sealers’ camps were made by Aboriginal
women resisting their servitude and maintaining their
cultural traditions. This is the subtext of many of the
archaeological interpretations. Indeed very little exploration
is given to alternate views as most archaeologists have
assumed (unproblematically) that the women were
responsible for manufacture (and probable use) of the stone
and glass scrapers which are found associated with these
sites (Campbell and Noone 1943; Draper 1999; Harvey
1941; Marsden 1991; De Leiuen 1998).
Kangaroo Island and other sealing communities provide
the opportunity to explore the process of creolisation in
which ‘indigenous and adopted cultural elements blend[ed]
into a new mixed culture, of extreme vigour, which differs
from its predecessors’ (Birmingham 2000:362). Using this
model, and arguing for ambiguity, I would like to consider
the possibility that the implements (and hence the
archaeology of the period) are not so easily explained, not
least because we do not know which members of these new
communities made them or who used them. The women
might well have made and used the stone artefacts. Early
European observers in Tasmania recorded Aboriginal
women working with stone (Roth 1899:151; Robinson 1834
cited in Plomley 1966:897) and it would appear that there
was no cultural prohibition for women using stone tools.
However, in the new context of Kangaroo Island sealing
community, it is equally possible that the women made but
did not use the artefacts or perhaps they even taught their
Russell
‘husbands’ the techniques for stone knapping. One could
also tend similar arguments for knapped glass tools. Did the
men make these items, or the women? Were the artefacts
used by or made by their children? In which case what
ethnic designation should we ascribe to the objects? In
addition, perhaps a much more important intervention is to
ask why we would want to ascribe an ethnicity at all? Rather
than label the objects Aboriginal (if they were made and
used by the women) or European (if they were made and
used by the men) it may well be more useful to consider
these artefacts as belonging to a new creolised category that
of, for the want of more appropriate terms hybrid or
creolised.
A creolised object from a creolised society
It is indeed exciting that Walshe and Loy report the
discovery of a telegraph insulator flaked to form an adze,
used for working wood, which had been hafted using an
unidentified resin (which was definitely neither Xanthorrea
nor Spinifex) (2004:38). The adze must post-date 1876 if it
was collected on Kangaroo Island (as this was the year
when the telegraph was installed), or 1871 if it was secured
from the mainland when the telegraph inched its way
through South Australia. Walshe and Loy (2004: 39)
designate the adze ‘Indigenous’:
the presumption around Indigenous versus nonIndigenous origin [they mean modification and use]
is drawn purely from the weight of evidence for
Indigenous use of introduced materials such as glass
and ceramics and to manufacture traditional tools
against the scant evidence for non-Indigenous use of
the same materials, post-contact.
Yet interpreting an absence of evidence (in the historical
record) as a mirror of the past is conceptually restrictive
given that a key aim of historical archaeology is to explore
the omissions and contradictions of the historical record. I
would argue that this object perhaps better than any other
epitomises hybridity or creolisation. However I would
question the label ‘Indigenous’. As I have observed above it
is virtually impossible to tease apart Kangaroo Island
society into its constituent racial, gender and class elements
based on twenty-first century’s notions of each category.
Furthermore I suggest that such attempts to break down the
society into its preformation constituents of Aboriginal and
European ignores the reality that within a very short time of
arriving on the island this relatively new social entity would
have undergone substantial change. Concurrent with this
transition I suggest that the identities of the social agents
would have been likewise in flux. As a corollary the
material culture (and indeed the people) of Kangaroo Island
at this time and place similarly resists the essentialising
categories of race – Aboriginal or European and for that
matter male and female. Indeed very little exploration is
given to alternate views as most archaeologists have
assumed (unproblematically) that the women were
responsible for manufacture (and probable use) of the stone
and glass scrapers which are found at these sites.
In arguing for a creolised society, I am not suggesting
that the Pallawah women gave up their identity as
Aboriginal women but those aspects of their identity in
particular that which relates to their daily lives can be
considered to be creolised. Early settler George W. Walker
noted, of Aboriginal women living with Bass Strait sealers,
that they were not heard to speak English and they
continued to practice their ‘ancient customs’ (cited in
Plomley 1987:282). Walker referred to the women dancing
naked and practising their traditional rituals. The Pallawah
women living on Kangaroo Island most likely maintained
similar traditional activities, possibly out of sight of their
European ‘husbands’, perhaps even doing so while
undertaking hunting excursions. Importantly, the theoretical
construction of a creolised society should not be seen as
replacing all aspects of the constituent cultures. To the
contrary, creolised societies such as that suggested for
Kangaroo Island were based around community members
expanding their cultural repertoire for the purposes of the
new community. Kangaroo Island sealing communities
included Europeanised Aboriginal women as much as they
included Aboriginalised European men.
Clearly much of what I propose to explore is speculative
and perhaps even risks being labelled unanswerable,
however, by not asking these complex questions we risk
replicating simplistic dichotomous narrativist histories. By
failing to explore the undecidable or ambiguous within
archaeology we are almost certainly condemned to
reproducing binaristic paradigms. Furthermore, this simple
binarism lacks historical specificity as the women’s lives
would have undergone significant change throughout the up
to 40 years they worked and lived with the men. Our twentyfirst century understandings of the meaning of
Aboriginality are based on identity politics and debate
which emerged in the mid-part of the twentieth century. For
many (but by no means all), Aboriginality is a category of
exclusivity, it is of its very nature binaristic and definitive
(see Anderson 1997; Langton 1981, 1988). Elsewhere I
have discussed at length that not all identity is exclusive,
stable or authoritative (Russell 2001a, 2001b, 2001c, 2002,
forthcoming 2005) and have explored the possibility of new
identity positions which reject the definitive nature of the
binarism of black or white (cf: Anderson 1997;
Papastergiades 1995). Given the acknowledgment that
identity positions are utterly context specific and absolutely
modern then we must note that there is simply no evidence
that the definitive identity paradigm (of black or white/
Aboriginal or European) can be extended back in time to
the early or mid nineteenth century.
Conclusion
Australia was, for most of its history, readily divided
into the categories of coloniser and colonised: European
and Aboriginal. As such, these racial/ethnic categories do
have some crude meaning as signifiers of cultural
difference. However at times these categories become
misleading. The boundaries around the native and the
newcomer were, in some places, vague, ambiguous, context
specific and most importantly, also unstable. Interpreting
the archaeological materials from sites that date to this
phase present a great challenge to the archaeologist.
However these materials also present us with a unique
opportunity to explore a set of cross-cultural interactions,
which produced a hybridised and composite mix of cultural
traits. Care needs to be taken not to privilege the mediation
of cultural difference – we also need to recognise that
Kangaroo Island in the post-contact period represents a new
society modified, blended with aspects from several other
groups. Perhaps by developing theoretical models which are
sufficiently fluid, so as to account for a range of
3
Kangaroo Island sealers
possibilities, we might approach a more comprehensive
understanding of the past.
Acknowledgments
Ian McNiven read this paper and made very useful
comments as did the anonymous referees. Rebe Taylor and
others who have written on Kangaroo Island continue to
inspire debate. To this I would add Kerryn Walshe and Tom
Loy for stimulating and requesting discourse. The
arguments in this paper represent a much condensed and
distilled version of a chapter in the book Beyond
identification? The Archaeology of Plural and Changing
Identities, edited by Eleanor Conlin Casella and Chris
Fowler, Plenum Press (2005).
References
Anderson, I. 1997 I, the hybrid Aborigine: Film and
representation, Australian Aboriginal Studies 1:4-13.
Basedow, H. 1914 Relic of the last Tasmanian race – Obituary
notice of Mary Seymour, Man 81 and 82:161-163.
Birmingham, J. M. 1992 Wybalenna: The Archaeology of Cultural
Accommodation in Nineteenth Century Tasmania. Sydney:
The Australian Society for Historical Archaeology
Incorporated.
Birmingham, J. 2000 Resistance, creolization or optimal foraging
at Killalpaninna Mission, South Australia. In R. Torrence and
A. Clarke (eds) The Archaeology of Difference: Negotiating
Cross-Cultural Entanglements in Oceania, pp. 360 405.
London: Routledge.
Campbell, T. D. and Noone, H. V. V. 1943 South Australian
microlithic stone implements. Records of the South
Australian Museum 7:281-307.
Clarke, P. 1996 ‘Early European interaction with Aboriginal
hunters and gatherers on Kangaroo Island, South Australia’.
Aboriginal History 20(1):51-81.
Colley, S. M. 2000 The colonial impact? Contact archaeology and
indigenous sites in southern New South Wales. In R.
Torrence and A. Clarke (eds) The Archaeology of Difference:
Negotiating Cross-Cultural Entanglements in Oceania, pp.
278-299. London: Routledge.
Colley, S.M. 2002 Uncovering Australia: Archaeology, Indigenous
People and the Public. Stdney: Allen and Unwin.
Colley, S. and Bickford, A. 1996 ‘Real’ Aborigines and ‘real’
archaeology: Aboriginal places and Australian historical
archaeology. World Archaeological Bulletin 7:5-21.
Cumpston, J. 1970 Kangaroo Island, 1800-1836. Canberra: Roebuck.
Cusick, J. G. (ed) 1998 Studies in Culture Contact: Interaction,
Culture Change and Archaeology. Carbondale: Southern
Illinois University. Center for Archaeological Investigations
Occasional Paper 25.
Deagan, K.A. 1983 Spanish St. Augustine: An Eighteenth-Century
Colonial Creole Community. New York: Academic Press.
Deagan, K.A. 1990 Accommodation and resistance: The process
and impact of colonisation in the southeast. In D. H. Thomas
(ed.) Columbian Consequences, Volume 2, pp. 297-314.
Washington: Smithsonian Institution Press.
De Leiuen, C. 1998 The Power of Gender. Unpublished B.A.
Honours thesis, Department of Archaeology, Flinders
University, Adelaide.
Draper, N. 1999 Land use history: The history of Aboriginal land use
on Kangaroo Island. In A.C. Robinson and D.M. Armstrong
(eds) A Biological Survey of Kangaroo Island, pp. 38-48.
Adelaide: Department of Environment and Aboriginal Affairs.
Farnsworth, P. 1992 Mission Indians and cultural continuity.
Historical Archaeology 26(1):22-36.
Godwin, L. 2001 The fluid frontier: Central Queensland 18441875. In L. Russell (ed.) Colonial Frontiers: IndigenousEuropean Encounters in Settler Colonies, pp. 101-118.
Manchester: Manchester University Press.
4
Harvey, A. 1941 Flint implements of Tasmanian manufacture
found at Cape Hart, Kangaroo Island. Records of the South
Australian Museum 6:363-368.
Hemming, S., Wood, V. and Hunter, R. 2000 Researching the past:
Oral history and archaeology at Swan Reach. In R. Torrence
and A. Clarke (eds) The Archaeology of Difference:
Negotiating Cross-Cultural Entanglements in Oceania, pp.
331-359. London: Routledge.
James, K. 2002 Wife or Slave? The Kidnapped Aboriginal Women
Workers and Australian Sealing Slavery on Kangaroo Island
and Bass Strait Islands: Using Landscape Archaeology to
Search for Evidence of Pre-settlement Lifeways during the
Sealing Era 1802-1835 on Kangaroo Island, South Australia.
Unpublished Honours thesis. Department of Archaeology,
Flinders University.
Langton, M. 1981 Urbanising Aborigines: The social scientists’
great deception. Social Alternatives 2(2):16-22.
Langton, M. 1988 Medicine Square. In I. Keen (ed.) Being Black:
Aboriginal Cultures in ‘Settled’ Australia (Reprint 1991).
Canberra: Aboriginal Studies Press.
Leone, M. and Potter, P. 1987 The Recovery of Meaning.
Washington: Smithsonian Institution Press.
McBryde, I. 2000 ‘Barter … immediately commenced to the
satisfaction of both parties’: cross-cultural exchange at Port
Jackson 1788-1828. In R. Torrence and A. Clarke (eds) The
Archaeology of Difference: Negotiating Cross-Cultural
Entanglements in Oceania, pp. 238-277. London: Routledge.
McNiven, I. J. 2001 Torres Strait Islanders and the maritime
frontier in early colonial Australia. In L. Russell (ed.)
Colonial Frontiers: Indigenous-European Encounters in
Settler Colonies, pp. 175-197. Manchester: Manchester
University Press.
McNiven, I. J. and Russell, L. 2002 Ritual response: Place
marking and the colonial frontier in Australia. In B. David
and M. Wilson (eds) Inscribed Landscapes: Marking and
Making Place, pp. 27-41. Honolulu: University of Hawaii
Press.
Marsden, S. 1991 Heritage of Kangaroo Island. Adelaide:
Department of Environment and Planning, South Australia.
Papastergiades, N. 1995 Art and Cultural Difference: Hybrids and
Cultures. London: Academy Group.
Plomley, N. J. B. 1966 Friendly Mission: The Tasmanian Journals
and Papers of George Augustus Robinson. Hobart: Blubber
Head Press.
Plomley, N. J. B. 1987 Weep in silence: A History of the Flinders
Island Aboriginal Settlement; with the Flinders Island
Journal of George Augustus Robinson, 1835-1839. Sandy
Bay: Blubber Head Press.
Reynolds, H. 1989 Dispossession: Black Australians and White
Invaders. St Leonards: Allen and Unwin.
Rogers, J. D. and Wilson, S. M. 1993 Ethnohistory and
Archaeology: Approaches to Post-Contact Change in the
Americas. New York: Plenum Press.
Roth, H. L. 1899 The Aborigines of Tasmania. Halifax: King and
Sons.
Rubertone, P. E. 1989 The archaeology, colonialism and 17th
century Native America: Towards an alternative
interpretation. R. Layton (ed.) Conflict in the Archaeology of
Living Traditions, pp. 32-45. London: Unwin Hyman.
Rubertone, P. E. 1996 Matters of inclusion: Historical archaeology
and Native Americans. World Archaeological Bulletin 7:7786.
Russell, L. 2001a Savage Imaginings: Historical and
Contemporary Constructions of Australian Aboriginalities.
Melbourne: Australian Scholarly Publishers.
Russell, L. 2001b Introduction. In L. Russell (ed.) Colonial
Frontiers: Cross Cultural Interactions in Settler Colonies,
pp. 1-10. Manchester: Manchester University Press.
Russell, L. 2001c The instrument brings on voices: Life story
narratives and family history. Meanjin 3:145-152.
Russell
Russell, L. 2002 A Little Bird Told Me: Family Secrets, Necessary
Lies. St Leonards: Allen and Unwin.
Russell, L. (ed) forthcoming Boundary Writing: Living across the
Boundaries of Race, Sex and Gender. Honolulu: University
of Hawaii Press.
Singleton, T. 1998 Cultural interaction and African American
identity in plantation archaeology. In J. Cusick (ed.) Studies
in Culture Contact: Interaction, Culture Change and
Archaeolog, pp. 172-188. Carbondale: Southern Illinois
University. Center for Archaeological Investigations
Occasional Paper 25.
South Australian Government 1836 First Report of
Commissioners on the Colonisation of South Australia.
London: Public Records Office.
Taylor, R. 1996 Sticking to land- A history exclusion on Kangaroo
Island 1827-1996. Unpublished Masters thesis, University
of Melbourne.
Taylor, R. 2002 Unearthed: The Aboriginal Tasmanians of
Kangaroo Island. Adelaide: Wakefield Press.
Thomas, N. 1991 Entangled Objects: Exchange, Material
Culture, and Colonialism in the Pacific. Cambridge:
Harvard University Press.
Torrence, R. and Clarke, A. 2000 Negotiating difference: Practice
makes theory for contemporary archaeology in Oceania. In
R. Torrence and A. Clarke (eds) The Archaeology of
Difference: Negotiating Cross-Cultural Entanglements in
Oceania, pp. 1-31. London: Routledge.
HISTORIC SITES AND LANDSCAPES
THE HILLS FACE ZONE
CULTURAL HERITAGE
PROJECT REPORTS
Edited by
Pamela A. Smith, Susan Piddock and Donald Pate
Kopi Books, Adelaide
Order form at www.milnwalker.com.au
2004 (Vol. 1)
2005 (Vols 2-5)
ISBN 0-9757-3590-x
RRP A$19.95 each
(paperback)
5
‘BELIEF’ in the past: Dempster-Shafer theory, GIS and archaeological
predictive modelling
Shaun Canning
Abstract
This paper introduces the use of a new technique in
archaeological predictive modelling which has particularly
wide ranging appeal and application in the cultural heritage
management sphere. Most predictive modelling programs
are restricted to the use of certain types of data with
a range of untested assumptions or caveats. Traditional
probability based models are difficult to construct, and are
often limited by the use of data sets which render the results
statistically suspect. The method introduced here makes use
of non-parametric statistical methods which are
not hampered by imperfect raw data. In particular, this
paper introduces the use of Dempster-Shafer theory in
archaeological predictive modelling
Introduction
The majority of contemporary archaeological data in
Australia are produced through cultural heritage
management survey projects. These survey projects are
conducted by contract archaeologists on behalf of
exploration companies, developers or land managers.
Academia has often criticised non-academic archaeologists
for a lack of rigour, and the production of vast quantities of
essentially useless archaeological survey data. Some of the
perceived shortcomings of this corpus of data include poor
sampling methods, insufficient survey coverage, nonstandardised survey methods, visibility constraints and
numerous other bias producing flaws or errors. Given the
criticism aimed at the consulting archaeologists and the
identified biases in the data, can the data produced by the
consulting industry be put to good use? Or, as is currently
the case, is this ‘grey literature’ simply dismissed as too
problematic to be useful? This paper explores methods of
utilising these data in predictive models, and discusses ways
of allowing for the known biases, and incorporating this
‘error’ into the resultant models.
The study area and data
The project (Canning 2003) this paper describes was
conducted in the Melbourne Metropolitan area between
1999 and 2003. One of projects main aims was to construct
predictive models of Aboriginal archaeological site location
within the relevant study area using the existing Aboriginal
Affairs Victoria (AAV) sites database. The majority of the
archaeological data contained in this database are the result
of contract projects, and as such, contains many of the
identified biases. Over 1000 sites and approximately 100
archaeological reports held by AAV were the source data for
the predictive model developed later in this paper. The
purpose of the predictive model is to provide local
government planners with management tools that will
enhance the identification, protection and management of
P.O. Box 21, Dampier, WA 6713, Australia.
Email: [email protected].
6
Aboriginal archaeological sites within the relevant local
government areas.
Model building caveats
Models are essentially abstract representations of an
observed or hypothesized phenomenon (Winterhalder
2000). The models developed here focus on very specific
elements of Aboriginal behaviour in prehistory. They make
use of ethnographic analogy, and utilise the theoretical
perspectives of human ecology. It is specifically noted,
however that the archaeological record cannot simply be
read in terms of ethnoarchaeological understandings of past
Aboriginal behaviour. Such an approach would ignore
depositional and post-depositional processes, discard rates,
erosion, and innumerable other factors which have created
the archaeological record. All models ultimately provide
nothing more than a generalised view of the archaeological
record perceived by us through innumerable filters and
obstructions. The methods presented later in this paper aim
to extract meaning from data which may otherwise be of
limited use in other forms of predictive modelling
endeavours, and uses new techniques to achieve this end.
These are therefore ‘self-consciously reductionist’ models’
(Winterhalder 2001: 14).
Predictive modelling
Predictive modelling in archaeology has its origins in
the settlement pattern analyses of Julian Steward (1938) and
Gordon Willey (1953). These pioneering studies focused
primarily upon the relationship between regional
environments and settlement patterns (Dalla Bona 1994;
Kohler and Parker 1986). Out of the development of
settlement pattern studies, and the increasing emphasis on
scientific research methods, catchment analysis techniques
were developed to investigate regional processes (Higgs and
Vita-Finzi 1972; Vita-Finzi and Higgs 1970) that emphasise
the relationships between people and their environment
(Roper 1979). The ‘New Archaeology’ of the 1960s and the
heightened interest in archaeological sampling techniques
and data analysis methods (Binford 1964) led to a shift by
some archaeologists away from ‘single-site’ archaeology to
broader regional questions. The introduction of new
technological tools (i.e. computers) has given practitioners
the ability to interrogate greater quantities of data than had
previously been possible (Dalla Bona 1994).
Amongst the first studies to explicitly state that a
research goal was to predict actual archaeological site
location was that of the South-western Anthropological
Research Group (SARG) in the United States (Plog and Hill
1971). The SARG members reasoned that if the structure of
a particular settlement system under consideration was
known, then it should be possible, a priori, to predict the
location of unknown archaeological sites (Plog and Hill
1971). Similar prediction-based research questions began to
appear in the archaeological literature during the early
1970s (Green 1973; Thomas 1973, 1975). Although it has
Shaun Canning
been claimed that the use of predictive modelling in pure
research based activities has gradually declined (Dalla Bona
1994), the perceived benefits of predictive modelling in
cultural heritage management (CHM) applications has
fuelled the continued development of predictive modelling
method and theory (Kohler 1988; Kohler and Parker 1986;
McManamon 1984; Warren 1990).
Cultural heritage management agencies are generally
mandated with the responsibility of managing and
protecting a finite archaeological resource base for a
relevant region, state or nation. This is attempted using
various combinations of techniques, including excavation,
site discovery and recording programmes, and more
recently the application of predictive modelling. The
responsibility to manage and protect the archaeological
record means many agencies are constantly seeking
improved methods to locate and record archaeological sites,
particularly in the face of rapid development and urban
expansion (MacNeill 1998). In this environment, the
benefits of archaeological predictive modelling to cultural
heritage managers appear obvious, even though no one is
entirely certain as to how predictive modelling should be
approached (Kohler 1988), and the approach eventually
chosen is dependant on a great many variables. Paramount
among these is determining the purpose for the model. Is
the model to be a purely academic exercise, or is the model
designed to ‘red flag’ (Altschul 1990) areas of
archaeological sensitivity for planning and management
agencies?
Types of predictive models
There are many types of predictive models in CHM and
archaeology (Kohler 1988), which will be briefly
summarised, together with their various theoretical
perspectives, and the range of decisions involved (Gibbon
1998). Two major approaches are used in the construction of
archaeological predictive models (inductive and deductive),
particularly those developed within a CHM framework
where their primary purpose is not explanation, but usually
prediction. Explanation is perhaps the more desirable
outcome of research activity, as theoretically grounded
explanation is a more powerful tool than prediction alone.
Deductive Modelling involves deductive logic where the
researcher moves from the abstract (theory) to the nonabstract (archaeological reality). Deductive predictive
models commence from a certain theoretical perspective
and proceed towards an understanding of extant
archaeological data or phenomena primarily via
explanation. Indeed, Kohler (1988:37) defines a deductive
modelling approach as one that begins ‘with a theory as to
how people use a landscape and to deduce from that theory
where archaeological materials should be located’. For
example, a model of archaeological site location may be
constructed using the theoretical perspectives of
behavioural or human ecology (Butzer 1982). Once the
model has been constructed, and middle-range theoretical
issues such as discard rates, depositional and postdepositional processes have been incorporated, the
modelling process can then turn to data gathering and the
interpretation (explanation) of that data (Ebert 1988).
Kohler and Parker (1986:432) state that a deductive model
must:
(1) ‘Consider how humans make choices concerning
location. This requires considering a mechanism for
decision-making, and an end for decision making;
what is the goal?
(2) Specify the variables affecting location decisions for
each significant chronological or functional subset
of sites; and
(3) Be capable of operationalisation’.
In the realm of CHM predictive modelling, deductively
based models are comparatively rare.
Inductive Modelling involves the researcher moving from
detailed data to more generalised theory. In inductive
predictive modelling, the researcher begins with site data
and then makes estimates or inferences regarding the
overall spatial distribution of archaeological material in that
sampling universe (Kohler 1988; Neuman 1997). This type
of modelling exercise makes use of existing data, such as
site records held by management agencies, and is the most
common approach to predictive modelling (Ebert 2000;
Kohler 1988) particularly in CHM applications. This
approach is also known as correlative or inferential
modelling (Kohler 1988).
It has been argued that inductive predictive models are
simply ‘formal devices of pattern recognition’ (Warren and
Asch 2000:8). Inductive predictive modelling in CHM and
archaeology is primarily concerned with attempting to find
correlations between site location and the numerous
environmental attributes considered relevant. These
attributes are normally ‘harvested’ from existing
contemporary geospatial data. Once (or if) any correlations
are established, the process then moves on to ‘modelling’
the probable locations of further unknown sites (Ebert
2000). There are severe theoretical failings in the inductive
modelling approach. Arguably, the most problematic issue
with inductive or ‘correlative’ (Church et al. 2000)
modelling is the almost implicit assumption that settlement
decisions made by people in the remote past are somehow
directly linked to geospatial attributes that can be derived or
deduced from modern maps (Ebert 2000). While certain
environmental attributes undoubtedly influenced prehistoric
human settlement choices, not all of these attributes are
static through time. Change in the environmental structure
of a place through time is seldom taken into account in
inductive models.
As an approach to predictive modelling, the inductively
based model also has major appeal. The principal reason for
the popularity of this approach is that the majority of the
data required already exists in the form of site databases and
geospatial map data. This significantly reduces the costs of
any modelling project, which is another area of
considerable appeal to the agencies that fund predictive
modelling exercises (Church et al. 2000).
Mathematical vs. graphical modelling
In relation to the use of models, Rindos (1989:13)
comments ‘…their potential weakness lies in their tendency
to make us believe we have an insight into the data when we
merely have created a mathematical epiphenomena’.
Dalla Bona (1994) discusses the second significant
decision that is required when building a predictive modelthe choice between a mathematical or graphical modelling
methodology. Although the two may be combined, it has
been the usual practice for modellers to choose one over the
other.
Mathematical predictive models make use of any of the
7
‘BELIEF’ in the past: Dempster-Shafer theory
numerous multivariate statistical methods available in order
to determine if correlations between archaeological site
locations and the variables under analysis exist (Dalla Bona
1994). Graphical techniques often make use of modern
computer hardware and software (particularly GIS) to
develop a model as a series of map overlays of the relevant
variables under consideration.
A third category of predictive model often seen in CHM
literature and reports are known as intuitive models.
Intuitive models are based upon a practitioners experience
in the field, and their ‘feel’ for the archaeology of a
particular area. Intuitive models are seldom tested (or are
indeed testable) in an empirical sense. The ‘predictions’ will
mostly be a series of statements such as ‘sites will occur on
terraces above waterways’. These are intuitive statements,
and are not a ‘predictive model’ in the true sense of the term
(Kohler 1988). This type of ‘model’ is common in
Australian CHM, and many consulting reports (i.e. du Cros
1989, 1990, 1991) contain ‘predictions’ such as the
preceding example. These types of intuitive statements are
based upon the archaeologist’s notion of where sites will be
located (intuition), rather than any formalised research or
sampling design (Altschul 1988; Moon 1993). These type
of data are often referred to as ‘expert knowledge’.
been developed in recent years to model the process of
frontier settlement in the eastern United States (Zubrow
1990), to model the development of trade in the Great Lakes
area of the United States (Allen 1990) and to address
Palaeolithic and Mesolithic archaeological site location in
the Southern Netherlands (Kamermans and Rensink 1999).
The major limitations on predictive modelling are the
availability of the appropriate data (in the case of inductive
models) or the appropriate theoretical perspectives (in the
case of deductive models) upon which to base the model(s).
The scope of a model is also dependant upon the required
outcomes of the modelling process. If the model is to
predict archaeological site location on a contemporary
landscape for management agencies to aid in the
preservation of the archaeological record, then the model is
constructed primarily for amenity, and not to answer
specific research questions regarding human behaviour and
archaeological processes in the past. However, overarching
research paradigms are an absolute necessity as the ultimate
goal of any CHM activity is the preservation of
representative samples of complexly constructed
archaeological landscapes. Research facilitates a better
understanding of the nature of these complex
archaeological landscapes.
Attributes of predictive models
While predictive models may make use of a variety of
methods and techniques, they can generally be divided
between deductively or inductively based, and are either
mathematical or graphical in design (this delineation is not
absolute). All models however, regardless of method, should
share a number of common attributes. Models should be
testable, be simple enough to be useful, and must be able to
be operationalised (Kohler 1988; Kohler and Parker 1986;
Kvamme 1988a, 1988b; Moon 1993). Finally, because
models are simple representations of reality, they are always
fallible (Kohler 1988; Moon 1993). The choice between
different approaches and styles of modelling depends as
much upon the required outcomes, and the available data, as
it does on the methods utilised. Either way, ‘there is nothing
inherently unscientific about either approach’ (Warren
1990:90) and each method has its champions and its critics
(Ebert 2000; Westcott and Kuiper 2000).
Predictive models, regardless of type, can be
constructed to almost any scale. Contemporary predictive
models of prehistoric archaeological site location have been
constructed utilising a study area as small as a few hectares,
up to very large undertakings such as the Minnesota
Department of Transport’s ‘Mn/Model’ (Brooks et al. 2000)
which models prehistoric archaeological site location for
the entire state of Minnesota (22,000,000 ha). Projects of
this scale and budget are rare. The Minnesota Department
of Transport (MN/Dot) spent $US4.5 Million on the project
between 1995 and 2000, employing 49 people in various
capacities. Although a rare level of commitment to one
project, the Mn/Dot model has been shown to have saved
the Minnesota Department of Transport $US 12 Million
since 1998 (Minnesota Department of Transport 2000). The
majority of predictive modelling projects however, are
nowhere near this scale or scope.
The scope of predictive modelling projects is also broad.
Models may be constructed to predict the location of
archaeological sites from any temporal period or
archaeological class. For example, predictive models have
Modelling sensitivity
It is generally considered impossible to construct a
predictive model with the necessary spatial resolution (van
Leusen 2002: 5-16) to predict the location of all individual
hunter-gatherer-fisher archaeological sites within a region,
particularly as the geographic scale of the model increases
or the resolution of the spatial data decreases. Nor is the
traditional reliance upon individual site based assessment
well suited to the development of broad scale models of
archaeological sensitivity or significance. This situation has
led to the development of zone-based assessments of
archaeological sensitivity and significance at landscape or
regional scales (Altschul 1990; McConnell 1995, 2002; van
Leusen 2002). Rather than attempting to predict the
location or significance of individual sites, (or indeed the
presence or absence of individual sites) zone-based
assessments highlight those zones within a region that are
expected to contain archaeological materials of various
classes, and, a priori, significance. It must be remembered
however, that it is the interpretation of archaeological
material that determines significance, and not geomorphic
or ecological predictions.
For instance, the deeply stratified alluvial deposits of the
Maribyrnong Valley in the Australian state of Victoria
(location of the Keilor and Green Gully sites) could
reasonably be expected to contain buried prehistoric
archaeological material. These comparatively rare
occurrences are significant for their ability to illuminate
past human behaviour in detail. However, it is impossible to
predict the exact location of these rare phenomena with any
chance of success. Knowing the geomorphic context in
which these sites are likely to occur means the entire
geomorphic unit can be regarded as archaeologically
sensitive, and therefore likely to contain significant
archaeological materials. This zone-based approach
provides a superior method of identifying part or whole
landscapes where archaeologically significant sites may be
located rather than continuing to rely solely upon sporadic
site surveys or test excavations.
8
Shaun Canning
Altschul (1990) utilised a zone-based methodology
when developing the ‘red flag models’ of Mount Turnball in
Arizona. Altschul (1990) viewed this approach as a more
powerful tool to be used in everyday cultural heritage
management contexts than individual site based
assessments. Altschul’s (1990) methodology consisted of
simply modelling three environmental variables (elevation,
slope and aspect) believed to influence archaeological site
location, and then plotting the relationship between these
variables and the existing state archaeological database.
The result of Altschul’s (1990) project was a series of
‘favourability’ maps which corresponded well to where
archaeological sites were expected to occur, and did in fact
occur. In terms of end usage, management agencies were
handed a tool (favourability or sensitivity maps) that
allowed them to ‘flag’ areas of greater sensitivity well in
advance of any development related activities.
Anne McConnell (1995; 2002) developed similar zonebased models of archaeological sensitivity for management
agencies in Tasmania and Victoria. McConnell (2002)
assessed a series of environmental variables thought to
have some bearing upon prehistoric Aboriginal
archaeological site location. These attributes included
distance to fresh water, slope, and access to flakeable stone.
McConnell (2002), with the assistance of the Forest
Modelling Branch of the Department of Natural Resources
and Environment (DNRE), created a series of sensitivity
maps using GIS that are to guide DNRE in the planning of
logging and general management operations in Victorian
forests. Josephine McDonald also utilised a similar method
of sensitivity zoning in her study of a site on the
Cumberland Plain near Sydney. McDonald (1996) based
the sensitivity zones in her study primarily upon the level
of previous ground disturbance that was observed
throughout the study area. In this way, McDonald proposed
that it was possible to identify entire landscapes that had
undergone little disturbance since European settlement, and
were thus more likely to contain undisturbed Aboriginal
sites or ‘potential archaeological deposits’.
The major attraction of zone-based methods is that
valid wide-area sensitivity models can be formulated (at a
‘macro scale’) where much of the data critical to
statistically-based models is absent or cannot be
determined (i.e. site absence vs. presence, statistically valid
samples). For instance, the attempt by Lewis, MacNeill,
and Rhoads (1996) to create a predictive model of
archaeological site location in East Gippsland was not
completely successful because of limitations in the
available data sets and statistical complexity (McConnell
2002:29). Considering the frustrating array of limitations
(i.e. visibility constraints) that can confront archaeological
projects the sensitivity zone approach is arguably the most
appropriate means of modelling the location of certain
prehistoric archaeological materials. Rather than
attempting to predict the locations of individual sites (as
many models do), it is argued that a method of determining
archaeological sensitivity based upon the relationship
between existing known site data and key environmental
attributes is a productive means of firstly identifying, and
secondly preserving, archaeological material. In areas
where it is possible to isolate particular geomorphic
features that are known to be archaeologically sensitive, the
zone-based approach is particularly useful. For example,
the alluvial deposits of the Maribyrnong River valley are
known to contain Pleistocene archaeological deposits of
great scientific and cultural significance. Rather than
attempting the impossible task of predicting where
‘individual’ buried sites lie, it is far simpler to zone this
entire geomorphic context as ‘sensitive’ and impose limits
to the type and extent of land altering development
activities permitted within (or impacting upon) this zone.
Within the various ‘zoned’ areas, it will still be necessary to
conduct archaeological survey and research in order to
update and improve the data for the model, as well as to
ensure that any archaeological material located is recorded,
and afforded the full protection of the relevant state CHM
policies and legislation.
Establishing a predictive modelling program is not a
particularly easy task. The researcher is faced with a myriad
of choices, which have little bearing upon problems of an
archaeological nature. The process is further complicated
by the necessity of determining whether the model
developed is to address management questions or provide
insight into, or explanations of, human behaviour in the
past. While both are equally valid pursuits, they are not
always compatible aims. A management model that is
designed to predict the most likely locations of prehistoric
archaeological sites on a contemporary land surface is not
seeking to explain why the archaeological material is
located where it is. This type of model aims to identify
where archaeological material is likely to be so as to avoid
inadvertently destroying this material through development.
Case study
The model of archaeological sensitivity presented here
was developed as part of doctoral research in the
archaeology programme at La Trobe University (Canning
2003). This model is based upon the extensive collection of
available archaeological data held by Aboriginal Affairs
Victoria, and publicly available geospatial data sets. One
map sheet has been chosen upon which to construct the
model of archaeological sensitivity as it meets many of the
criteria considered important for the modelling exercise.
The VicMap 7822-1-3 1:25,000 Mapsheet was chosen for
the modelling example because:
(1) It contains the important Keilor and Green Gully
sites,
(2) It contains 276 registered Aboriginal sites,
(3) Approximately 10.5% of the total map sheet has
been subject to archaeological survey, and
(4) It has been the subject of considerable previous
archaeological research.
Summary of available data
The following environmental attributes were extracted
by interrogating various spatial data sets using ArcView 3.2
to determine if relationships existed between archaeological
site location and the variable in question. For many of the
environmental variables analysed (i.e. soil drainage) it was
simply impossible to determine any correlation between the
archaeology and the variable (predominantly because of
poor spatial resolution in the environmental data). The
following section summarises the major environmental
variables often believed to have considerable bearing upon
archaeological site location.
Geology
The overall effect of geology on the distribution of
9
archaeological sites in the area in question is difficult to
determine. However, flakeable siliceous stone (i.e. silcrete)
sources commonly occur at the junction of the basalt plains
and the river valleys (Webb 1995) common in the area, and
quartzite river cobbles are prolific in the various waterways.
Topography
A greater concentration of archaeological materials has
been recorded at lower altitudes in the study area, and this
is problematic in the construction of any sensitivity models.
It is not clear if the lack of sites at higher altitudes
accurately reflects prehistoric Aboriginal behaviour
patterns, or is simply a product of bias in survey or the sites
database. The effect of elevation on prehistoric Aboriginal
site distribution is poorly understood, thus elevation is not a
particularly strong predictive variable, particularly in areas
that display relatively little topographic variation through
large tracts of the subject area (as is the case here).
Distance to water
Distance to fresh water is the most often used
environmental variable (van Leusen, 2002) in Australian
hunter-gatherer archaeological modelling. Distance to water
is used here in much the same manner as in any other
project. The importance of access to potable water is
considered one of the primary environmental factors
affecting prehistoric land use decisions. Nearly two thirds of
the recorded AAV sites (n = 276) for the area in question are
within 0-100 metres of a fresh water source, and
approximately 80% are within 200 metres (see Fig. 1).
Slope
Slope is a direct function of the topography of a region.
In the present study area, slope is a variable with little real
‘predictive power’. While almost 90% of all sites within the
study area are located on landforms where the slope is
between 0º and 10º, over 90% of the study area exhibits a
slope of between 0º and 10º. The effect of survey bias on the
distribution of sites per slope ‘class’ is also uncertain. While
it would seem likely that Aboriginal occupation areas would
more frequently be located on landforms displaying limited
slope, it is not possible to quantify the relationship further.
The areas displaying the greatest slope throughout the study
area feature the least number of recorded sites. Again, the
effect of uneven or biased survey coverage is not known.
Aspect
There is no clear pattern in the site data to suggest that
one ‘aspect’ was preferred over any other. However, factors
such as cold air drainage may still be shown to affect the
choice of occupation sites.
Previously surveyed areas
One problematic attribute of the various data sets is the
relationship between areas previously surveyed and the
apparent proximity of sites to fresh water. While the
proximity to fresh water is an important factor in the
location of prehistoric archaeological sites, the location and
extent (availability) of this resource will have changed
markedly through both time and space. Contemporary
survey coverage has tended to concentrate on those areas in
Figure 1 Distance to water for all known AAV sites types. The graph shows that 62.2% of all known AAV sites within the
study area occur within 100 metres of a fresh water source. The 1:25,000 hydrology layer used in ArcView 3.2 for
these calculations was modified to remove modern water features such as dams, reservoirs or drains.
10
Shaun Canning
close proximity to water, as most archaeologists ‘know’ that
this is the area likely to yield the most sites. While this
practice is common sense to a certain extent, it must also be
remembered that a reliance on such ‘expert’ bias may result
in an unrepresentative sample of the archaeological record.
For instance, the AAV digital survey data of the study area
were used to determine that a large proportion of survey
activity has been undertaken within 200 metres of a source
of fresh water (approximately 69%). Therefore, the
observed site patterning may be a product of archaeological
survey method as much as the result of prehistoric
Aboriginal behaviour (Witter 1992:270).
‘Weight of evidence’ and Dempster-Shafer models
Management is essentially about an organizational
response to uncertainty and risk. If all the parameters,
choices and decisions of an organizations activity were
known then active management would be redundant. In this
regard, the management of archaeological material shares
the same uncertainty and risk vocabulary as all other forms
of resource management. Management uncertainty is
‘inevitable in the decision making process’ (Eastman
2001:23) and cultural heritage management operates within
boundaries of considerable uncertainty. Uncertainty in
archaeology can come from many sources. This paper
considers several sources of uncertainty – namely
uncertainty in the existing body of knowledge (i.e. no
formal sampling, inconsistent survey intensity, overall lack
of survey coverage, poor visibility) and uncertainty as to
where other resources (sites or non-sites) are likely to be
located. The ignorance of where undiscovered sites or nonsites are located introduces the risk that any existing but
undiscovered archaeological resources may be destroyed
through management processes that allow inappropriate
activities to take place (i.e. unplanned development).
Biases in various data sources make it inappropriate to
apply or utilise the wide range of parametric statistical
techniques that are available in other archaeological
pursuits (Orton 2000; Shennan 1997). This means that we
cannot formulate answers to the predictive modelling
questions in terms of binary opposites (yes/no - site/nonsite) or standard probabilities. However, the masses of
available data can be combined in a manner that produces
valid results for any given data. Essentially ‘weight of
evidence’ techniques are a means of combining various
forms of evidence to support a hypothesis or hypotheses.
These forms of evidence may be binary (i.e. presence or
absence of sites) or may introduce other non-binary
variables, which can be difficult to assimilate into models
because the values are not binary (i.e. distance to water).
For the purposes of the following discussion, sites are
defined as geographic cells within a GIS that are known to
contain archaeological materials and cover an area of
100 m2 (10 m x 10 m). This is essentially the same
definition of an archaeological site applied by AAV. Nonsites are the opposite of this, i.e. cells of 100 m2 where no
archaeological material is believed to exist. There are
approximately 1.5 million cells of this size (100 m2) in a
map sheet such as the 7822-1-3-map sheet introduced
previously.
Given the body of knowledge for the study area (the
‘expert’ knowledge and existing data) it is possible to begin
to build a series of GIS layers that can be combined using
various processes to produce a site ‘likelihood’ surface. A
likelihood surface is not a quantitative probability
statement. It does not state that a site will or will not exist
at a specific point in space with a mathematical degree of
precision. A likelihood surface is an indication that, on the
balance of all the available evidence, a site is likely or
unlikely to exist at that point in space. This type of analysis
is particularly suited to cultural heritage management
where so many of the parameters are either impossible to
define, or where previous models are based upon untested
hypotheses. The weight of evidence approach allows for the
use of existing evidence in a manner that utilises particular
aspects of Bayesian probability theory.
The GIS layers constructed here are based upon the
enormous quantities of data generated by consultants and
academics in the study area over the last 25 years. However,
statements such as ‘sites will occur on prominences in the
landscape overlooking waterlines’ are not easily converted
to Boolean statements or queries for analysis in GIS. This
is where the use of the raster GIS IDRISI32 and its
‘BELIEF’ module becomes indispensable. ArcView 3.2
was used for the majority of this project. While this is an
adequate piece of software in its own right, IDRISI32 offers
a suite of powerful tools based upon Dempster-Shafer
belief theory, which is an extension of Bayesian probability
theory. ‘The basic assumptions of Dempster-Shafer theory
are that ignorance exists in the body of knowledge, and that
belief for a hypothesis is not necessarily the complement of
belief for its negation’ (Eastman 2001:34). The workings of
the IDRISI32 ‘BELIEF’ module are largely beyond the
scope of this paper, however the ‘Dempster-Shafer rule of
combination provides an important approach to
aggregating indirect evidence and incomplete information’
(Eastman 2001: 36) in GIS-based modelling.
In order to model a likelihood surface we need to decide
what is being hypothesized. In this case, the relatively
straightforward binary opposites (‘site’ and ‘non-site’) are
the two basic elements (hypotheses) of the decision frame.
Evidence to support one or other is proffered from
numerous sources. In this case, the evidentiary layers are
distance to water, slope, and proximity to known sites.
None of these attributes is easily described by internal
binary relationships (i.e. they are not interval
measurements, but are more like ratio measurements). For
instance, the statement ‘sites will occur at between 0 and
200 metres from a source of potable water’ cannot easily be
transformed into statements (map algebra) understood by
GIS. Prior knowledge and experience would suggest that
this is a valid statement for much of the archaeology of
Australia, but this does not allow us to determine the
relationship between distance to water and sites (i.e. are
more sites really located closer to water than further
away?).
The ‘BELIEF’ module in IDRISI32 contains numerous
procedures that allow for the variable nature of the model
attributes to be accounted for. When these processes are
used on a ‘distance to water’ layer for example, the
‘BELIEF’ module can be programmed to take into account
that the further away from a source of potable water we
move the more likely it is that each cell will be a non-site.
From the available sites data for the study area, we know
that approximately 80% of all known sites occur within 200
metres of a source of potable water (inclusive of biases).
The falloff in site frequency at distances greater than 200
metres from water is shown in Figure 1. Site frequencies
11
Figure 2 Graph showing the sigmoidal nature of the
relationship between archaeological sites and
water sources as the distance to water increases
(0 = site 1 = non-site).
decline at distances greater than 200 metres from potable
water, reaching almost zero beyond 1000 metres. The
relationship of site proximity to water is shown as a
sigmoidal (s-shaped) curve (Fig. 2).
Figure 2 shows the manner in which distance decay can
be viewed graphically. This type of curve best represents the
relationship between the distance to water and the number
of sites, as there are no ‘hard’ boundaries delineating where
site distributions and densities change or do not change.
Close to water sources, the probability of encountering a
non-site is low (i.e. nearer 0). As we move further away
from a source of water, the probability of encountering a
non-site increases to the point where it is theoretically
100%. A sigmoidal curve demonstrates this cumulative
nature of distance to water and site distribution. As we move
further away from the water source, the closer we are to the
theoretical point at which no further sites will be found (i.e.
the likelihood of a non-site approaches 100%). Naturally,
we also move closer to encountering the next source of
potable water, so there are limits to the application of this
technique.
Other landscape attributes may be modelled in a similar
manner. Slope is the other variable for which we have a
significant amount of prior or existing expert knowledge, as
well as the limited (and biased) quantitative data from GIS
analyses. The accumulated data suggests that sites will
occur in areas where the slope is between 0º and
approximately 10º and that the sites will most commonly
occur near an area of topographic change (i.e. where the
plains meet the hill slopes of the river valleys). The same
GIS processes described above can be run on these
attributes to create a series of ‘likelihood’ surfaces to be
incorporated into the final weight of evidence model. The
known data for the relationship between site location and
slope, for instance, can be processed to create two separate
surfaces that show the likelihood of the occurrence of both
sites and non-sites.
Because there is a degree of uncertainty in the data, and
the completed modelling exercise should reflect this, the
layers must be ‘scaled’ or weighted to ensure that the results
do not indicate 100% certainty for any predicted value.
IDRISI32 makes this process comparatively easy. Layers
12
can be scaled (i.e. multiplied) by any factor to reflect the
degree of uncertainty. For instance, the Distance to Water
layer used in the modelling exercise here has been weighted
using a factor of 0.8 (80%). This simply means that the
known distribution of archaeological sites (i.e.
approximately 80% within 200 metres of water) has been
accounted for, while factoring in an estimate of the
uncertainty (i.e. the other 20% that occur at varying
distances greater than 200 metres from water). ‘FUZZY’
logic is applied within IDRISI32 to model those cells where
it is unlikely that a site will occur (non-site). Table 1
presents a summary of the layers that were created in
IDRISI32 for incorporation into the final aggregated
‘BELIEF’ model. A comprehensive discussion of the
operation of the IDRISI32 ‘BELIEF’ and ‘FUZZY’ functions
is provided by Eastman (2001).
When the various layers are entered into the IDRISI32
‘BELIEF’ module, the surface produced shows the
likelihood of a cell being a non-site. Because uncertainty
has been factored into this model, no values greater than 0.8
are used. Where a value of 0.8 is shown, the model predicts
that there is a 80% likelihood that the cell in question will
be a non-site. Where the value returned by the model is low,
i.e. 20%, the model predicts that there is an 80% likelihood
that the cell is a site (i.e. 20% likelihood of non-site equals
an 80% likelihood of a site).
Interpreting the model
The site likelihood surface generated from the available
data should not be seen as a definitive probabilistic model.
Interpreting this likelihood surface is relatively
straightforward. Where the resultant value for any cell is
high (i.e. >0.70) there is a high likelihood of encountering
cells (remembering that each cell represents 100m2) that do
not contain any archaeological sites (i.e. non-sites). Where
the value is low (i.e. <0.20) there is a high likelihood of
encountering cells that will contain archaeological material.
It is explicitly recognised that other environmental and
socio-cultural attributes will affect the presence or absence
of archaeological sites in any area. In the current study area,
for example, the location and distribution of siliceous lithic
material can be used as a predictor variable for locating
prehistoric silcrete quarries. The difficulty however, is that
this material outcrops within the river valley slopes already
known to be archaeologically significant (Webb 1995). The
river valleys are already given significant ‘weight’ within
the model due to the inclusion of ‘slope’ as a variable, so no
extra ‘weight’ was thought to be required.
Conclusion
While it is undeniable that statistically rigouress
quantitative predictive models should be the ultimate aim of
any research driven modelling exercise, the reality is that
this is often an impossible goal. In the case presented here,
the majority of the data available for use already existed in
the AAV database. However, these data are not suited for
use in traditional statistical analyses, as various biases have
been introduced during their collection (i.e. no standardised
survey methods, poor sampling techniques, no visibility
quantification). Rather than ignore this huge data set, the
methods presented here specifically acknowledge that the
data are biased, and allow them to be incorporated into a
model of archaeological sensitivity. Nor are models of this
type intended to supplant archaeological survey. The
Shaun Canning
intention of the model presented here is to focus
management attention on those areas believed to be the
most archaeologically sensitive. The resultant ‘likelihood’
surface is an ideal tool for use by local government planners
and other statutory authorities to flag those areas that must
have archaeological investigations undertaken prior to any
land altering activities. The models are not intended to
explain prehistoric human behaviour of the area in question,
but to further assist in the preservation of representative
samples of the archaeological record for the researchers of
tomorrow.
Acknowledgements
The Basalt Plains Archaeology Project was made
possible through the generous support of the Australian
Research Council via an APA (I) SPIRT grant incorporating
a three-year PhD scholarship. Generous financial support
was also forthcoming from the industry partner, Aboriginal
Affairs Victoria. Dr Richard Cosgrove and Dr David
Frankel of La Trobe University supervised the project, and
their contribution is gratefully acknowledged. Thanks are
due to the huge number of individuals who were involved
with the project in one-way or another. Special thanks to
Ken Mulvaney, Sean Ulm, Christine Martin, Brian Egloff
and Harry Allen for comments on earlier drafts of this
paper.
References
Allen, K. 1990 Modelling early historic trade in the Eastern Great
Lakes using Geographic Information Systems. In K. Allen,
S. Green and E. Zubrow (eds) Interpreting Space: GIS and
Archaeology, pp. 319-329. London: Taylor and Francis.
Allen, J., Cosgrove, R. and Brown, S. 1988 New archaeological
data from the Southern Forests Region, Tasmania: A
preliminary statement. Australian Archaeology 27:75-88.
Altschul, J. 1988 Models and the modelling process. In J.W. Judge
and L. Sebastian (eds) Quantifying the Present and
Predicting the Past: Theory, Method and Application of
Archaeological Predictive Modelling, pp. 61-96. Denver: U.S.
Department of the Interior, Bureau of Land Management.
Altschul, J. H. 1990 Red Flag Models: The use of modelling in
management contexts. In K. Allen, S. Green and E. Zubrow
(eds) Interpreting Space: GIS and Archaeology, pp. 226-238.
London: Taylor and Francis.
Binford, L. R. 1964 A consideration of archaeological research
design. American Antiquity 29: 425-441.
Brooks, A., Hudak, G. J., Gibbon, G. and Hobbs, E. 2000
Management summary. In Mn/Model: A Predictive Model of
Pre-contact Archaeological Site Location for the State of
Minnesota, pp. 2-8. St Paul: Minnesota Department of
Transport.
Butzer, K. 1982 Archaeology as Human Ecology. Cambridge:
Cambridge University Press.
Canning, S. 2003 Site Unseen: Archaeology, Cultural Resource
Management, Planning and Predictive Modelling in the
Melbourne Metropolitan Area. Unpublished PhD thesis,
Department of Archaeology, School of European and
Historical Studies, La Trobe University. Bundoora,
Melbourne.
Church, T., Brandon, R. J. and Burgett, G. R. 2000 GIS application
in archaeology: Method in search of theory. In K.L. Wescott
and R.J. Brandon (eds) Practical Applications of GIS for
Archaeologists: A Predictive Modelling Kit, pp. 135-155.
London: Taylor and Francis.
Dalla Bona, L. 1994 Cultural Heritage Resource Predictive
Modelling Project. Volume 1: Introduction to the Research.
Ministry of Natural Resources, Ontario.
du Cros, H. 1989 The Western Ring Road Archaeological Study.
Stage 2, Volume 2. Report to VicRoads, Melbourne.
du Cros, H. 1990 The Sydenham Corridor: A Cultural Heritage
Study. Victoria Archaeological Survey, Melbourne.
du Cros, H. 1991 The Werribee Corridor: An Archaeological
Survey. Aboriginal Affairs Victoria. Melbourne.
Layer Name
Cell Hypothesis
Description
Known Site
Site
Those cells where a known site exists, plus
all cells within 300 m of a known site.
FUZZY logic applied, using sigmoidal
monotonically decreasing curve. The further
away from a known site, the less likely it is
that a cell will be a site.
Distance to Water
Non-Site
Slope
Non-Site
Cells greater than 300 m to a source of
potable freshwater. Cells between 0-300 m
have FUZZY logic applied using a
sigmoidal monotonically increasing curve.
The greater the distance away from potable
water, the higher the likelihood a cell is a
non-site.
Cells where the slope angle exceeds 25º are
more likely to be non-sites. Those less than
25º are more likely to contain sites. FUZZY
logic is applied, using a sigmoidal
monotonically increasing curve. Those
values between 0º and 25º are weighted
more heavily than those greater than 25º.
Justification(s)
Other sites will occur in close proximity to
existing sites. As the distance between sites
increases, so does the likelihood that a cell
will be a non-site. The distribution of
material from prior surveys demonstrates
that the presence of sites in a cell is
strongly influenced by the location of other
archaeological material.
Distance to freshwater affects the
distribution of site(s). The exact pattern is
not known, although the overwhelming
majority of sites in the study area (~90%)
that occur within 300 m of a permanent
water source. This layer is weighted to
reflect this phenomenon.
The distribution of archaeological sites
shows that sites tend to occur on slopes of
between 0º and 25º. This is not to say that
no sites will occur on slopes greater than
25º, rather that it is less and less likely as
the slope increase. The FUZZY logic
applied factors this into the aggregation of
evidence.
Table 1 The various layers created for the 7822-1-3 Mapsheet, and the processes applied to them within IDRISI32.
13
Eastman, J. R. 2001 IDRISI32 Release 2: Guide to GIS and Image
Processing. Worcester, Massachusetts: Clark Labs, Clark
University.
Ebert, J. 1988 Modelling human systems and “Predicting” the
archaeological record: The unavoidable relationship between
theory and method. American Archaeology 7:3-8.
Ebert, J. 2000 The state of the art in ‘Inductive’ predictive
modelling: Seven big mistakes (and lots of smaller ones). In
K. L. Westcott and R. J. Brandon (eds) Practical Applications
of GIS for Archaeologists: A Predictive Modelling Kit, pp.
129-134. London: Taylor and Francis.
Gibbon, G. 1998 Archaeological predictive modelling: An
overview. In Mn/Model: A Predictive Model of Pre-contact
Archaeological Site Location for the State of Minnesota. St.
Paul: Minnesota Department of Transport.
Green, E. L. 1973 Location analysis of prehistoric Maya sites in
Northern British Honduras. American Antiquity 38:279-293.
Higgs, E. S. and Vita-Finzi, C. 1972 Prehistoric economies: A
territorial approach. In E. S. Higgs (ed.) Papers in Economic
Prehistory, pp. 27-36. Cambridge: Cambridge University Press.
Kamermans, H. and Rensink, E. 1999 GIS in Paleolithic
archaeology: A case study from the Netherlands. In L.
Dingwall (ed.) Computer Applications and Quantitative
Methods in Archaeology Proceedings of the 25th Anniversary
Conference, Archaeology in the Age of the Internet,
University of Birmingham, April 1997. Oxford: Archeopress.
Kohler, T. A. 1988 Predictive locational modelling: History and
current practice. In J. W. Judge and L. Sebastian (eds)
Quantifying the Present and Predicting the Past: Theory,
Method and Application of Archaeological Predictive
Modelling, pp. 19-60. Denver: U.S. Department of the
Interior, Bureau of Land Management.
Kohler, T. A. and Parker, S. C. 1986 Predictive models for
archaeological resource location. In M. B. Schiffer (ed.)
Advances in Archaeological Method and Theory, Volume 9,
pp. 397-452. New York: Academic Press.
Kvamme, K. 1988a Development and testing of quantitative
models. In J. W. Judge and L. Sebastian (eds) Quantifying the
Present and Predicting the Past: Theory, Method and
Application of Archaeological Predictive Modelling, pp.
325-428. Denver: U.S. Department of the Interior, Bureau of
Land Management.
Kvamme, K. 1988b Using existing Archaeological survey data for
model building. In J. W. Judge and L. Sebastian (eds)
Quantifying the Present and Predicting the Past: Theory,
Method and Application of Archaeological Predictive
Modelling, pp. 301-323. Denver: U.S. Department of the
Interior, Bureau of Land Management.
Lewis, A., MacNeill, R. and Rhoads, J. 1996 East Gippsland
Aboriginal Cultural Site Modelling. Melbourne: Department
of Conservation and Natural Resources, Aboriginal Affairs
Victoria, and Centre for Resource and Environmental
Studies, Australian National University.
MacNeill, R. 1998 GIS on the brink: GIS applications at
Aboriginal Affairs Victoria. Paper presented at AURISA ’98
- the 26th Annual Conference of AURISA. Perth, Western
Australia, November 1998.
McConnell, A. 1995 Archaeology Potential Zoning: A Strategy for
the Protection of Aboriginal Archaeological Sites in
Tasmanian State Forests. Forestry Tasmania, Hobart.
McConnell, A. 2002 Aboriginal Heritage Management in
Victorian Forests. Draft Project Report Volume 2. Central
Highlands Region. Department of Natural Resources and
Environment, Melbourne.
McDonald, J. 1996 The conservation of landscapes: A strategic
approach to Cultural Heritage Management. In S. Ulm, I.
Lilley and A. Ross (eds) Australian Archaeology:
Proceedings of the 1995 Australian Archaeological
Association Conference, pp. 113-121. St. Lucia: University
of Queensland.
14
McManamon, F. P. 1984 Discovering sites unseen. In M.B.
Schiffer (ed.) Advances in Archaeological Method and
Theory, Volume 8, pp. 223-292. New York: Academic Press.
Minnesota Department of Transport 2000 History of the Mn/Model
Project Minnesota Department of Transport, On-Line Report
at http://www.mnmodel.dot.state.mn.us/pages/history.html.
Moon, H. 1993 Archaeological Predictive Modelling: An
Assessment. Victoria, British Columbia: Resources Inventory
Committee.
Neuman, W. L. 1997 Social Research Methods: Qualitative and
Quantitative Approaches. Boston: Allyn and Bacon.
Orton, C. 2000 Sampling in Archaeology. Cambridge Manuals in
Archaeology. Cambridge: Cambridge University Press.
Plog, F. and Hill, J. 1971 Explaining variability in the distribution
of sites. In G.J. Gummerman (ed.) The Distribution of
Prehistoric Population Aggregates. Prescott, Arizona:
Prescott College Anthropological Reports 1:7-36.
Rindos, D. 1989 Diversity, variation and selection. In R.D.
Leonard G.T. and Jones (eds) Quantifying Diversity in
Archaeology, pp. 13-23. Cambridge: Cambridge University
Press.
Roper, D. C. 1979 The method and theory of Site Catchment
Analysis: A review. In M. B. Schiffer (ed.) Advances in
Archaeological Method and Theory, Volume 12, pp.119-140.
New York: Academic Press.
Shennan, S. 1997 Quantifying Archaeology. Edinburgh: Edinburgh
University Press.
Steward, J. 1938 Basin-Plateau Aboriginal Socio-Political Groups.
Washington, D.C.: Bureau of American Ethnology Bulletin
120.
Thomas, D. H. 1973 An empirical test for Steward’s model of
Great Basin settlement patterns. American Antiquity 38:155176.
Thomas, D. H. 1975 Non-site sampling in archaeology: Up the
creek without a site. In J.W. Mueller (ed.) Sampling in
Archaeology, pp. 61-81. Tucson: University of Arizona Press.
van Leusen, M. 2002 Pattern to Process: Methodological
Investigations into the Formation and Interpretation of
Spatial Patterns in Archaeological Landscapes. Unpublished
PhD thesis, Groningen Institute for Archaeology, University
of Groningen, The Netherlands.
Vita-Finzi, C. and Higgs, E. S. 1970 Prehistoric economy in the
Mount Carmel area of Palestine: Site Catchment Analysis.
Proceedings of the Prehistoric Society 36:1-37.
Warren, R. 1990 Predictive modelling in archaeology: A primer. In
K. M. Allen, S. W. Green and E. Zubrow (eds). Interpreting
Space: GIS and Archaeology, pp. 90-111. London: Taylor
and Francis.
Warren, R. E. and Asch, D. L. 2000 A predictive model of
archaeological site location in the Eastern Prarie Peninsula.
In K. L. Wescott and R. J. Brandon (eds) Practical
Applications of GIS for Archaeologists: A Predictive
Modelling Kit, pp. 5-32. London: Taylor and Francis.
Webb, C. 1995 The Identification and Documentation of Silcrete
Quarries. Aboriginal Affairs Victoria, Melbourne.
Westcott, K. L. and Kuiper, J. 2000 Using GIS to model prehistoric
site distributions in the Upper Chesapeake Bay. In K. L.
Wescott and R. J. Brandon (eds) Practical Applications of
GIS for Archaeologists: A Predictive Modelling Kit, pp. 5972. London: Taylor and Francis.
Willey, G. 1953 Prehistoric Settlement Patterns in the Viru Valley,
Peru. Washington, D.C.: Bureau of American Ethnology
Bulletin 155.
Winterhalder, B. 2000 Models. In J.E. Terrell (ed.) Darwin and
Archaeology: A Handbook of Key Concepts, pp. 201-223.
Westport, Connecticut: Bergin and Garvey Publishers.
Winterhalder, B. 2001 The behavioural ecology of huntergatherers. In P. Rowley-Conwy (ed.) Hunter-Gatherers. An
Interdisciplinary Perspective, pp. 13-38. Cambridge:
Shaun Canning
Witter, D. 1992 Region and Resources. Unpublished PhD thesis.
Department of Prehistory, Australian National University,
Canberra
Zubrow, E. 1990 Modelling and prediction with Geographic
Information Systems: A demographic example from
prehistoric and historic New York. In K. M. Allen, S. W.
Green and E. Zubrow (eds) Interpreting Space: GIS and
Archaeology, pp. 307-317. London: Taylor and Francis.
FORENSIC
ARCHAEOLOGY:
A Textbook
by
John Hunter and Margaret Cox
Routledge
2005
ISBN 0-4152-7311-0
RRP US$115.00
(hardcover)
A RECORD OF STONE:
THE STUDY OF AUSTRALIA’S
FLAKED STONE ARTEFACTS
Simon Holdaway and Nicola Stern
Aboriginal Studies Press
2004
ISBN 0-8557-5460-5
RRP A$49.95
(paperback)
15
Book Reviews
Water views: Water-based survey methods on Cowan Creek, New South
Wales
Abstract
A survey of shell midden sites within the Cowan Creek
area, which lies to the north of Sydney, was conducted over
a period of five months. The survey design was based upon
a site location printout obtained from the New South Wales
National Parks and Wildlife Service Aboriginal Heritage
Information Management System. Due to the ruggedness of
the terrain within the area, the sites were first identified
from the water, and GPS was used to verify the coordinates.
The location of the sites was then confirmed on foot.
During the survey potential habitation sites were also
examined, and a total of thirty-eight new sites containing
middens, were identified and recorded for inclusion on the
site register with National Parks. Recorded site density, as a
result of the survey, has increased along Coal and Candle
Creek from 0.5 sites /linear km, to 8.5 sites/linear km.
Surface examinations of the midden deposits in this area
suggest that Trichomya hirsuta and Crassostrea
commercialis were strongly favoured by Aboriginal people
in the past. The results of the survey indicate that the waterbased survey method adopted for the purposes of this
project was highly efficient in the location of sites
containing middens in Cowan Creek, within a short period
of time. The results also indicate that a great deal of this
information is being rapidly degraded within the area, with
the majority of sites displaying damage due to both natural
and anthropogenic influence, despite the protection
afforded by inclusion of all of these sites within a National
Park.
Introduction
Shell midden analysis has been crucial to the
understanding of many aspects of prehistoric life within
Australia. The study of occupation sites of this type may be
advantageous to a researcher for a variety of reasons. Shell,
being the primary component of middens, is highly durable,
which in combination with the frequent large size exhibited
by many deposits, results in a highly visible source of
archaeological information. The size and depth of a midden
deposit may also contain a wealth of valuable
archaeological material, providing chronological evidence
of cultural adaptation and resource utilisation. The
environment within a midden itself is generally conducive
to the long-term preservation of such evidence, including
faunal remains, stone tools, shell scrapers and charcoal. The
structure of the midden, and the taphonomic processes
involved in its deposition, also provide a rather unique
representation of the environmental and cultural conditions
that existed during the period of occupation. This
combination of features has provided a valuable medium for
study in the past, with midden analysis receiving attention
Department of Biological Sciences. Division of Environmental and Life
Sciences, Macquarie University, Sydney, NSW 2109. Australia. Email:
[email protected].
16
from a variety of disciplines, attempting to answer questions
relating to Aboriginal occupation.
In the past excavations have been performed on shell
middens in order to obtain information about diet,
technological adaptation, demography and social structure
of prehistoric cultures. While the results of such analyses
have certainly acted to broaden the knowledge and
understanding within the archaeological community, in
order to access this information there inevitably results in
some degree of disturbance/damage to the integrity of the
site or sites in question. While it may be argued that the
archaeological significance of a site is based upon the
information it contains, and that the only way to access this
information is to excavate, it has been suggested that in
many cases this has acted to reduce, rather than increase the
archaeological significance of the site itself (Bowdler
1984). The basic dilemma, it would seem, is whether a site
is more valuable if it remains intact, or does its value only
become truly appreciated once the information it contains is
examined and assessed through excavation?
In some instances the answer is simple. For example,
where an impending threat exists that will lead to the total
destruction and inevitable loss of the site/sites, it may be
argued quite reasonably that excavation is the only course of
action in order to salvage archaeological information. The
most common threat to Aboriginal sites comes from the
increasing pressures of industrial development and
population growth. Within many coastal regions of
Australia the location of Aboriginal midden sites
unfortunately demonstrate a propensity to coincide with
land desirable for real estate, which has resulted in the
destruction and loss of many sites. Within some areas
however, middens appear to have been granted a small
reprieve due to their fortuitous location within the bounds
of a National Park, or other area deemed unsuitable for
development due to the topographic or geological nature of
the terrain. Within these locations the archaeological
integrity and significance of sites may remain virtually
intact and undisturbed.
The New South Wales National Parks and Wildlife
Service (NPWS) are the legal custodians of Aboriginal sites
in New South Wales. Their most recent approach to site
conservation includes a reluctance to allow any form of
excavation for research purposes, on sites within some
national park areas around Sydney (K. Przywolnik, NPWS
pers. com.). Although this may protect the sites from the
impacts of researchers, various agents of disturbance can
frequently still impact upon the integrity, and thereby the
significance of sites located within the national park
boundaries. The NPWS are also responsible for maintaining
a site register, which lists all reported sites and their
location, and has often provided a useful source of
information to researchers who are interested in a particular
area. Attenbrow (1991) noted during her 1989/1990 survey
of Aboriginal sites located within the Port Jackson area, that
the information recorded on the register frequently comes
from different sources, often resulting in variation in terms
of the quality and level of information recorded for each
site. The large number of sites recorded on the register
would also make monitoring and assessment of the
individual sites in terms of their condition, deterioration and
level of preservation, virtually impossible. It was also noted
during the 1989/1990 survey, that previous site recording
within the Port Jackson area also displayed an imbalance in
terms of the types of site that had been recorded, and the
environmental zones that had received attention. This
resulted in areas of the catchment that, according to the
register, had either not been surveyed or contained very few
sites. The consequences of such disjointed surveying and
recording techniques may therefore result in a misleading
and biased representation of occupation and site usage
within an area.
Traditional midden analyses have been conducted using
a site by site strategy, however it has been argued that the
archaeological significance of a site may lie in its role
within the contextual framework of the grouping of sites, or
site complexes that exist in the area (Vinnicombe 1980).
The implications of this are important in terms of previous
and future research, and for the design and implementation
of appropriate land management strategies, suitable for the
conservation of archaeological and cultural heritage.
Although it is impossible to protect every known Aboriginal
site within a particular region, it may be more practical and
beneficial (as Vinnicombe recommended in 1984:117) to
preserve a small group of sites, which reflect a variety of
activities within an area. The first step towards achieving
this aim is to devise a suitable method to enable an
assessment of the distribution, condition, and content of all
sites located within a particular area.
In view of the rapidity at which information relating to
many aspects of site usage has, and may continue to be lost,
together with the fact that in many instances a thorough
analysis through excavation may no longer be feasible, a
new rapid-assessment technique is needed. In this way it
may be possible to create a catalogue of information that
attempts to include a representative sample of sites situated
within a region, which would provide a useful aid for the
identification of areas in need of protection. An appropriate
technique should incorporate a re-assessment of the
condition of sites previously recorded within an area, and
also an intensive spot survey within other areas that
apparently are devoid of sites yet display good potential, in
order to fill in any ‘gaps’ that exist in the record. A suitable
strategy should also aim towards obtaining the maximum
information with minimum disturbance to the sites
themselves in order to maintain their integrity. The survey
method devised for the project conducted in the Cowan
Creek area attempts to satisfy all of these criteria.
The study area
Cowan Creek lies at the lower end of the Hawkesbury
River, and forms part of an extensive drowned river valley
system located 32 km north of Sydney (see Fig. 1). It is
surrounded by Ku-ring-gai Chase National Park and as such
as been largely unaffected by development. The entrance to
Cowan Water lies approximately 3 km from the ocean inlet
at Broken Bay, and extends some 15 km in a southwestern
direction. Various smaller creeks, inlets and bays exist in the
Cowan Water area including Cowan Creek, Smiths Creek,
Coal and Candle Creek, Jerusalem Bay, Refuge Bay and
America Bay. Many natural drainage channels and small
creeks also exist along its reaches and contribute the
freshwater component to the environment, which is
especially evident during times of heavy rainfall. Tidal
influence extends to the farthest ranges of Cowan Water,
with some of the smaller creeks displaying stands of
mangroves and mudflats at their upper reaches. Small
pockets of seagrass beds also exist within many of the bays
and inlets in the area.
Due to the scenic quality of the area, it is a popular and
frequently utilised stretch of water favoured by recreational
visitors, especially boating and fishing enthusiasts. Despite
this, due to the inaccessible nature of the terrain, combined
with its characteristic poor soils associated with the
sandstone environment, the area has experienced minimal
impact, either during initial colonial settlement or as a
result of the subsequent expanse and spread of European
urbanisation. The only exception to this, and occurring
prior to 1894 when Ku-ring-gai Chase was declared as a
reserve for recreational use, involved the logging of some
stretches of forest along Cowan Creek and Bobbin Head
(Fairly 1972).
Hawkesbury sandstone predominates in the area, which
weathers to produce the numerous caverns that are
synonymous with the landscape within the region. Despite
the associated poor sandy soils, vegetation varies from
dense forest within gullies and better-watered sheltered
areas to heath, scrub and woodland communities restricted
to exposed regions displaying shallow poorly drained soils.
Throughout most of the study area, which has not
experienced bushfires since 1994, it was noted during the
survey period that the groundcover was extremely thick
with leaf litter, Casuarina needles and small shrubs.
Survey design and methodology
The design of an archaeological survey method that
would satisfy all of the criteria set out in the project was an
important factor. The survey was initially based upon a
printout obtained from the NPWS sites register, also known
as the Aboriginal Heritage Information Management
System (AHIMS). This register contained information
about the Aboriginal sites that have been recorded in the
Ku-ring-gai Chase National Park since 1980, and indicated
that 65 registered Aboriginal sites containing middens,
were located within the boundaries of the study area. It did
not provide an indication of the condition, or the rate of
deterioration to the sites since they were initially recorded.
Traditionally survey strategies have been based upon a
stratified surveying technique (Byrne 1984; Vinnicombe
1980). However where the terrain is rugged and
inhospitable it is difficult to adhere to such predetermined
strategies. Taking this into account, and for the purpose of
this type of project, a different strategy was devised which
was expected to yield the optimum results in terms of site
location and assessment. Initially the project was based
upon the location and examination of known sites, and their
assessment in terms of a simple review of their condition
and surface content. A sufficient strategy was therefore
simply to work through the site printout and locate the sites
using the specified map coordinates (given as eastings and
northings), and brief description of the location and
accessibility of each site.
The survey was also unusual in that the focus of the
project was initially on the examination of midden sites
17
Water views: Water-based survey methods
only, which by their very definition tend to be situated
within close proximity to the water, and are frequently
conspicuous. As a consequence it was an advantage to
conduct the survey by boat using the site coordinates, a
map, and a handheld GPS. Subsequent to GPS verification
that a site was within close proximity, the actual location of
the site was confirmed on foot. As with many survey
strategies however, what appears to be a simple procedure
on paper proves to be something quite different when put
into practice in the field.
The problems encountered during the survey were
twofold. Firstly the nature of the terrain, which was
frequently steep and covered in a dense understorey of
18
Percentage
Figure 1 Location of study area and associated tributaries of the Hawkesbury River: Windsor to Broken Bay. (Source:
adapted from original in Johnson 1997.)
Site Type
Figure 2 Site type/abundance of new sites recorded in
Cowan Creek.
shrubs and leaf litter, proved to be a formidable physical
impediment to site location in some areas. Secondly, many
of the coordinates given for the sites on the printout tended
to be inaccurate. This may be a consequence of the fact that
many of the sites had been recorded during the 1980s, when
the GPS system was subject to selective availability, or
reflect a lack in continuity of recording methods as sites
were recorded by a number of different people, and for a
variety of projects. This greatly delayed the location of
many sites. In some instances it was impossible, despite
intensive searching of the bushland to locate some sites that
were listed on the register.
Site no.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Conducting the survey by boat was beneficial within
this area due to the increased visibility it afforded in this
type of terrain. It allowed areas to be examined for the
presence of sites, which would otherwise have been almost
impossible in a land-based survey. During the initial stages
of surveying it became apparent that the site density in
some areas differed significantly from the information
suggested by the register. It became clear that where one site
was listed frequently there would be others close by. As a
result the technique was expanded in order to include
intensive surveying for sites from the water, combined with
spot checks of areas located within the vicinity of registered
Site name
Site type
Contents
Cockle Creek
Smiths Ck, S1
Cotton Tree Bay/ Cowan Ck 3
Charcoal Rock Shelter
Coal & Candle Ck, S1
Fig Tree Pt
Xanthorroea R.S
Red Fish Cave, C & C Ck
Coal & Candle Ck, Waterfall
Mangrove Midden
Halletts Beach S1
Coal & Candle Ck, OS1
Sandy Bay 1
Sandy Bay 2
Coal & Candle Ck, RS1
Midden Point
RS
RS
M
ORS
OH
OH
OH
ORS
RS
RS
ORS
RS
ORS
RS
RS
OM
ORS
RS/M
OM
OM
OM
OM
OM
OM
OM
OM
OM
RS
OM
ORS
ORS
RS
OM
OM
RS
ORS
ORS
OM
A, S, T, P, Ch, Bs
A, S, T, P, Ch
A, S, T,
S,T,C, Ch, Art
A, S. T,C, Ch, Bs, Us
T,C,S,A, Fl
T, S, A, C, P,
T, S, A, C, Ch, Us, Wa
T, S, C, O, Ch, Art
T, S, Ch
T, S, P, Tt, Fl, Us,Ch, Art
T, A, S, Bs, Ch
A, T, S, P, Ch
S, C, T,A, Bs, Ch
A, S, T, C, Us, Ch
A, P, S, T, Ch
A, S, P, T, C, Fl, Ws
A, S, T, C, B, St, Ch
A,S,T,C
A,S,T,C
A,S,T,C
A,S,T,C
S,T,C, Ch
S,T
A,T,S,C, Ch
T,S,C, Ch
T,S,C,A,Ch
T,S,C,Ch
T,S,A,P,Ch, Us
T,S,A,C,P,Ch, Art
T,S,Ch, Art
T,S,C, Ch
T,S,P,A,C, Ch
T,S,C,A, Ch
T,S,C, Ch,Us
T,S,A,C, St, Fl
T,C,S, Ch
T,A,C,S
Key to symbols
RS – rock shelter
OM – open midden
P – Pyrazus ebeninus
S – Saccostrea commercialis
Bs – burnt shell
Us – utilised shell
OH – overhang
Art – drawings
T – Trichomya hirsuta
Ch – charcoal
St – stone artefact
Ws – worked stone
ORS – open rock shelter
A – Anadara trapezia
C – Chama fibula
O – ochre
Fl – stone flake/s
Table 1 Sites recorded during the survey in the Cowan Creek area, New South Wales.
19
% of total
Sites per linear km
Location
Ranking
Figure 3 Comparison of site density with other areas along
the Australian coast.
Note: this information compiled from Coutts et al.
(cited in Vinnicombe 1980); Stockton 1972, 1977;
Sullivan 1976; Vinnicombe 1980.
Figure 4 Site condition recorded during the survey in the
Cowan Creek area.
sites. In this way the register provided an indicator of both
the location in which sites were known to occur, and also
areas in which sites were likely to be found. One area that
particularly stood out in terms of an apparent dearth of sites
already identified was Coal and Candle Creek, with only
two sites listed on the register. Ultimately this area, which
provided an environment that was favourable for Aboriginal
occupation in terms of resource availability and habitable
rockshelters, proved to be a focus for site surveying (see
Fig. 1 for location).
Once located, the condition of each site was assessed
according to its level of preservation, species composition
and integrity. The level of preservation was estimated
according to the percentage of intact shells evident on the
surface. A determination of species composition was made
according to the number of intact and/or identifiable
fragments present within a subjectively selected 30 cm
square. Site integrity and condition was assessed according
to the proportion of the site as a whole that displayed
disturbance or degeneration, and where possible included
the nature of the cause, such as trampling, digging or
erosion. For the purposes of this survey, midden locations
were classified as either open, within rock shelters, or open
rock shelters. This third category was necessary in order to
discriminate between middens identified within rock
shelters and those situated under rock overhangs of less than
1 metre (measured from the rear of the shelter).
Results
The survey, due to the constraints of time, was limited to
one day per week over a period of five months, with the
surveying team comprising three people (one person
driving the boat and two people surveying). During this
time 60 out of the registered 65 sites were located and
examined, and 38 new sites were identified and recorded
(Table 1). The survey included an assessment of potential
habitation sites, primarily sandstone rock shelters which
were numerous and commonly associated with this type of
environment due to natural weathering processes. The
identification of these sites was based on an inspection of
the shelter floor for any visible signs of occupation,
including faunal remains, artefacts, manuports, or the
presence of art on the rock walls. The majority of the newly
identified sites were located along the eastern side of Coal
20
Key to ranking system:
1 >80% intact: majority of midden undisturbed and
displaying good preservation
2 60-80% intact: good preservation with large areas
intact
3 40-60% intact: some areas eroded/degraded, but
with pockets of intact shells visible in
places
4 20-40% intact: majority
of
surface
highly
fragmented/eroded, some small
patches intact
5 0-20% intact: little/no intact shells evident,
generally midden composed of grey
powdery substrate with isolated
larger shell fragments visible
and Candle Creek (western side not surveyed), and included
rockshelters with art and shell middens, open middens and
rockshelters containing artefacts. The most abundant
occupation site type that was identified during the survey
was open midden sites (42%), followed by rock shelters
containing midden deposit (see Fig. 2). This may be a
consequence of Aboriginal site selection and usage within
the area, or simply a reflection of preferential visibility. Site
density was calculated along Coal & Candle Creek, and
yielded a density of 8.5 sites per linear km, which is very
high compared to the results obtained from other studies
along the Australian coast (Fig. 3). This may be due to a
higher level of site preservation in this area compared to
other regions that have experienced a greater level of impact
from urban development. Most of the sites were located
within 20 m of high tide, with the exception of one large
rock overhang displaying numerous charcoal drawings,
which was situated approximately 100 m above the water
level. Although accurate measurements of depth were not
practicable for the purposes of this survey, estimates were
made and indicated that the majority of middens (70%)
displayed a deposit depth >30 cm. The majority of new sites
displaying good preservation and integrity were located
within rockshelters or shallow rock overhangs, with open
midden sites generally evidencing a greater level of
disturbance and erosion.
Many of the sites that were located and examined
showed some degree of disturbance. The types of
disturbance varied from site to site, but included the impact
of animals (including humans) digging holes into the
deposit, trampling of the surface, gullying due to natural
drainage through the deposit, bioturbation caused by the
growth of plants, and the erosional impacts of wave action.
Each of the sites was assessed for damage and subsequently
assigned a ranking between 1 to 5 according to their level of
preservation and integrity based upon an examination of the
surface. A rank of 1 given to a site indicates the highest
level of integrity, and 5 the lowest (see Fig. 4). The majority
of sites recorded during this survey appeared to be in a poor
condition, with most assigned a ranking of three or more.
Despite the apparent poor condition displayed by many sites
within this area, their potential in terms of further
archaeological work, and from a cultural heritage value, is
not refuted. This is due to the fact that although the general
condition of a site may be poor, frequently ‘pockets’ of
relatively intact or undisturbed midden material remain
within many of the rockshelter sites that were examined,
offering an opportunity for more detailed analysis.
Species composition and midden contents
The co-dominant shellfish species based on the surface
counts of the middens in the study area, were the Sydney
rock oyster (Saccostrea commercialis) and the Hairy mussel
(Trichomya hirsuta). Other species that were identified but
occurred in lesser amounts within the deposits were the
whelk (Pyrazus ebeninus), Sydney cockle (Anadara
trapezia), and Spiny oyster (Chama fibula). Each of these
species is commonly represented within midden deposits
around Sydney and are strongly associated with an
estuarine environment. This may indicate that similar
conditions have persisted within the area throughout the last
occupational period. Individual broken specimens of the
giant turban shell Turbo torquatus, were also identified at
two sites. This is not an estuarine species, and must have
been transported from its rocky shore habitat situated
around Broken Bay. Its presence may be suggestive of fishhook production occurring at these sites, with other midden
studies indicating it was a species commonly chosen for this
purpose (Attenbrow 2002:119).
Although the majority of the archaeological material
identified was shell, some stone artefacts were also noted,
including numerous chert flakes, a basalt pebble tool, and a
shaped and utilised Xanthorroea flower stalk (possibly part
of a fishing spear), together with several Anadara shells
displaying use-wear. This suggests that tool manufacture
and maintenance was occurring at many of the sites, and
may be indicative of ‘base camp’ as opposed to ‘dinnertime’ (Meehan 1982) site usage in some areas. Most of the
art observed during the survey was in the form of charcoal
drawings, some of which were hard to distinguish with any
degree of certainty due to erosion. One exception to this
was a large red ochre fish that was painted on the wall of a
small rockshelter (with the piece of ochre rock used for the
drawing still in situ). Axe grinding grooves were also noted,
and although not abundant in the area tended to occur close
to the shoreline, within close proximity to some of the sites.
This low number may be a reflection of the limited
distribution of sandstone exhibiting a grain size conducive
to grinding within the study area.
Discussion
The Cowan Creek area contains a rich and varied
collection of Aboriginal sites, displaying a great deal of
information relating to the resource utilisation and
subsistence strategies adopted by the Aboriginal people who
have inhabited the area The density of sites observed,
together with the apparent depth of many midden deposits
recorded during the survey along Coal and Candle Creek,
will greatly enhance the level of information on the site
register, with the results of this survey indicating that this
was a location strongly favoured for occupation in the past.
This may be due to the combination of abundant and
accessible resources, together with the prevalence of
suitable inhabitable rock shelters within the area. Whether
or not this was a region that was occupied intensively over
a short period of time, or sporadically over a longer time
period is open to conjecture, and is difficult to determine
without excavation. The material accumulated, together
with the size and depth displayed by many of the deposits,
may be interpreted as an indication that occupation patterns
in this area varied. The evidence is supportive of both shortterm visits, during which perhaps only a single meal of
shellfish was consumed, to more long-term campsites that
displayed evidence of a variety of activities taking place
during periods of occupation.
The information available from ethnohistorical accounts
supports the archaeological evidence for Aboriginal
occupation in this region. During Governor Phillips
exploration of Broken Bay and its surrounding waterways in
March 1788, both written accounts and painted scenes of
the area, indicate a large Aboriginal presence during this
time. By June 1789 however, Phillips accounts also attest to
the decimation wrought upon the local Aboriginal people in
this region due to the impact of smallpox, and resulting in
few sightings of ‘live natives’ during his second excursion
(Champion 1990). Although this may provide an indication
of the last time that the area was occupied by a relatively
large number of Aborigines, it has been suggested that
people were still living in a traditional manner in the
vicinity up until the early 1900s (Jacobs 2003:59; Read
2000:24-25). The archaeological material still visible upon
the surface of the middens today could therefore be
occupational debris deposited anywhere between late
eighteenth to early twentieth century. Although some of the
material was also indicative of European visitation, and
included glass and plastic bottles, and other forms of litter,
the shellfish remains evident throughout many of the
deposits are consistent with Aboriginal usage of sites within
this area extending to the uppermost layers.
The middens indicate that the favoured shellfish species
for collection include rock oyster and the hairy mussel,
which must it seems, have been abundant in the past based
on the quantities present in the deposits. Although rock
oysters are still prolific, and are obvious in their domination
of the rocks along the high water line, the hairy mussel
appears to be more cryptic. Isolated individuals were
located growing interspersed between the densely packed
oyster shells, but diving indicated that they were also
present at the sub-littoral level, where they formed clumps
on the soft sediment. With this in mind, and if the ‘timedistance’ factor often associated with gathering strategies is
invoked (Bailey, 1975), together with the often large
quantities that are evident within many of the deposits, it
seems likely that the ‘clumps’ would have been favoured for
collection by Aboriginal people camping in the area. A
suitable collection strategy therefore may have been to dive
into the water, either from canoes or the shoreline (which is
extremely steep and rocky in many places), in order to
gather this mussel. Alternatively at low tide and during
periods of low visibility a similar strategy may have been
21
adopted to that utilised by the Anbarra women in their
collection of oysters attached to mangrove roots in the
Northern Territory. This involves the women wading into
the water and using their feet to ‘feel’ around in the
sediment in order to locate clusters of shellfish prior to
diving down to retrieve them (see Meehan 1982:100).
In the past there has been a great deal of debate
concerning the observation of large amounts of mussel
remains within archaeological deposits (Bowdler 1976;
Sullivan 1987; Mackay and White 1987). Although the
focus has generally been on the common mussel (Mytilus
edulis), and the associated social implications of its
presence and the introduction and adoption of hook and line
fishing, the question remains as to why Trichomya has not
received similar attention? Although Mytilus was not
observed within any of the middens in this particular area,
it is reasonable, due to the similar environments occupied
by these two species, to evoke a similar argument to that
proposed by Bowdler and Sullivan, in that the increases of
mussel remains observed within these midden deposits may
be indicative of the introduction of fishing with hooks and
lines. The abundance of Trichomya observed within the
deposits may indicate a change in collection strategies as a
consequence of the introduction, and increasing popularity
of hook and line fishing. The presence of large pieces of the
Turban shell (Turbo torquatus), found on the surface of two
deposits (one at Coal and Candle Creek, the other at Smiths
Creek) may be supportive of this hypothesis, and attest to
the fact that the manufacture of shellfish hooks was taking
place in this area. Ethnohistorical accounts also provide
supportive evidence of fish-hook production occurring in
the area, with William Bradley clearly documenting an
Aboriginal woman demonstrating how the hooks were
fashioned out of ‘pearl’ shell, during his visit to Broken Bay
in the late 1700s (Bradley 1786-92: cited in Attenbrow
2002:118). Alternatively, and in association with the
‘fishing hypothesis’, Trichomya may also have been
collected and used as bait, a suggestion that was initially
proposed by Sullivan (1987), in reference to the late
occurrence and abundance of Mytilus within other deposits.
Although the small number of Turban shell pieces observed
is not suggestive of large-scale hook manufacture taking
place, other evidence, including the shaped Xanthorrhoea
flower stem (possibly part of a fishing spear) and the ochre
fish drawing, does imply that fish were an important
resource for the Aboriginal people occupying the area. In
order to clarify the role played by fish and shellfish for the
people that occupied this area, it is important that a more
detailed examination of the midden material be conducted
in the future, however this may only be achieved through
excavation. Perhaps, in order to minimise the impact of such
an investigation, excavations could be limited to the
removal of column samples, which will enable a
determination of the presence/abundance of fish remains,
and shellfish-hooks to be made.
Although only a quarter of the rockshelters examined
during the survey showed evidence of art, mostly charcoal
drawings, this was consistent with other studies in the
region (McDonald 1992; Vinnicombe 1980). The images
depicted were often fragmented and eroded, which made
them difficult to interpret. However, andropomorphic
figures, marine animals and weapons were images
decipherable at some shelters. The most easily identifiable
image was that of a large fish, measuring 33 cm in length
22
and drawn in red ochre on the rock wall of a small
rockshelter situated along Coal and Candle Ck. Although
the image was clear it was not possible to identify to the
species level, and the purpose behind the drawing is open to
conjecture. This, and the other faunal images that are
depicted, may have totemic implications, or more simply
were created in order to demonstrate the suitability of the
shelter and its proximity to particular resources, as a guide
for future occupation.
Conclusion
This survey has produced a dramatic increase in the
number of sites along Coal and Candle Creek from the
previously recorded 0.5 to 8.5 sites per linear kilometre of
the creek. Other research on site density along the east coast
of Australia would suggest that this level of density is
unusually high. For example, slightly to the north of this
study area in the Gosford-Wyong region, Pat Vinnicombe
surveyed for sites along a 2 km transect of coastline, and
determined a site density of 0.5 sites/linear km (1980). This
was also in keeping with Stockton’s work on the Central
Coast (1972, 1977) which indicated a site density in this
area of 0.51 sites/linear km. Sullivan (1976) calculated 1.0
sites/linear km associated with sandstone environments
during work in the South coast, and 0.85 sites/linear km on
the Victorian coast (Coutts et al cited in Vinnicombe 1980).
There are many plausible explanations to account for this
apparent variation. Firstly, the high site density observed in
this area may simply be due to a pattern of intense
Aboriginal occupation and utilisation of the region over a
period of time. It is perhaps more plausible to suggest that
the lower densities observed in other areas is a reflection of
the loss of sites due to the impacts of urbanisation and
development. This study area, which, due to its inhospitable
and rugged nature, appears to have experienced very little
such impact, and could therefore provide a more accurate
depiction of the site density that existed in other areas prior
to the expanse of European settlement. Site visibility and
location in this area may have been enhanced due to the
nature of the terrain, which is open forested with rocky
outcrops extending down to the water line, combined with
the preponderance of sites within close proximity to the
water. The survey technique adopted here could also be
beneficial for other researchers to adopt, enabling the
coverage and examination of large areas that would
otherwise be inaccessible, and result in the maximum site
identification suited to this type of environment. The high
density of sites identified in this study may be attributed to
one or more of these factors, however it is undeniable that
the absence of development and cultivation within the
immediate area is a significant factor.
The methods adopted for this survey achieved the initial
aims of the project, in that a large number of sites were
located and examined within the study area, over a
relatively short period of time. The results of the survey
indicate that this area contains a wealth of information
reflecting a variety of Aboriginal social and economic
activity. The results also suggest that the majority of both
registered and newly recorded sites within the Cowan Creek
area appear to display a disturbing level of degradation. This
is despite the protection afforded by their inclusion within
the bounds of a National Park. Although open midden sites
by their very location are more vulnerable, and therefore
generally display the greatest level of degeneration caused
by both natural and anthropogenic means, there are also
many sites within rockshelters that appear to be suffering
damage due to the impacts of fossickers and ‘day-trippers’.
Although it may be impossible (given the magnitude of the
area) for management strategies to be implemented for all
of the Aboriginal sites, it is important that the significance
of the sites within Cowan Creek and other areas around
Sydney, be thoroughly assessed and recorded, before this
valuable source of information becomes irretrievably lost.
Acknowledgements
This project was conducted through and supported by,
Macquarie University, Sydney. We would like to
acknowledge the Metropolitan Local Aboriginal Land
Council and Guringgai Tribal Link for their support of this
work. We would also like to thank the staff within the
Aboriginal Heritage section of the NSW National Parks and
Wildlife Service for their assistance, Val Attenbrow at the
Australian Museum for her invaluable advice and support,
and Patricia Bourke for her useful comments on the final
draft of this paper.
References
Attenbrow, V. 1991 Port Jackson archaeological project: a study of
the prehistory of the Port Jackson catchment, New South
Wales. Stage I – site recording and site assessment.
Australian Aboriginal Studies 2:40-55.
Attenbrow, V. 2002 Sydney’s Aboriginal Past: Investigating the
Archaeological and Historical Records. Chapters: 2, 3, 10.
Sydney: University of New South Wales Press Ltd.
Bailey, G. N. 1975 The role of molluscs in coastal economies: The
results of midden analysis in Australia. Journal of
Archaeological Science 2:45-62.
Bowdler, S. 1976 Hook, line and dilly bag: An interpretation of an
Australian coastal shell midden. Mankind 10(4):248-258.
Bowdler, S. 1983 Sieving seashells: Midden analysis in Australian
archaeology. In G. Connah (ed.) Australian Field
Archaeology. A Guide to Techniques, pp. 135-144. Canberra:
Australian Institute of Aboriginal Studies.
Bowdler, S. 1984 Archaeological significance as a mutable
quality. In S. Sullivan and S. Bowdler (eds) Site Surveys and
Significance Assessment in Australian Archaeology, pp. 1-9.
Canberra: Australian National University.
Byrne, D. 1984 A survey strategy for a coastal forest. In S.
Sullivan and S. Bowdler (eds) Site surveys and Significance
Assessment in Australian Archaeology, pp. 61-70. Canberra:
Australian National University.
Champion, G. S. 1990 Journey to Broken Bay and the Hawkesbury
River – 6th June to 16th June 1789. Monogram.
Fairly, A. 1972 The Beaten Track: A Guide to the Bushland around
Sydney. Hong Kong: Dai Nippon Printing Co., (Int’l) Ltd.
Jacobs, I. 2003 A History of the Aboriginal Clans of Sydney’s
Northern Beaches. Sydney: Northside Printing.
Mackay, R. and White, J.P. 1987 Musselling in on the New South
Wales coast. Archaeology in Oceania 22(3):107-111.
McDonald, J. 1992 The Great Mackerel Rockshelter excavation:
Women in the archaeological record? Australian
Archaeology 35:32-50.
Meehan, B. M. 1982 Shell Bed to Shell Midden. Canberra:
Australian Institute of Aboriginal Studies.
Read, P. 2000 Belonging: Australians, Place and Aboriginal
Ownership. Cambridge University Press.
Rowland, M. J. 1994 Size isn’t everything. Shells in mounds,
middens and natural deposits. Australian Archaeology
39:118-124.
Stockton, E. D. 1972 A Central Coast Survey. Canberra Australian
Institute of Aboriginal Studies Newsletter 3(5):20-24.
Stockton, E. D. 1977 Middens of the Central coast. Australian
Archaeology 7:20-31
Sullivan, M. E. 1976 Archaeological occupation site locations on
the south coast of New South Wales. Archaeology and
Physical Anthropology in Oceania 11(1):56-69.
Sullivan, M. E. 1987 The recent prehistoric exploitation of edible
mussel in Aboriginal shell middens in southern New South
Wales. Archaeology in Oceania 22:81-96.
Vinnicombe, P. 1980 Predilection and Prediction: A Study of
Aboriginal Sites in the Gosford-Wyong Region. Report to
the National Parks and Wildlife Service of New South Wales,
Sydney.
Vinnicombe, P. 1984 Single sites or site complexes? A case study
from north of the Hawkesbury, New South Wales. In S.
Sullivan and S. Bowdler (eds) Site Surveys and Significance
Assessment in Australian Archaeology, pp. 107-117.
Canberra: Australian National University.
THE ARCHAEOLOGY OF RANK
by
Paul K. Wason
Cambridge University Press
2004
ISBN 0-5216-1200-4
RRP £18.99
(paperback)
23
Book Reviews
Aboriginal occupation at Hawker Lagoon, southern Flinders Ranges,
South Australia
Keryn Walshe
Abstract
Two Indigenous archaeology field schools were
conducted by this author and Pauline Coulthard, an
Adnyamathanha elder, during 2001 and 2002 for a total of
four weeks. The schools were held at Hawker Lagoon, in the
southern Flinders Ranges with participation of students
from the Department of Archaeology, Flinders University.
The archaeology program continued earlier work
undertaken by Ron Lampert in the 1980’s at the same site.
Excavation revealed a disparity between earlier
stratigraphic patterns and dating outcomes. The surface
material is subject to significant environmental disturbance.
Three surface hearths returned dates ranging from about
1500 to 550 years BP for associated charcoal. The lagoon is
discussed within the broader context of occupation, trade
and response to the LGM rather than within the narrow
context of disturbed archaeological assemblages.
Introduction
During the 1970’s-1980’s archaeologist Ron Lampert
was particularly interested in ‘Kartan’ sites (assemblages
dominated by large pebble or block core tools) on Kangaroo
Island and the mainland of South Australia. Whilst
investigating the Flinders Ranges region of South Australia,
Lampert was introduced to Hawker Lagoon, where in
association with geologist Phil Hughes, he carried out
archaeological investigations during the 1980’s. Hawker
Lagoon provided a wealth of archaeological material
eroding out onto deflated, exposed surfaces below remnant
dunes and a lunette.
The idea of a distinctive ‘Kartan’ culture has not been
widely accepted amongst archaeologists (Mulvaney and
Kamminga 1999) and is currently absent from popular
discourse. However, the archaeology present at Hawker
Lagoon certainly confirmed Indigenous occupation in the
region and the carbon samples obtained during excavation
indicated Pleistocene occupation (Lampert and Hughes
1988). The interpretative discussion presented by Lampert
and Hughes (1988) positioned Hawker Lagoon as a
Pleistocene outpost, a position later strengthened with the
emergence of the ‘refugia’ theory for Indigenous response
during heightened glaciation (Veth 1989, 1993). Hawker
Lagoon was pictured as a refuge for Indigenous people who
were forced to retreat from increasingly arid locales during
the Last Glacial Maximum (hereon referred to as LGM).
The lagoon, along with other Pleistocene outposts across
the continent offered less risk than more highly exposed,
resource depleted areas at the peak of glaciation. The idea
of the Flinders Ranges generally providing a refuge away
from the depleted, exposed sandy low lands during the
LGM was supported by later investigative work along the
Lower Cooper Creek (Veth et al. 1995).
School of Humanities, Flinders University, Adelaide, SA 5001, Australia.
Email: [email protected].
24
Between 1989 and 2000, no further comprehensive
archaeological or geological investigations took place at the
lagoon, other than tracking the movement of stone tools
across vastly eroded surfaces (Cameron et al. 1990). In
2001 and 2002 archaeological investigations at Hawker
Lagoon were revived in association with Indigenous
archaeology field schools conducted by the Department of
Archaeology, Flinders University. This paper discusses the
results of the field schools run by this author in conjunction
with Pauline Coulthard, Adnyamathanha elder and local
resident of Hawker. Pauline’s parents Pearl and John
McKenzie worked closely with Ron Lampert in the 1980’s
and 1990’s. Pauline’s contribution to the field schools was
of similar immense value.
Hawker Lagoon
Hawker Lagoon sits about 6 km west of the township of
Hawker (Fig. 1) in the Wilson Valley and is accessed over a
saddle in the steep Yourambulla Range. The Yourambulla
and Yappala Ranges run along the eastern and western sides
of the Valley, respectively, and reach a height of about 635
m at Mt Elm. The Wilson Valley is a narrow, v-shaped
valley measuring less than 2 km between the ranges and
approximately 3 km from north to south.
Pastoral activity commenced over 100 years ago in the
Wilson Valley and continues so today. Hawker Lagoon
generally retains some water in associated channels
throughout the year but fills completely only intermittently.
The degree of impact on flood regimes from the advent of
pastoralism can only be assumed and the lack of consistent,
longitudinal data since pastoralism prevents any valid
prediction for hydrological cycles. The current owner Mr
Trevor Jarvis, provides some anecdotal information in that
the lagoon appears to fill more frequently than it did 50
years ago but the drainage rate appears to be similar.
When the lagoon is dry, its exposed bed reveals deep
fissures in the hard, alluvial surface. Much of the surface is
vegetated with cane grasses and rushes. Scree accumulates
upslope on both the eastern and western margins where a
band of Callitris pines hug the upper foothills. The valley
floor is open to the south where perennially grass covered,
boggy channels trap the unwary. To the north of the lagoon
Figure 1 Hawker in location to Adelaide, Port Augusta.
Keryn Walshe
are deflated red-orange sand dunes, eroded hard-pan
surfaces and dense mallee cover until the valley converges
into a narrow gorge that plunges down to Yappala Waters.
This narrow gorge ensures that winds from the north are
funnelled out into the valley with immense force. This
funnelling effect has resulted in source bordering dunes
running an unconventional north-south.
Lampert and Hughes (1987, 1988) identified a lunette
feature on the southern side of the lagoon and associated
eroded dune surfaces along the western margin. It is here
that the archaeology is most visible and needless to say
most disturbed. Site preservation is extremely poor once
material is exposed due to the combined effects of wind and
water in this particularly narrow and steep sided V-shaped
valley. The lack of vegetation cover across the lunette
enables wind and water action to undermine the lunette
feature. Stone material is quickly transported by wind and
water (Cameron et al. 1990) and faunal material is entirely
absent barring the occasional shell fragment. Faunal debris
derived from ancient campsites has a limited chance of
survival once exposed to the elements. Cypress-pine
(Callitris glaucophylla), Black Oak and various eucalyptus
(mallee) trees cover the slopes and foothills of the Ranges,
but the valley floor is today essentially treeless. According
to Mincham (1980:143) “a dense growth of native pines,
mallee, black oak and brush covered much of the floor and
sides of the (Wilson) Valley… the run-off from the ranges,
actually made a swamp near the northern end.” Mincham
also mentions the difficulty faced by the early settlers to the
valley in around 1880 in clearing the land, except for the
likes of James and Lucy Ward who ‘…had the family to
cope with that…’, considering their fifteen children. It is
today difficult to imagine such a densely forested valley
with discrete water sources rather than unstopped flood runoff.
Previous archaeological investigations
Lampert and Hughes carried out a series of both
excavations and surface recordings at the southern end of
Hawker Lagoon (Fig. 2). The main excavation trench (HL1)
was:
“…opened up in the richest part of the concentration
of artefacts in the dunefield, just beyond the western
end of the lunette, at the southern end of the lagoon,
on the deflated surface along the western margin and
on the lunette” (Lampert and Hughes 1988:151).
This revealed four superimposed layers of sand
including a loose orange (unit IA), acompact grey brown
(unit IB), then ‘rock hard’ red sand (unit IIB) and finally
mottled yellow and grey clayey sand (unit III). Upper unit
IA, yielded microlithic material and unit IB was found to be
sterile. Unit IIb with its rock hard red sand produced
significant numbers of “…core tools, cores and large
flakes...” and Unit III was sterile (Lampert and Hughes
1988). Needless to say, Unit IIB became the focus.
Unit IIB, associated with hard red sand, carried a mix of
apparently in situ and loose artefacts. The challenge lay in
the incredibly resistant nature of the IIB stratum to trench
excavation. Three seasons were to pass before sufficient
material could be gathered by Lampert and Hughes (1988)
from this unit. Difficulty was also encountered in
distinguishing between clearly in situ material and loose
material, preventing clear recognition of contaminated
material. Thus in order to increase the sample of in situ
material associated with the IIB stratum, Lampert and
Hughes retrieved artefacts from a nearby gully where the
same unit appeared to be exposed. The sampled material
was added to the IIB results. A carbon sample believed to be
associated with a fireplace was retrieved from this same
unit IIB in the main trench and the result was an exciting
14,770 ± 270 BP (SUA:2131) (Lampert and Hughes 1988).
Two small trenches, HL30 and 32 were opened on the
lagoon bed by visiting researcher Richard V.S Wright to
investigate possible faunal remains in the swamp sediments
(Lampert and Hughes 1988). Wright found a similar
stratigraphic pattern for the upper units which yielded only
one artefact. Unlike the main trench however, Wright did
not locate any cultural evidence in association with the
equivalent of unit IIB.
Lampert and Hughes (1988) opened up another trench,
HL40, on the eastern side of the lunette, southwest of
Wright’s smaller trench and approximately 700m east of
HL1. Three strata matching Units IB, IIB and III were
identified in trench HL40. In order to test the consistency of
the stratigraphic pattern, Lampert and Hughes placed a line
of closely spaced auger holes between Wright’s HL32 and
their second trench HL40. Confirmation was claimed. They
then inspected section HL TT, exposed in a creek bank and
about 500 m south of HL1. Again stratigraphic consistency
was recorded and further charcoal samples retrieved for
dating. These yielded results ranging from ca. 14,000 BP to
8400 BP for units either side of the cultural horizon, unit
IIB. In addition, two controlled surface collections were
made in areas associated with two outflow channels
emerging from the lagoon. They reported that small tool
industries were clearly confined to these outflow channels,
unlike the widespread Kartan large tool industries.
In summarising the archaeological evidence at Hawker
Lagoon, Lampert and Hughes (1988:166) suggested two
broad phases of occupation- the earliest commencing about
15,000 years ago and characteristic of the Kartan and a later
phase beginning about 5000 years ago up until perhaps
quite recently. The oldest stratum was represented by the
‘rock hard’ red sand unit IIB with its in situ ‘Kartan’ tools
whilst the younger stratum, unit IB, consisted of loose
orange-red wind blown sand. The gap of 10,000 years
between these two significant strata was interpreted as both
an occupational and depositional gap in the site’s history.
Lampert and Hughes (1988:166) refer to palaeoclimatic
evidence for moister conditions during this ‘gap’ and
suggest that Indigenous people were able to “…spread
themselves more widely, occupying regions that had been
inhospitably arid”. This was supported by dates retrieved
from archaeological sites around Lake Frome, suggesting
that occupation commenced in the early Holocene. Lampert
and Hughes entertained the idea of ‘refugia’ as posited
earlier by others (cited in Lampert and Hughes 1987)
whereby the pattern of movement by Aboriginal people is
determined by water availability. Good rains allow people to
shift out beyond the confines of reliable, well-watered
sources. In the early 1980’s Adnyamathatha elder John
McKenzie introduced Ron Lampert to Eudlia Wagloona
waterhole on the south-eastern shore of Lake Frome.
According to Mr. McKenzie, this waterhole was used in
high rainfall seasons and allowed people to make brief
forays from the Ranges to the lakeshores.
25
Aboriginal occupation at Hawker Lagoon, southern Flinders Ranges, South Australia
By inspirational leaps Lampert and Hughes (1988)
linked Hawker Lagoon into the cultural fringes of the Lake
Eyre Basin. This mapping of the lagoon into a much wider
cultural geography has strongly influenced the most recent
investigations and interpretative context for human
occupation and land use at Hawker Lagoon.
Hawker Lagoon fieldwork 2001-2002
Three field seasons were held at Hawker Lagoon during
2001 and 2002. A multi-stepped approach was undertaken
whereby test pits, auger holes and trenches were placed in
different environmental zones along the Wilson Valley. In
all we investigated the far eastern side of the lagoon, the
Figure 2 Flinders University Archaeological Field School Investigation over Hawker Lagoon and Lunette.
26
Keryn Walshe
upper northern reaches of the lagoon and the southern flood
plain of the open valley floor with the primary focus
remaining on the lagoon basin and its immediate margins
including the lunette (Figs 3-5).
Nine test pits (TP1-9), six auger holes (L1-6) and eleven
excavations (EH1, HL02ED1-5, HL02EDH1) were under-
taken near the margins of or across the lagoon, on the
lunette and on the eastern sand dunes, as shown in Figure 3.
Three surface hearths identified about 500 m south of the
lagoon basin were excavated (HL02WH1-3), as shown in
Figure 5. Three surface hearths about 600 m north of the
lagoon basin were also excavated (HL02NH1-3) and the
Figure 3 Hawker Lagoon Lunette and Test Pits HL02EDH1, EH1, ED1-5.
27
ground surface beneath an artefact scatter about 250 m
northeast of the lagoon basin was excavated (HL02ND1) as
shown on Figure 4.
Thus a total of nine test pits, 6 auger holes and 18
excavations were carried out between 2001 and 2002. Only
one excavation could be considered a deep trenchHL02WH3 which extended to 2 m. The remaining
excavations reached an average depth of 0.25 m due to the
resistant nature of the unit and the lack of cultural material
below the immediate surface.
Excavation results
The surface unit on which all excavations commenced
complied with the description for stratum IIB as identified
Figure 4 Hawker Lagoon (Northern Area) HL02NH1-4, HL02ND1.
28
Keryn Walshe
by Lampert and Hughes (1988). In no case were the upper
units, IA and B and IIA present, having been previously
washed off and/or blown out.
Nine test pits were placed along a bearing of 140E from
the lagoon basin to the top of the western edge of the
lunette. These were all found to be sterile. The stratigraphy
for the lagoon basin was uniformly alluvial with occasional
beach sands below the alluvium. Six of these test pits were
placed on the lunette and consistently revealed only red
compacted sand. No other strata were located despite
reaching depths of between 1-2 m. A carbonate horizon was
reached in two pits, matching the description given by
Lampert and Hughes (1988) for a calcium band
interrupting the IIB stratum.
Figure 5 Hawker Lagoon (Western Margin-Southern Area) HL02WH1-3 and HL02SH1.
29
The test pits revealed a complete lack of stratigraphy
beyond the IIB unit, which in all cases commenced at
ground surface and continued to at least 1.5 m below
surface. The compaction was highly resistant, analogous to
cutting through bricks with a butter knife. The uniform
nature of the single stratum was curious.
The main trench was established near HL1, the main
trench opened by Lampert and Hughes in the mid-1980’s.
Considering the outcome of the test pits across the lunette,
we were anxious to confirm the existence of the recorded
stratigraphy. However, after digging through the loose
orange sand stratum described by Lampert and Hughes as
IA, then the beach sands for 10 cm, mottled yellow/grey
sands for 30 cm, we encountered the incredibly hard,
compact red sand stratum IIB. No further change in strata
was encountered and after resorting to a crow bar to break
through this unit to a depth of 1.5 m below surface, a few
unmodified river pebbles were located at the base of the
trench. These were presumably associated with early stream
runoff into the valley floor. Scree continues today to roll
down from the western foothills and into the valley and the
lower most pebbles were explained similarly. The
discrepancy between the finds made by Lampert and
Hughes and recent fieldwork demanded further excavation.
One other trench was opened behind our main trench, closer
to the exposed strata in the gully noted by Lampert and
Hughes (1988). The same result was achieved with a single
unit of hard, compact red sand intersected by a band of
calcrete. As with the first trench, no artefacts were
encountered below the upper 20 cm of this trench. Artefacts
recovered from the upper sediments ranged from small to
larger tools. Certainly no tools were found in the IIB
stratum below the immediate surface.
HL02ED1-5 excavations were placed at the eastern side
of the lagoon directly onto loose orange sand that matched
the description for unit IA (Lampert and Hughes 1988).
Low density stone tool scatters of stone tools were found
clustered over discrete surface deflations. This provided
another focus for open site excavation. Trenches varied
from 1-2 m in depth and were easily dug but revealed only
one uniform, archaeologically sterile stratum.
After completing 9 test pits to an average depth of 1.5 m
and 5 excavations to 0.3 m and one to 2 m, a very clear and
consistent pattern emerged. Stone tools were clearly
isolated on the surface without any subsurface material or
other cultural horizons whatsoever. It was apparent that
deep excavations were highly unlikely to yield suitable
material. Considering the alarming degree of bio-turbation
on site integrity it seemed only logical to view the
artefactual material as a horizontal representation of
different occupation phases over time. That is, occupation
sites are unable to become ‘layered’ over time due to the far
greater rate of deflation over rates of deposition in this
exposed wind tunnel known as the Wilson Valley.
Occupation over many thousands of years is strewn across a
uniform surface, without any obvious chronological
indicator so that a site 500 years in age sits alongside one
5000 years old. Thus we turned our attention to the surface
hearths.
Five hearths were selected for excavation purely to gain
charcoal samples for dating. EH1 located on the lunette,
HL02WH1-2 located about 500 m south of the lagoon basin
and on the western margin of the valley, HL02EDH1 on the
dune east of the lagoon and HL02NH3 located on a remnant
30
dune about 1 km northeast of the lagoon basin (Figs 3-5).
These all proved to be shallow deposits ranging from 0.2 m
to 0.3 m below the surface situated burnt cobbles and
varying numbers of stone tools. Three of the five samples
collected from surface hearths were successfully dated, all
returning late Holocene dates. The results are as follows:
CS2332: 550 ± 110 yr BP
CS2331: 1500 ± 120 yr BP
CS2333: 1230 ± 120 yr BP
Two of these hearths were associated with unit IIB, the
hard red compact sand stratum dated by Lampert and
Hughes (1988) to about 15,000 years old. The third hearth
was located south of the lagoon on the open valley floor but
also on the same IIB unit.
Stone tools were observed, having fallen into deep
fissures on the lagoon basin surface. This prompted a series
of auger holes to investigate the probability of subsurface
material and the basin sediments. The artefacts had
presumably either washed in, or were left by Indigenous
people when camping on the basin surface during dry lake
times. The high, dense stands of cane grass across the
lagoon surface currently provide a welcome respite from the
interminable wind that bursts down this narrow valley. A
scatter of historic material including culturally modified
fragments of glass and ceramic suggest very recent use of
the Lagoon as an occupation surface.
A series of auger holes across the lagoon bed
consistently revealed a uniform alluvial deposit with
occasional beach sands and charcoal and no stone tools or
other cultural material.
Pedestrian surveys
As well as excavation work, Indigenous surface sites
and post-contact assemblages identified within the valley
and on the adjacent ranges were recorded. A number of
surface scatters of stone tools; a silcrete outcrop with
evidence for quarrying; rock art sites; artefact scatters on
both the valley floor and ridges and ruins associated with
early pastoral development were recorded.
All surface scatters of stone tools were associated with
the IIB surface and appeared rarely in situ, no matter how
loosely that term was applied. Our observations of the area
during times of high rain fall and subsequent rapid run off
strongly suggests that artefacts are quickly dislodged from
the surface and transported by fluvial activity to be
deposited later on the central valley floor. From here they
travel southwards at an unknown rate during times of
heightened water runoff.
Areas located east, north and south of the lagoon to a
distance of one km from the centre of the lagoon were
targeted. The recordings of surface material revealed a
similar density of small tools (predominantly flakes) over
larger tools (previously known as Kartan) across all
sampled surfaces. In all, approximately one square km of
surface material was recorded and some thousands of
artefacts. Consistently, smaller tools such as scrapers,
adzes, blades, points and cores were evident on the suface
alongside larger tools such as hammerstones, horsehoof
cores, knifes, anvils and grinding stones. Eroded surfaces
on the eastern and south-eastern foothills of the
Yourambulla Ranges were also investigated, particularly
along gullies. These revealed a similar distribution of tool
types and raw materials.
As regards raw material, a definite clear abundance of
Keryn Walshe
quartzite and silcrete over other materials was noted. In
particular grey/green silcrete dominates scatters to the south
of the lagoon and given the closer proximity of these
scatters to the silcrete outcrop this is hardly surprising. In
lesser numbers were quartz, flint and cherts.
Considering the relatively small, confined nature of the
Wilson Valley, it was possible to survey on foot all margins
of the lagoon up to break of slope on the east and west sides,
to Yappala Waters on the north side and to Youraumbulla
Cave on the south side. The ridges of the Youraumbulla and
Yappala Ranges were also inspected.
This survey revealed a painting site at Yappala Waters
where the gorge north of the lagoon meets a tributary of
Hookina Creek; an artefact scatter on the saddle above the
valley on the Yappala Ranges; a silcrete outcrop with
evidence of quarrying and tool making on site; numerous
artefact scatters increase in density when associated with
the harder, red sand surfaces; modified glass and ceramic
around ruins; scarred trees; stone cairns and engravings on
the western face of Yappala Ranges, along minor creek
tributaries of Hookina Creek.
Post-contact material
One of the earliest settler structures on the Jervis
property, a modest wattle and daub two roomed hut, was
found to have a dense scatter of glass fragments in front of
the door facing the lagoon. Similarly, a standing chimney
and fireplace indicating a collapsed hut to the south of the
lagoon and towards the adjoining property, ‘The Oaks’
presented a glass scatter. Between these two features, the
former schoolhouse today indicated only by a plaque, also
presented some glass fragments as well as stone tools and a
series of small hearths.
Discussion of the field results
The archaeological investigations over 2001-2002
recorded a silcrete quarry close to the occupation surfaces,
a painting site at Yappala Waters, occupational use of the
lagoon surface and other surface sites previously unknown.
Dates obtained from three surface hearths were all of
relatively recent age, being about 1500 to 550 years old.
Removing the ‘Kartan’ as a key discussion point, we can
sensibly confirm that the archaeology demonstrates
continuous use of the Wilson Valley by Indigenous people
over a considerable amount of time. Exactly what this time
stretch is remains at this point unclear considering the
discrepancy between the stratigraphy encountered by
Lampert and Hughes in the 1980’s and the more recent
investigations. This has undermined our confidence in the
status of the hearth feature identified by Lampert and
Hughes (1988) from which the 15,000 year old charcoal
sample was obtained at a depth of some 1.3 m below the
surface. At no stage did we encounter any subsurface
material below the upper ‘A’ horizon of some 20 cm. All
archaeological material was exposed on a hard pan surface.
The lunette and other low, deflated dunes on all sides of the
lagoon, the lagoon bed itself and the higher, more yellow
dunes on the eastern side did not contain any material below
the immediate upper surface. Despite completing 32
subsurface investigations into a broad and representative
range of environmental units with or without surface
artefacts displayed, not a single tool or other object was
located more than 15 cm below the surface.
Obviously the archaeology here cannot singularly
determine the chronological or dynamic context of
occupation and landscape use. Identifying Indigenous
response to significant environmental and climatic changes
over vast periods of time is similarly confounding when
remaining only with the archaeological assemblages and
few isolated, discrete dates for occupation. It is entirely
sensible to return to the broader cultural geography so
thoughtfully considered by Lampert and Hughes (1988) and
the pertinent regional questions raised by Veth et al. (1995).
The palaeoenvironment of Hawker Lagoon
Evidence from Lake Eyre suggests that fluvial activity
heightened 55,000-40,000 years ago and again at 26,00022,000 years ago (Nanson et al. (1996) in Gell and Bickford
(1996)). This fluvial activity resulted in the probable
megalake resulting from the coalescence of the present-day
Lakes Eyre and Frome. Such high lake levels is argued by
Nanson et al. (1996) to be caused by temperatures up to 8 C
cooler than present leading to a combination of monsoonal
rains and reduced evaporation. There is now valid
archaeological evidence for occupation at points along the
entire Australian coastline and on major river systems by at
least 30,000 years ago (Mulvaney and Kamminga 1999,
Flood 1997). We also have early occupation dates from the
interior, for example at Puritjarra, Central Australia (Smith
1989) and closer to this study area, the JSN site in the
Strzlecki Desert during the late Pleistocene (Smith et al.
1991).
Obviously then, Indigenous people had explored much
of the continent and embarked on long-term occupation at
selected locations prior to the Last Glacial Maximum.
Indeed the combined data from palaeoenvironmental and
archaeological research implies that exploration of the
continent was taking place during heightened fluvial
activity at about 50,000-40,000 years ago, obviously prior
to the LGM. An obvious corridor of movement in eastern
Australia at this time is along the drainage systems of the
Diamatina River and Cooper Creek, as postulated by
Tindale (1974). This corridor connects the key resource rich
zones such as the Darling River system and Lake Eyre
Basin. If well-watered resources were also available in the
Flinders Ranges around this same time, it would be highly
likely that occupation in the Ranges, including Hawker
Lagoon commenced well before the LGM and continued
during the LGM as suggested by Veth et al. (1995).
Evidence has been presented by Williams et al. (2001)
for a late Pleistocene wetland at Brachina, 65 km northwest
of Hawker in the Flinders Ranges. They provide strong
geomorphological evidence for some places being wetter
during the LGM than today. This resulted in features such as
the Brachina wetland to exist at the height of the LGM, in
similar fashion to Lake George near Canberra. Williams et
al. (2001:130) attribute the existence of a wetland at
Brachina, where there are no wetlands today, to the
divergent responses of various lakes and rivers during the
LGM to local effects on water balance. Clearly, regional
climatic differences appear to have operated during the
LGM on both a micro- and a macro-level. This is also
supported more broadly by the evidence from Lakes Eyre
and Frome (Gell and Bickford 1996).
If Brachina wetland was able to exist during the LGM
due to an increase in local fluvial activity, perhaps Hawker
Lagoon with its similar geological context experienced a
similar microclimate during the LGM. Hawker Lagoon may
31
in fact have been a much larger expanse of water well before
(55,000-40,000 years ago) and during the LGM than today.
Indigenous people following the riverine corridors in north
eastern Australia and down into the Lake Eyre Basin may
have encountered the equivalent of a vast inland sea. The
Flinders Ranges, standing some 700 m above sea level
offered not just dry land but also high quality stone
materials and high quality ochre (Jones 1984; McBryde
1987). For those people continuing their journey southward,
Brachina Wetland and Hawker Lagoon surely offered water,
food, shelter and raw materials for stone and wood tool
making.
Lake Eyre has been identified as the epicentre for the
exchange of goods and cultural elements (McBryde 1987,
2000). In fact and as regularly remarked upon, the entire
continent appears to have been inter-linked by far reaching,
complex exchange routes allowing the transfer of goods and
culture both sacred and mundane (McCarthy 1939; Jones
1984; Jones and McEntee 1996; McBryde 1984, 1987,
2000; McEntee 1991; Veth 1993; Mulvaney 2002). The
narcotic pituri formed the pivot around which much Lake
Eyre exchange was focussed and within the complexities of
exchange ochre from the Pukartu ochre mine, just north of
Brachina Gorge in the Flinders Ranges was renowned
(Jones 1984). Sandstone grinding slabs, hatchet heads,
chisels and knives, spears and other wooden implements
also entered the Lake Eyre and subsidiary exchange routes.
Such items were drawn from prized and usually guarded
potent sources (McBryde 1984, 1987, 2000; Jones 1984;
Jones and McEntee 1996).
“The Flinders Ranges project north into the arid
grasslands and sandy deserts of Central Australia.
Their complex geology has generated surface
archaeological evidence of at least three important
Aboriginal cultural practises throughout this broader
region. Sandstone slabs and sandy shales
impregnated with silica were mined in the eastern
Ranges near Wertaloona over a period which may
have spanned several thousand years, providing
grindings stones of fine quality. These stones were
quarried and traded as far north as the central
Simpson Desert and as far west as the western Lake
Eyre Basin. They were used for wet-milling grass
seed, a staple in the Aboriginal diet of the entire
region. …Also traded extensively was the distinctive
dark pink ochre used for ceremonial purposes,
mined from the western side of the Ranges, just
north of Brachina Gorge. The Pukardu ochre mine
attracted annual expedition of up to 200 men whose
journey south from localities as distant as Birdsville,
Innamincka and Oodnadatta followed the ancestral
paths of Dreaming Emu and Dingo Ancestors” Jones
and McEntee (1996:163).
It is more than reasonable to consider Hawker Lagoon
with its supply of abundant fresh water as somehow linked
to the exchange business.
Along with water, the Wilson Valley also ensured a
range of animals and seeds associated with the forests of
Cypress-pine, Black Oak and Mallee gum and plains of
Spinifex grasses and sedges. The trees in turn were able to
supply hard woods for various implements and the close
proximity of good quality silcrete ensured stone tool
resources. Ceremony has been found to be closely
32
associated with exchange routes, often demonstrated by
specified open spaces, stone arrangements and/or art sites.
One of the most significant rock painting sites in the
southern Flinders Ranges, Yourambulla Caves complex is at
the end of Wilson Valley, within 3 km of Hawker Lagoon.
These caves, particularly the main chamber offer an ideal
medium for extensive paintings.
Discussion
The question now rotates around two key points- the
exchange of sacred and mundane objects and other cultural
facets as a continuous practise throughout the LGM and
into ethnographic times. Discontinuous presence of ochre in
deep archaeological deposits excavated from sites well
outside of the Lake Eyre and Flinders Ranges regions
suggests a break in the exchange mechanism during the
LGM (Veth 1993; Veth et al. 1995). How feasible is it to
suppose that around Lake Eyre and on its south eastern
corridors, the exchange mechanism was so finely tuned
prior to the onset of the LGM that it managed to continue
throughout and beyond the entire event? Intrinsic to this
question is the availability of sufficient micro ‘refuges’ in
the lowlands away from the Ranges and along the 200-300
km stretch to the eastern margins of Lake Eyre north and
south. The ability of ethnographically recorded populations
to travel through this same stretch of country with sufficient
regularity so that exchange routes were adequately
maintained, clearly relied on a cognitive map of every
rockhole, soak, spillage, good season creek and so on. As
importantly Indigenous travellers (or merchants perhaps) in
this ‘inhospitable’ region also cognitively mapped water
bearing trees and roots and were adept at building and
maintaining dams (Johnston 1941). The existence of
Indigenous people in the arid zones of South Australia and
their incomparable ability to travel across it with immense
rapidity certainly suggests that the alternating cycles and
intensity of fluvial patterns over the last 50,000 years
merely shifts locations on the cognitive water map rather
than eliminate them.
Pollen analysis for samples from Holocene strata
indicates a down turn in conditions more amenable to
humans and other mammals. By the mid-Holocene it
appears that in the Lake Eyre region and adjacent Flinders
Ranges drier conditions were constant with resultant
changes to vegetation, including loss of forests and the
spread of grasses and sedges (Gell and Bickford 1996;
Singh and Luly 1991). Therefore the Pleistocene, before
and during the LGM, rather than the Holocene appears at
this point far more favourable for humans to establish
strategies around long distance and complex exchange
routes, linking lowlands and ranges.
Hawker Lagoon may well have seen its florescence
30,000-15,000 years ago followed by a slow distancing
from the pulse of the exchange route until by a few
thousand years ago, the lagoon heard no more than the
occasional sound of distant traffic. The modest stone tool
scatter strewn on a saddle in the Yappala Range, just above
the southern-western end of the lagoon suggests a crossing
point between the valley and a small ephemeral creek on the
western side of the ranges that heads off toward Lake
Torrens. Hawker Lagoon becomes then no more than a stop
over when water was available.
Arguably, population numbers may have simply been
too low to maintain exchange routes through country made
Keryn Walshe
more challenging, requiring people to travel further
between resource areas. Certainly, it does require a critical
mass to maintain distribution centres and pass on down the
line various trade objects and bits of communication. This
is an area for far more informed discussion than the
archaeology at Hawker Lagoon can possibly hope to
provide at this point in time. Our collective theoretical
ignorance about the patterns and dynamics of Indigenous
hunter-gatherer convergence and mobility at 50,000, 30,000
or even 10,000 years ago is equally unhelpful. The
archaeology at Hawker Lagoon is a miniscule vision of
debris, severely eroded and sorted over time and yet so
tantalisingly linked to a vibrant landscape. Reaching into
that story of the past and merely attempting to brush up
against its profound complexity remains the challenge.
Acknowledgements
The field schools at Hawker Lagoon were made
enormously informative, entertaining and multi-layered by
the immense contribution from Pauline Coulthard and her
partner, Phil Stewart. Pauline in particular shared so much
with us, from her heart and her land. Academic teaching
provides one level, but the unique contribution made by
Indigenous women like Pauline and quiet Indigenous men
like Phil, cannot possibly be matched. All those who
participated benefited in a way that I am certain will stay
with them for many years to come. My utmost thanks to
Pauline and Phil and the Andymathanha ancestors of the
southern Flinders Ranges. Many thanks also to Mr Trevor
Jarvis who graciously allowed us unrestricted access to the
lagoon. Much gratitude to Matt Schlitz, Technical Officer,
Department of Archaeology, Flinders University for
mapping the archaeological works. Kindest thanks to those
who supported the field schools by providing unstinting
hard work, skills, intellect and good humour- Greg Carver
first and foremost, Pam Smith, Tiana Jenkins and all the
students of 2001-2002. Lastly, the Hawker community - my
particular thanks.
References
Cameron, D., White, P., Lampert, R. and Florek, S. 1990 Blowing
in the Wind. Site Destruction and Site Creation at Hawker
Lagoon, South Australia.
Flood, J. 1997 Rock Art of the Dreamtime: Images of Ancient
Australia. Sydney: Angus and Robertson.
Gell, P. A. and Bickford, S. 1996 Vegetation. In M. Davies, C. R.
Twidale and M. J. Tyler (eds) Natural History of the Flinders
Ranges, pp. 86-101. Adelaide: Royal Society of South
Australia Inc., DENR.
Hercus, L. A. 1992 A Nukunu Dictionary. Canberra: Australian
Institute of Aboriginal and Torress Strait Islander Studies.
Johnston, T. H. 1942 Some Aboriginal routes in the western
portion of South Australia. Royal Geographic Society of
Australia, Proceedings 42:33-65.
Jones, P. G. 1984 Red Ochre Expeditions: An ethnographic and
historical analysis of Aboriginal Trade in the Lake Eyre
Basin. A progress report – South Australian Museum.
Journal of the Anthropological Society of South Australia Vol
22(7):3-10, Vol 22(8):10-19.
Jones, P. G. and J. McEntee. 1996 Aboriginal People of the
Flinders Ranges. In M. Davies, C. R. Twidale and M. J. Tyler
(eds) Natural History of the Flinders Ranges pp. 159-173.
Adelaide: Royal Society of South Australia Inc., DENR.
Lampert, R. J. 1981 The Great Kartan Mystery. Terra Australis 5,
ANU, Canberra
Lampert, R. J. and Hughes, P. J. 1987 The Flinders Ranges: a
Pleistocene outpost in the arid zone? Records of the South
Australia Museum 20:29-33.
Lampert, R. J. and Hughes, P. J. 1988 Early Human Occupation of
the Flinders Ranges. Records of the South Australia Museum
22:139-168.
McBryde, I. 1984 Exchange in south-eastern Australia: an
ethnohistorical perspective. Aboriginal History 8:132-153.
McBryde, I. 1987 Goods from Another Country: Exchange
Networks and the People of Lake Eyre Basin. In J. Mulvaney
and J. P. White (eds) Australians to 1788 pp. 253-273.
Sydney: Fairfax, Syme & Assoc.
McBryde, I. 2000 Travellers in storied landscapes: a case study in
exchanges and heritage. Aboriginal History 24:152-174.
McCarthy, F. D. 1939 “Trade” in Aboriginal Australia and “Trade”
relationships with Torres Strait, New Guinea and Malaya.
Oceania 9:405-48, 10:80-104, 171-195.
McIntee, J. 1991 Lake Frome (South Australia) Aboriginal trails.
Transactions Royal Society South Australia 115(4):199-205.
Mincham, H. 1980 The Hub of the Flinders. The Story of the
Hawker District embracing the towns of Cradock, Wilson,
Hookina and Wonoka. Hawker Centenary Committee,
Adelaide.
Mulvaney, J. 2002 ‘…these Aboriginal lines of travel’. Historic
Environment 16(2):4-11.
Mulvaney, J. and Kamminga, J. 1999 Prehistory of Australia.
Sydney: Allen & Unwin.
Nanson, G., Price, D., Page, K., Roberts, R., East, J., Nott, J.,
Callen, R. and Chen, X-Y. 1996 TL evidence of
palaeoclimate and flow regime changes in Australia, pre
stage 7 to the present. Extended Abstract, CLIMANZ IV.
Dept. Geology, ANU, Feb 26-28.
Smith, M. A., Williams, E. and Wasson, R. J. 1991 The
archaeology of the JSN Site: some implications for the
dynamics of human occupation in the Strzlecki Desert
during the late Pleistocene. Records of the South Australia
Museum 25(2):175-192.
Singh, G. and Luly, J. 1991 Changes in vegetation and seasonal
climate since the last full glacial at Lake Frome, South
Palaeogeography,
Palaeoclimatology,
Australia.
Palaoecology 84:75-86.
Tindale, N. B. 1974 Aboriginal Tribes of Australia. Their terrain,
environmental controls, distribution, limits and proper
names. Uni. California Press, London.
Veth, P. 1989 Islands in the Interior: a model for the colonisation
of the arid zone of Australia. Archaeology in Oceania 24:8192.
Veth, P. 1993 Islands in the Interior: The dynamics of Prehistoric
adaptations within the arid zone of Australia. Michigan:
Ann Arbor. International Monographs in Prehistory.
Archaeological Series 3.
Veth, P., Hamm, G. and Lampert, R. J. 1990 The archaeological
significance of the Lower Cooper Creek. Records of the
South Australian Museum 24(1):43-66.
Williams, M., Prescott, J., Chappell, J., Adamson, D., Cock, B.,
Walker, K. and Gell, P. 2001 The enigma of a late Pleistocene
wetland in the Flinders Ranges, South Australia. Quaternary
International 83-85:129-144.
33
Book Reviews
Socializing stone artifact assemblages: Regionalization and raw material
availability in northern Queensland
David R. Guilfoyle
Abstract
During the late Holocene in northern Queensland,
Australia, a process of increasing regionalization appears to
have resulted in a decline in the access to raw material
sources, and as a result, an increase in the concern for the
conservation of raw material has occurred at certain sites.
This pattern suggests that socio-demographic processes are
the primary agent affecting variation in stone artifact
assemblages. Moreover, it suggests that traditional models
linking stone artifact assemblage variation to environmental
conditions and/or mobility systems are to be amended.
Introduction
Most regional archaeological studies of hunter-gatherer
societies are limited by the dominance of universal cultural
ecological models and a lack of theoretical and
methodological perspectives allowing the incorporation of
social processes into interpretations of land-use systems. In
particular, interpretations of any spatio-temporal patterning
in stone artifact assemblages are restricted to the
environmental/economic realm and any possible social
process affecting variation is often ignored.
For instance, a number of studies have interpreted the
mid-to-late Holocene changes in the archaeological record
of Australia as representing a process of intensification and
increasing regionalization (c.f. Lourandos 1997; David and
Chant 1995). According to Lourandos (1997:319), sociodemographic processes during the mid-to-late Holocene
resulted in the development of more closed social or
alliance systems, involving the use of smaller-sized
territories, logistical strategies, and economic
intensification or specialization (Lourandos 1997:319).
However, the effects that these processes had on stone
artifact assemblage structure and organization has been
largely ignored, restricted to evidence of increased discard
rates and the appearance of new stone-working strategies
and forms. The limited behavioural relationships proposed
for the process of socio-demographic change and stone
artefact assemblage organisation is due in part to the
persistence of the conception of stone artefacts in Australia
as amorphous, with specific technological practices often
assumed to be widespread (David and Chant 1995:363). In
addition, current theoretical models of Australian stone
artefact assemblages are based on principles of cultural
ecological (forager-collector) settlement models and as a
result, major social changes across diverse regions are
proposed with little or no reference to patterns in stone
artefact assemblage structure and organisation; and changes
in stone artefact assemblage organisation are seen as
independent of the social realm. Little attempt has been
made to examine the interaction of environmental,
Research Unit, Cultural Heritage Division, Department of Environment and
Conservation (NSW), PO Box 1967, Hurstville, NSW 2220, Australia.
E-mail: [email protected]
34
economic and social factors on spatio-temporal variability
in stone artefact assemblage structure.
Thus, there is little discussion as to the behavioral
significance underlying these changes in assemblage
structure and organization, except to say that it represents
yet another aspect of the increasingly dynamic nature of the
mid-to-late Holocene archaeological record and, for this
reason alone, supports the notion that major sociodemographic change was taking place. Indeed, little
explanation has been offered to account for the appearance
of the variety of typologically regular forms of retouched
flakes – bifacial points, backed blades and tula adzes – that
appear in the Australian archaeological record between
approximately 7000 to 4000 years ago. These implements
are depicted as components of a pan-continental entity
known as the Australian Small Tool Tradition (Hiscock
1994:268). Investigation of the socioeconomic function of
these implements has been discouraged by an ongoing
focus on exploring the reasons behind the ultimate origins
of the Tradition, that is, whether they reflect an independent
Australian invention or were introduced from elsewhere
(Hiscock 1994:268; Bowdler and O’Connor 1991:54).
The only attempt to address this neglected phenomenon
is Hiscock’s (1994) model, which is based on a cultural
ecological interpretation suggesting that stone implements
such as points, backed blades and tula adzes reduced the
risks imposed by climate change during the mid-to-late
Holocene. In this paper, consideration was given to the
mechanical effectiveness of these implements for particular
uses (involved in resource procurement), and how they are
consistent with organizational properties associated with
the broader settlement-subsistence system in which they are
embedded (i.e. to assist in the conservation of raw material).
For example, according to Hiscock (1994), technological
strategies involving the manufacture and use of backed
blades represent the modus operandi for diversified mobile
foraging patterns that minimize the risks involved with
exploiting a landscape where the location of resources is
unpredictable. However, these implements are associated
with assemblages in productive and non-productive
environments, and date to the mid-Holocene as well as the
late Holocene, implying that a simple correlation between
the Australian Small Tool Tradition and contexts of high
mobility is not relevant.
In the subsequent sections, the limitations of traditional
interpretations of stone artifact assemblage variability are
identified, and an argument for the incorporation of social
process to assemblage variation is proposed. An exploratory
examination of the regional archaeological data from Cape
York, in northern Queensland, Australia, is used to illustrate
how a movement away from cultural-ecological models
reveals an additional mechanism for linking social process
and assemblage variation. In addition, this reorientation
provides an explanation for the spatio-temporal variation
underlying the use of implements associated with the
Australian Small Tool Tradition. It identifies an association
David R. Guilfoyle
between social processes of regionalization and increased
territoriality as they relate to changing access to raw
materials. Such an explanation is presented to challenge the
traditional deterministic linkage between curated,
formalized assemblages and increased mobility due to
climate change; that does not take into account the marked
spatio-temporal variation in the use of these implements.
Traditional approaches
Soon after Binford’s (1980) forager-collector model was
published, the principles were applied to the analysis of
stone artifact inter-assemblage organization; it remains the
dominant theoretical framework for comparing and
contrasting assemblages across space and time. Presumably,
patterns in the structure and organization of stone artifact
assemblages directly relate to mobility and settlementsubsistence strategies. This premise has been used to link
variation in flaked stone artifact assemblages between
particular land-use systems. Borrowing principles of
Binford’s (1980) settlement model, the degree of effort
associated with the manufacture and maintenance of flaked
stone tools has been related to contrasting levels of mobility
and settlement organization. Studies adopting these
principles suggest that highly mobile groups tend to have
curated, formalized assemblages, whereas sedentary groups
tend to have expedient, un-formalized assemblages (c.f.
Bleed 1986; Henry 1989; Kuhn 1995; Morrow and Jefferies
1989; Shott 1986; Torrence 1989; Parry and Kelly 1987).
The logic underlying this association is that mobile
groups provision themselves with tools that are not only
portable, but multi-functional and readily modified; so that
they can be applied to a range of tasks as need be
(Andrefsky 1999:214). Thus, the assemblages produced by
mobile groups should differ from those of more sedentary
groups, with evidence for more formal tools that are heavily
rejuvenated and recycled. Formal tools require more effort
in the manufacturing process, “whether the production has
occurred over the course of several re-sharpening or hafting
episodes or in one episode of manufacturing from raw
material to finished product” (Andrefsky 1999:214). The
amount of “effort” in tool production is often implied in the
terms expedient and curated assemblages.
In the application of this model, it is assumed that
different “camp” types within the systems will leave
different archaeological signatures. For example, base
camps of the logistical settlement model may be dominated
by specialized assemblages with various tasks that are
logistically organised and “geared” from the base camp.
Residence camps, associated with the residential settlement
system, are dominated by generalized or extensively curated
assemblages, where perhaps tool production and use was
restricted by the necessity of tool portability.
Disputing traditional interpretations
These models of stone artifact assemblage variation are
firmly fixed in the environment/economic realm and
examine only the major structural changes between broad
time periods, or between diverse regions. As regional
settlement patterns are consistently characterized in an
either/or manner, so to have regional stone artifact
assemblages, whereby diverse regions, or different time
periods, are contrasted on the basis of expedient- or curatedtype assemblage structure and organization. The emphasis
on documenting the relationship between particular patterns
of mobility, technology, and environment ensures that social
systems are viewed as having little influence, and much of
the variation remains unaccounted for. Thus, there is
currently a theoretical and methodological gulf in describing
the relationship between stone artifact technology and actual
socio-demographic processes. Changes in past huntergatherer societies are documented without any attempt to
examine the behavioral significance of changes in patterns
of stone artifact assemblage organization.
In Australia, the only behaviorally significant change
attributed to stone artifact assemblages during the Holocene
relate to perceived changes in mobility, based on cultural
ecological models that simply contrast “curated” and
“expedient” assemblages. Indeed, this distinction forms the
basis for almost all regional studies of stone artifact
assemblages in Australia. For the Holocene, a distinction
has been made between mid-Holocene curated assemblages
and late Holocene, expedient assemblages, said to represent
a change toward increasing sedentism in many regions of
Australia (c.f. Torrence 1989; Hiscock 1994); fitting neatly
with the notion of intensification. However, as
demonstrated below, curated assemblages are in some
regions characteristic of late Holocene assemblages, and
this simple observation means that the sedentary/expedient
and mobile/curated dichotomy needs to be re-evaluated or
dropped.
Furthermore, a number of studies have contradicted the
traditional link between mobility and stone artifact
assemblage structure and organization. A range of
ethnographic and archaeological literature documents that
raw material availability/abundance is perhaps the primary
agent in creating differences in the structure and
organization of stone artifact assemblages (c.f. O’Connell
1977; Gould 1980; McNiven 1993; Andrefsky 1994).
Studies such as these suggest that the inter-assemblage
variation is directly related to access to raw material, which
may or not reflect the level of mobility or type of
settlement-subsistence in operation. That is, the level of tool
curation in stone artifact assemblages is telling us less about
forager or collector site types than it is about the differential
use of stone raw material (Daniel 2001).
In Andrefsky’s (1994) analysis of inter-assemblage
variation in three different areas in the western United
States, the abundance and quality or raw material is seen as
the dominant factor in structuring tool production.
Andrefsky (1994) observes only minor differences in the
structure and organization of stone artifact assemblages
between mobile and sedentary populations that had equal
access to raw material. Thus, where raw material is
available, similar assemblage structures result, no matter
what level of mobility is being practiced.
McNiven’s (1993) analysis of six excavated shell
midden sites along Teewah Beach, southeast Queensland,
demonstrates that increasing reduction of artifacts
corresponds with increasing distance from replacement
stone. In the analysis of the unmodified flakes from the six
sites, it was shown that with increasing distance from raw
material source, there is a decrease in the amount of cortex
present on the dorsal surface of flakes and/or platform, and
a decrease in the overall weight and length of flakes
(McNiven 1993:141). McNiven documents increases in the
use of local stone and faunal resources during the last 900
years, which he interprets as the development of a more
regionally focused exploitative systems.
35
Socializing stone artifact assemblages: Regionalization and raw material availability in northern Queensland
Bamforth (1986) investigated the stone tool technology
of the !Kung San, observing that tools made of locally
available raw materials “have shorter use lives, are
maintained less frequently for a small time investment, and
require a smaller time investment overall for procurement
and subsequent maintenance than do tools made of nonlocal materials” (Bamforth 1986:40). Bamforth suggests
that tool raw material availability directly affects the
intensity of maintenance and recycling, and that these
factors account for much of the variation between stone
artifact assemblages.
Studies such as these suggest that the inter-assemblage
variation is directly related to access to raw material, which
may or may not reflect the level of mobility or type of
settlement-subsistence in operation. That is, much of the
variation observed in the structure and organization of stone
artifact assemblages can be attributed to differential
emphasis on the conservation of raw material. This
suggestion is also confirmed by ethno-archaeological
investigations of Aboriginal groups from the Western
Desert and Central Australia (Gould 1980; O’Connell 1977)
which demonstrate that the abundance/availability of raw
material has a strong affect on inter-assemblage variation.
Therefore, it is in this realm that the link between sociodemographic change and stone artifact assemblage
structure and organization should be investigated.
Socio-demographic process and stone artifact
assemblages
An approach is required that allows one to examine how
the interaction of environmental, economic, and social
factors structure raw material procurement strategies and
stone artifact assemblage organization. The link between
theory and method may derive from an examination of the
environmental and social factors behind raw material
distribution, supply and access. For instance, as discussed
above, both environmental and socio-demographic processes
would be expected to have an affect on the access to, or
supply of, stone raw material sources due to the fact that:
Shortages of material are caused not only by
regional geology but also by patterns of behavior
that can increase or decrease the amount of raw
material available in specific contexts (Bamforth
1986:48-49).
For instance, the oft-cited change occurring in the midto-late Holocene of Australia is the development of more
closed territorial practices (Lourandos 1997; David and
Chant 1995). This would create a reduction in the spatial
range of particular groups. As a result, access to stone raw
material sources, a fixed resource of the landscape of any
particular region, is likely to decrease with the development
of regionalization and “increasingly bounded regional
practices” (David and Chant 1995:514).
It has been argued that a key strategy for overcoming
raw material shortages is to use artifacts in more
economical ways to extend their “normal” use lives and
offset replacement costs (Hiscock 1993, 1994), that is, to
adopt strategies which assist in the conservation of raw
material. Therefore, if a process of intensification/
regionalization has been proposed for a particular region,
then this process should be reflected in those aspects of
stone artifact assemblage structure and organization
relating directly to the strategies of raw material
procurement.
36
Therefore, one prediction is that the development of
regionalization through time results in a hypothesized
diminution of raw material available at source and,
therefore, tool production and use will become more
economic which will be reflected in more economical use
of raw materials; and therefore more reduced assemblages
in the upper levels. The Cape York, northern Queensland,
Australia (Fig. 1), archaeological record is used as a case
study to test his notion, and to more clearly explain the
arguments presented here.
Late Holocene regionalization and stone artifact
assemblage change in Cape York
In their analysis of the archaeological record of this
region, David and Chant (1995: 515) state that, beginning
around 3000 years ago, “social factors such as settlement
patterns, patterns of resource exploitation and the like
became more bounded in tandem with an increased
formalization of territorial boundaries.” They argue that
these changes reflect a process of regionalization of sociocultural networks in southeast Cape York, with “major
changes in the configuration of social landscapes (including
inter-regional behavior), involving a closure of territorial
structures, after ca. 3500-2500 BP.” The emergence of
distinct regional rock art styles, new site types, a broadening
of artifact typologies, and increased discard rates represent
the archaeological manifestation of this process.
One may predict that, in this region, changing access to
raw material sources may coincide with the development of
regionalization and a restriction of territorial land-use
practices. The development of territorial boundaries may
restrict movement between regions and result in a general
decrease in the access/availability of raw material sources.
If so, at certain sites, where raw material sources are not
available in the immediate vicinity the (apparent) increase
in regionalization should involve an increase in raw material
economizing behavior; both in the manufacturing process
and in the use of more formal implements. In other words,
the tools made from the source outside control are more
“economized” – local tool stone, however, need not reflect
this. If the expected patterns of more economic use of raw
material are found to exist in the upper layers of the sites
from this region, then this would add substance to the
interpretation of increased regionalization, and provide a
more effective means of linking stone artifact assemblage
change to behavioral changes.
The object of this analysis is not to evaluate the
argument for increasing regionalization, or provide a
detailed examination of raw material sources and stone
artifact assemblage change. Rather, the aim here is to adopt
this model as the theoretical framework to be tested through
an exploratory analysis of the available data. The premise is
that a process of regionalization in this area begins after
3500 BP and new territorial boundaries are established by
around 1000 years BP. The following summary of the
available stone artifact data from the excavated sites in this
region is taken largely from David and Chant’s (1995) own
summary.
Interpreting the Cape York data
Only two detailed site reports from the RookwoodMungana-Chillagoe limestone belt are available. At the site
of Echidna’s Rest, the upper layers (last 700 years before
present) show an increase in the proportion of modified
David R. Guilfoyle
Figure 1 SE Cape York Peninsula showing location of sites mentioned in text (after David and Lourandos 1999).
flakes, compared to the stratigraphic layer dated to between
3000 BP and 700 BP. There is also a two-fold proportional
increase in the number of artifacts with edge modification
on artifacts during the last 700 years than there are in the
older layer (David and Chant 1995:406). The increase in
retouch and modified edges is consistent with an increase in
the concern for the conservation of raw material; increasing
the discard threshold by using tools for a longer length of
time, and investing greater time into tool
maintenance/rejuvenation activities such as re-sharpening
and or reshaping. In addition, the first evidence for the use
of formal implements at this site is represented by nine
burren adzes appearing after 700 years BP (David and
Chant 1995:407).
The use of formal tools at this site coincides with the
perceived development of regionalization in southeast Cape
York. The traditional interpretation has been a late
Holocene decrease in the use of formal tools due to
increased sedentism (Torrence 1989; Hiscock 1994).
However, at this site, and increase in sedentism has resulted
in an increase in the use of formal tools and, therefore, this
traditional interpretation is invalid.
In the Koolburra Plateau region, two sites have been
investigated. At the site of Echidna Shelter (note, different
from Echidna’s Rest) Flood and Horsfall (1986:47-48)
argue that the last 1000 years witnessed a significant shift
in the dominant use of raw materials from quartzite to chert.
The use of better quality raw materials is also an indication
of an emphasis on the conservation of raw materials as it
provides greater precision in knapping, and produces more
cutting edge per area. At the site of Green Ant Rockshelter,
Flood and Horsfall (1986) observe a “broadening of the
tool-kit” with the use of burren adzes occurring at around
1400 BP, which, incidentally, does not coincide with major
climatic changes that appeared to have occurred in this
region around 3000 BP; suggesting that the changes in
assemblage composition are not related to changing
environmental/ecological structures at this site.
37
In the Laura and Bare Hill region, the site of Early Man,
burren adzes make their first appearance sometime between
5500 and 1800 BP, with the appearance of smaller tools and
a broadening of the tool-kit. David and Chant (1995:420)
suggest that the slow deposition rates from layer 7 (11,8505500BP) to Layer 6 (5500-100BP) indicates that “it is
unlikely that these increases commenced much before
1800BP”. Rosenfield et al. (1981) note that there is a
gradual trend towards an increasing use of chert, as well as
towards the manufacture of smaller artifacts thought time.
At the site of Yam Camp, Morwood (1989:97-98) notes “a
general trend towards smaller artifact size in all raw
materials was also evident in the upper part of the
sequence” (i.e. from the last 1000 years BP), suggesting the
economical use of raw material. At Mushroom Rock, a
decrease in the overall weight of the stone artifacts was
noted, beginning around 3000 years ago (David and
Lourandos 1999:115). Also, at Sandy Creek 1, burren adzes
occur between 1890 and 1230 years BP, coinciding with the
development of new territorial boundaries in this region.
In the Ngarrabullgan region (Mount Mulligan) two sites
have been investigated – Ngarrabullgan Cave and Initiation
Cave. For the upper units of Ngarrabullgan Cave, David and
Chant (1995:395) note that major continuities occur in terms
of the stone artifact assemblage, “the most striking of these
concerns the proportions of raw material types used” with a
‘basalt-like’ material used in approximately equal proportion
throughout all layers since the Pleistocene. David and Chant
(1995:395) state that basalt does not occur in the immediate
vicinity of the site, but has to be introduced from the base of
the mountain. However, the base of the mountain is only 2-8
kilometers away, so it might be considered ‘local’ for the
people using the mountain. David and Chant (1995: 395)
also state: “Good quality chert nodules occur in abundance
in the immediate vicinity of the cave.” The abundance of
quality raw material in close proximity would offset any of
the expected effects of economizing behavior associated
with regionalization at this site. The stone artifacts at
Initiation Cave (located at the base of Ngarrabullgan), dating
from approximately 4000 years BP, are classified as
amorphous, with no formal tool types. Again, the proximity
of quality raw material sources would provide no pressure to
produce formalized, curated assemblages.
Likewise, at Fern Cave, chert is found in the immediate
vicinity, while a quartzite source is located around 10
kilometers away. At this site, no formal tool types occur in
the deposit, suggesting that the availability of raw material
sources in close proximity to the site offset any pressure to
practice economizing behavior, if a process of
regionalization occurred. In the Mitchell-Palmer Limestone
Belt, three sites have been excavated (Mordor Cave, Hearth
Cave, Mitchell River Cave), however, no technological
analysis has been undertaken. David and Chant (1995:393)
state: “It may be worth noting that the only artifact type
recognized from this region was a single burren adze slug,
dated to 700-1100 BP, approximately contemporaneous
with those found at Laura and in the Koolburra Plateau to
the north.”
In sum, although the data is limited, broadly
synchronous changes occur in this region around 1500 BP,
involving an increase in, or the first evidence for the use of,
formal implements, an increase in edge modification,
and/or a switch to better-quality raw materials (David and
Chant 1995). On the other hand, sites showing little change
38
in stone artifact assemblage organization are located
adjacent to, or in close proximity to, quality raw material
sources. At Early Man Rockshelter, Echidna Shelter, and
Yam Camp, there is a gradual decrease in the size of
artifacts through time (David and Lourandos 1999:115).
Thus, there is some evidence to suggest an increase in stone
raw material economizing behavior in this region in areas
where tool stone sources were perhaps located outside
territorial boundaries.
Thus, as these changes are perceived to coincide with
the development of regionalization, and associated
development of territorial boundaries and more closed
alliance systems, access to raw material is expected to have
been reduced for some groups, thus explaining the change
in the structure and organization of assemblages toward the
conservation of raw material and the production of more
formalized assemblages. Thus, the late Holocene pattern
involving a shift toward raw material conservation strategies
in southeast Cape York may represent a subsequent decline
in the access to raw materials resulting from a process of
increased regionalization and territorial exclusion. In an
overview of the changes occurring throughout the wider
region, David and Lourandos (1999:121) speculate that
“more dynamic, competitive processes” emerged during he
late Holocene, including the development of more closed,
more formally bounded behaviour.”
Importantly, the characteristics of these assemblages
contradict existing interpretations of assemblage change in
Australia. That is, despite an apparent increase in sedentism
associated with increased regionalization and territoriality,
certain assemblages would be characterized as curated,
formalized assemblages. Furthermore, the use of formal
tools at these sites does not appear to correspond with
climatic changes that occurred during the mid-Holocene.
Therefore, cultural ecological models linking assemblage
variation directly to climate change and/or mobility are not
a valid explanation.
Discussion
The pattern of increasing regionalization and changing
access to, and use of, stone raw materials, may also be
observed in other regions of Australia, where a general
continental-wide trend of socio-demographic process
involving changing inter-group alliances has been
suggested (Lourandos 1997). For instance, in McNiven’s
(1993) analysis of six excavated shell midden sites along
Teewah Beach, southeast Queensland, an increase in the use
of local stone and faunal resources during the last 900 years
is interpreted as evidence for the development of a
increasing regionalization and territoriality, based on a
perceived decline in the level of integration with inland
areas.
Similarly, Morwood (1979, 1987) argues that around
4300 years ago, a distinctive Central Queensland rock art
type began to emerge, and this coincided with the
appearance of formal tools associated with the Australian
Small Tool Tradition. Morwood concludes that “changes in
Central Queensland stone-working technology at this time
can be seen as a regional manifestation of a Pan-Australian
change, with important implication for inter-group
relationships within Australia” (Morwood 1979: 409). Such
implications, he argues, may have included the development
of increasingly regionalized social networks.
Thus, these alternative interpretations suggest a general
David R. Guilfoyle
process of regionalization reduced access to raw material
sources. In northern Queensland, the perceived pattern of
increased regionalization broadly coincides with a change
in the structure and organization of stone artifact
assemblages toward a greater concern for the conservation
of raw material. Thus, the (apparent) relationship between
the adoption of formal tools and regionalization in Cape
York provides one possible (behavioral) explanation for the
widespread adoption of, and spatio-temporal variability in,
implements associated with the Australian Small Tool
Tradition (ASTT).
That is, implements such as points, backed blades and
tula adzes, are not only mechanically effective for particular
uses associated with resource procurement, but also possess
organizational properties within the broader settlementsubsistence system that assist in the conservation of raw
material (Hiscock 1994). That the new stone-working
strategies coincide with the development of regionalization
in some areas may suggest that, while performing similar
functions as the existing technology (sharp edges for
cutting, engraving, wood-working), they have the advantage
of producing more cutting edge per area, thereby,
conserving raw material. Technological strategies involving
the manufacture and use of such implements would
represent, therefore, the modus operandi for groups faced
with a decrease in the availability of replacement tool stone.
If one advantage of the adoption of formal implements
associated with the ASTT is in the conservation of raw
material (Hiscock 1993, 1994), then the spatio-temporal
occurrence of such implements in various regions might be
expected to coincide with the development of increased
regionalization, and subsequent changing patterns of access
to raw material. However, one would expect marked
regional variation in the processes described above. For
example, in arid or marginal areas, the effects of
regionalization may be somewhat restricted spatially, due to
the large range sizes needed to acquire widely dispersed
water and animal resources. Thus, access to raw materials
may not be adversely affected and, therefore, major changes
assemblages are not required.
Also, with increasing regionalization, trade/exchange
relations may become more formalized and efficient (David
and Chant 1995:357), suggesting that trade and exchange of
raw materials may have increased, creating a subsequent
increase in the access to raw material. The result would be
a subsequent decline in strategies aimed at raw material
conservation; including a reduction in the use of formal
implements. A further possible scenario creating substantial
regional variation in the occurrence of formal implements is
that the development of resource intensification strategies
(such as the mass harvesting of eels) may have resulted in a
decline in the importance of stone implements themselves.
Further analysis of the northern Queensland data would
be required to examine, in more detail, the relationship
between raw material procurement strategy, stone artifact
assemblage organization, and socio-demographic process.
However, the above example suggests that examining
strategies of raw material procurement is an effective
method for linking stone artifact assemblage organization to
the past settlement systems and overall land-use patterns.
Conclusion
A major conclusion of this descriptive, exploratory
analysis is that a more effective means of linking stone
artifact assemblage structure and organization to the
general process of socio-demographic processes (such as
intensification), that has heretofore been neglected, is
required. This involves a shift in the emphasis placed on
stone artifact organization, in terms of its relevance to
aspects of past socio-economic behavior. As distance to raw
material source is a major factor causing variation in the
structure and organization of stone artifact assemblages, in
assessing the range of possible explanations for intra- and
inter-regional change or variation, one must consider the
possible factors causing change in the access to raw
material. Analysis of spatio-temporal variability in the use
of raw materials can provide a more effective means of
linking stone artifact assemblage structure and organization
to actual socio-demographic processes.
For instance, the development of regionalization is one
possible event that may reduce access to raw material
sources. In northern Queensland, the perceived pattern of
increased regionalization broadly coincides with a change
assemblages toward a greater concern for the conservation
of raw material. This pattern may also be observed in other
regions of Australia, where a general continental-wide trend
of socio-demographic process involving changing intergroup alliances has been suggested (Lourandos 1997).
Few regional studies document the range of variation in
economic strategies based on patterns in stone artifact
assemblages during particular time periods, or within any
particular environmental setting. Instead, regions are often
pigeonholed into either a residential or a logistical
settlement system, based on broad, subjective ecological
criteria. As opposing economic strategies, the entire
patterning in the archaeological record of a particular region
is characterized as either logistical or residential. Much
variation is consequently overlooked in favour of
homogeneity. While the concept of logistical and residential
settlements remains a useful framework for categorizing
archaeological data and individual sites, it should not be
applied in a strict either/or sense to characterize regional
data, and interpretations should extend beyond simple
ecological criteria affecting perceived levels of mobility.
Numerous social processes are expected to have played
a major role creating variation in stone artifact assemblage
structure and organization, with regionalization and a
decline in raw material access being but one possibility. By
incorporating social processes into the analysis of variation,
the interpretative potential is enhanced while the limitations
of traditional cultural ecological models are exposed.
Rather than viewing the Australian Small Tool Tradition as
simply a pan-continental entity that is loosely associated
with a process of intensification, or as an response to risks
associated with climate change, situating the use of these
implements within a particular socio-demographic context
provides a more effective means of interpreting variation in
stone artifact assemblages.
References
Andrefsky, Jr., W. 1994 Raw material availability and the
organization of technology. American Antiquity 59:21-35.
Andrefsky, Jr., W. 1999 Lithics: Macroscopic Approaches to
Analysis. Cambridge: Cambridge University Press.
Bamforth, D. B. 1986 Technological efficiency and tool curation.
American Antiquity 51: 38-50.
39
Binford, L. R. 1980 Willow smoke and dogs’ tails: Hunter-gatherer
settlement systems and archaeological site formation.
American Antiquity 45:4-20.
Bleed, P. 1986 The optimal design of hunting weapons:
Maintainability or reliability. American Antiquity 51:737747.
Bowdler, S. and O’Connor, S. 1991 The dating of the Australian
Small Tool Tradition, with new evidence from the Kimberley,
WA. Australian Aboriginal Studies 1991/1: 53-62.
Daniel, I. R., Jr. 2001 Stone raw material availability and early
Archaic settlement in the Southeastern United States.
David, B. and Chant, D. 1995 Rock art and regionalization in north
Queensland prehistory. Memoirs of the Queensland Museum
37:357-528.
David, B. and Lourandos, H. 1999 Landscape as mind: Land use,
cultural space and change in north Queensland prehistory.
Quaternary International 59:107-123.
Flood, J., Horsfall, N. 1986 Excavations at Green Ant and Echidna
Shelter, Cape York Peninsula. Queensland Archaeological
Research 3: 4-64.
Gould, R. A. 1980 Living Archaeology. Cambridge: Cambridge
University Press.
Henry, D. O. 1989 Correlations between reduction strategies and
settlement patterns. In D. O. Henry and G. H. Odell (eds)
Alternative Approaches to Lithic Analysis, pp. 139-212.
Boulder: Westview Press.
Hiscock, P. 1993 Bondaian technology in the Hunter Valley, New
South Wales. Archaeology in Oceania 28:64-75.
Hiscock, P. 1994 Technological responses to risk in Holocene
Australia. Journal of World Prehistory 8(3):267-292.
Kuhn, S. L. 1995 Mousterian Lithic Technology. Princeton:
Princeton University Press.
Lourandos, H. 1997 Continent of Hunter-Gatherers: New
Perspectives in Australian Prehistory. Cambridge:
Morrow, C. A. and Jefferies, R. W. 1989 Trade or embedded
procurement? A test case from southern Illinois. In R.
Torrence (ed.) Time, Energy, and Stone Tools, pp. 27-33.
Cambridge: Cambridge University Press.
Morwood, M. J. 1979 Art and stone: Towards a Prehistory of
Central Western Queensland. Unpublished Ph.D. thesis,
Department of Prehistory and Anthropology, Australian
National University, Canberra.
Morwood, M. J. 1987 The archaeology of social complexity in
south-east Queensland. Proceedings of the Prehistoric
Society 53:337-350.
McNiven, I. J. 1993 Raw material proximity and bevel-edged tool
use, Teewah Beach, southeast Queensland. Archaeology in
Oceania 28(3):138-143.
O’Connell, J. F. 1977 Aspects of variation in Central Australian
lithic assemblages. In R.V.S. Wright (ed) Stone Tools as
Cultural Markers: Change, Evolution and Complexity, pp.
269-281. Canberra: Australian Institute of Aboriginal
Studies.
Parry, W. J. and Kelly, R. L. 1987 Expedient core technology and
sedentism. In J. K. Johnson and C. A. Morrow (eds) The
Organization of Core Technology, pp. 285-304. Boulder:
Westview Press.
Rosenfield, A., Horton, D. and Winter, J. 1981 Early Man in North
Queensland. Canberra: Australian National University. Terra
Australis 6.
Shott, M. J. 1986 Settlement mobility and technological
organization: An ethnographic examination. Journal of
Anthropological Research 42:15-51.
Torrence, R. 1983 Time, Energy, and Stone Tools. Cambridge:
Veth, P. 1995 Aridity and settlement in northwest Australia. In J.
Allen and J. F. O’Connell (eds) Transitions: Pleistocene to
Holocene in Australia and Papua New Guinea. Antiquity
69:733-746.
UNCOMMON GROUND:
WHITE WOMEN IN ABORIGINAL HISTORY
Edited by
Anna Cole, Victoria Haskins, and Fiona Paisley
Aboriginal Studies Press
2005
ISBN 0-8557-5485-0
RRP A$34.95
(paperback)
40
Sourcing stone from the Sydney region: A hatchet job
Tessa Corkill
Abstract
This paper provides an analysis of 326 edge-ground
hatchet heads collected from the region surrounding
Sydney, New South Wales. A study of attributes based on
raw material and form reveals that the majority of blanks in
all areas are likely to have originated in gravels of the
present Nepean/Hawkesbury River and abandoned palaeochannels, which are mainly located about 50 km inland
from Port Jackson. This finding supports sparse
ethnographic and some archaeological accounts but
contradicts a number of reports that have postulated sources
for these hatchets considerably further afield. Results of the
study also invalidate a commonly held belief about raw
material type, which has influenced the identification of
sources in the past. The findings have implications for
studies of social factors such as trade and exchange,
selection criteria, the accessibility of raw materials and time
budgeting.
Introduction
The Australian Museum in Sydney holds hundreds of
Aboriginal stone hatchet heads that were obtained in diverse
ways from the region surrounding Sydney (Fig. 1) since the
nineteenth century. Apart from those found during formal
excavations and surface collections, records in the Museum
have little locational information for most apart from a
general provenance (e.g. “ploughed up at Riverstone”
[Museum Registration E77081]). A few are more precisely
located (e.g. “Innes Orchard” [Museum Registration
E31527]) but practically none has a map reference. And
there is rarely any mention, let alone discussion, of what
kind of stone they were made from or where the raw
material might have originated.
If we knew the source or likely source of the raw
material for a particular hatchet, in addition to its collection
location, we would have a basis for looking at aspects of
social organization such as trade and exchange. For
example, how far from sources of the materials were
hatchets found? And, does increased distance indicate long
distance travel for collection, or trade with people who
owned or controlled the sources?
In this paper I present and discuss data derived from The
Australian Museum’s collection of Sydney region hatchet
heads and from raw material samples I collected during the
course of research between 2001 and 2003. The project
aimed to identify the rock type and original form1 of each
hatchet and locate a source or potential sources of its raw
material (see Corkill 2003 for full data). I also review
previous reports that suggested raw material sources, and
discuss the veracity of their findings.
Results of the project demonstrate that most Sydney
hatchets are almost certainly locally derived and that earlier
claims for long distance trade for their raw materials are
unsupported.
Historical background
As soon as members of the Cook expedition set foot on
the shores of Botany Bay in 1770, followed later by the First
Fleet in 1788, they began to gather information about the
Indigenous inhabitants of the Sydney area. In official
documents, journals and casual notes they set down details
of their appearance, language, rituals and daily life,
including what foods they ate and how they obtained them.
Among the items of material culture recorded by these
first British explorers and settlers were hafted hatchet heads
or Mogo. Hatchets were used for various tasks, including
bark removal to make “canoes, shelters and shields, and to
get wood to make clubs, containers and other implements
and weapons” (Attenbrow 2002:89; see also Collins
1798[1975]:487). They were also used to cut toeholds and
enlarge holes while climbing trees to collect honey and
catch possums (e.g. Attenbrow 2002:89-90; Hunter
1793[1968]: 61).
The hatchet heads were made of stone, with at least one
end ground to a sharp edge (see for example Bradley
1792:89, 129, 170; Collins 1798:487, 510; Flannery
1996:52,193; Hunter 1793[1968]:43,61,147, 515,519;
Stockdale 1789[1982]:114; White 1790[1962]:136,157,
200-1; Worgan 1788[1978]:51). Very few details about their
construction, shape and size are given, but Bradley
(1792[1969]:129) calls the implement “a miserable blunt
tool”. In a somewhat less disparaging tone, White
Anthropology, Australian Museum, 6 College Street, Sydney, NSW
2010, Australia. Email: [email protected].
1
For example a water-rolled cobble or a piece quarried from an in situ
lava flow.
Figure 1 Study area.
41
Figure 2 Hafted hatchet (Stockdale 1789[1982]: part Plate 13).
(1790[1962]:157) states that during one of the colonists’
first visits to Broken Bay at the mouth of the Hawkesbury
(Fig. 1), the stone hatchet he saw in a “native” hut was “of
a very superior make to what they usually have”.
Stone hatchets also appear in early illustrations (Fig. 2
shows one example, from Stockdale 1789[1982]:Plate 13;
see also Beaglehole 1968:397; Collins 1798[1975]:368;
White 1790[1962]:Plate 37), but only some of these have
scales by which the size of the implement can be estimated.
The illustrations are mainly engravings that were produced
in England or elsewhere, based on drawings and specimens
brought back from the new colony, and were often of
dubious accuracy - for example the head of an “ax” (sic)
shown in Stockdale appears to be more a stylized than an
accurate representation.
Better indicators of size and basic shape are stencils of
hafted hatchet heads on rockshelter walls, for example those
near Wisemans Ferry, at Maroota (Fig. 1, this paper) and
Canoelands (McCarthy 1961:115; Stanbury and Clegg
1990:104), and at Long Island in the Hawkesbury River
(Mathews 1896:95, Plate I, Fig. 9a).
There are only two temporally separate early accounts
that document where hatchet raw material was obtained. In
the first, Bradley (1792[1969]:170) wrote that in July
1789 Governor Phillip’s expedition reached “very shoal
water with very large hard stones (of which the Natives
make their hatchets &c)” near Richmond Hill on the
Hawkesbury River (Fig. 1). However, how Bradley
obtained this information is unclear (Attenbrow
2002:123).
The second account states that in April 1791 a
government party, accompanied by two coastal Aborigines,
Colebe and Ballederry, was exploring land in the same area
when they met up with a group of inland Bu-ru-be-rong-al
people, including Go-me-bee-re, Yal-lah-mun-di and a child
Jim-bah. According to Governor Phillip they had apparently
“come this journey in order to procure stone hatchets, as the
natives get the stones whereof they make their hatchets
from that part of the river near Richmond-Hill” (Phillip in
Hunter 1793[1968]:513-525). The gist of this report was
42
repeated in several contemporary accounts, for example
Collins (1798[1975]:487). However, it should be noted that
no actual observations of Aboriginal people collecting stone
from this location or from others in the Sydney region
appear to have been recorded.
Since the late eighteenth century, many hatchets, usually
unhafted, have been collected from various parts of the
region. In the early days some were presented by their
owners or bartered for, but as the Aboriginal population
dwindled, due to disease, conflict and forced removal from
their homelands, many hatchets were found abandoned,
discarded or cached, either in rockshelters, during
ploughing of newly enclosed fields, or during professional
and amateur excavations. Some ended up in sheds and
private museums but many are now in more formal
collections.
The study area
The Sydney region can be defined in many ways but for
the purposes of this research it was taken to extend
southwards from Broken Bay and the northern reaches of
the Hawkesbury River to the southern parts of the Royal
National Park, and eastwards from the lower Blue
Mountains to the coast (Fig. 1).
The geology of the area varies from the sandstones of
the Hawkesbury and Narrabeen Group rocks (mainly in the
north, south, along parts of the coastline and to the west of
the Hawkesbury/Nepean River) to the shales of the
Wianamatta Group (mainly in the central areas), volcanic
diatremes and dykes (widely scattered), gravelly and sandy
Tertiary palaeo-channels (mainly in the west) and
Quaternary river and beach deposits.
To identify the area in which each hatchet head was
found, the study area was divided into nine Project Zones;
their code names are based on an alphanumeric grid system
(Fig. 3). Although the inland zones appear to be of equal
size it should be noted that maps illustrate horizontal area,
whereas the presence of hills increases the actual land
surface area (Corkill 1992). The three eastern zones also
comprise varying amounts of land, due mainly to the
Tessa Corkill
Figure 3 Hatchet counts in each Project Zone (1-3 = west
to east; A-C = north to south).
orientation of the coastline2. However, the discrepancies in
land area were found to be irrelevant to this project (see
below - Spatial distribution - Overall).
Previous research
Since the late nineteenth century a number of
researchers have suggested raw material sources for
hatchets found in various parts of the region. Some of these
suggestions seem credible but others are improbable. Some
of these hatchets (often called axes in the reports) are in the
Australian Museum and were included in this research.
In 1880 James C. Cox exhibited eight stone axe heads
at a meeting of the Linnean Society of New South Wales
(Cox 1880:271-272). They were ploughed up at
Castlereagh on the Nepean River and Cox thought (from
their position) that they had probably been buried in an
Aboriginal grave (though no skeletal material was
mentioned as having been found). He also mentioned that
30 similar axe heads had been discovered on the other side
of the river “under somewhat the same circumstances”. The
raw material was identified as “dioritic”. Although Cox
himself did not speculate on the source of the raw material,
this is an area where suitable material is abundantly present
(see below). Many years later, Australian Museum
Anthropology Curator F.D. McCarthy was in no doubt that
the hatchets originated from the Nepean River “axe
factory” and “quarry” shown to him in the 1930s by G.E.
Bunyan, an interested local resident (McCarthy 1978:50).
In 1889 geological surveyor T.W. Edgeworth David and
palaeontologist R. Etheridge Junior wrote that the raw
material for an edge-ground implement which was found
2
3
4
In Zone 3A it is also due to the fact that areas north of Broken Bay were
not included in the research.
The Nepean and Hawkesbury are different sections of the same river
(see Fig. 1).
Although the text says there were seven thin-sections made, the Table
stipulates there were only six.
with human remains at Long Bay, near Botany Bay, and
was made of “a dark micaceous flagstone …may have been
obtained from local beds in the Hawkesbury Series”
(Edgeworth David and Etheridge 1889b:12 and Plate I).
This source seems unlikely as none of the rock in
Hawkesbury Sandstone strata conforms with the criteria
necessary for the production of serviceable edge-ground
artefacts.
Also in 1889, Edgeworth David and Etheridge found
two hatchets in a “kitchen midden” at Forty Baskets Beach,
Port Jackson. They reported they were made from
“travelled pebbles, not representing local rocks”
(Edgeworth David and Etheridge 1889a:144 and Plate XX).
The description “travelled pebbles” implies that waterrolled, evenly-shaped stones were used. This, plus the
suggestion of a non-local source, is significant, as will
become clear later in this paper.
In 1894 Chemistry Professor A. Liversidge described
18 hatchet heads from various parts of Sydney, mainly
coastal but some from further inland (Liversidge 1894).
Fifteen were made from “pebbles of spotted altered
claystone” (hornfels) and one each from dolerite, diorite
and quartzite. He suggested that the altered (i.e.
metamorphosed) claystone pebbles had probably been
brought from the old riverbed at Lapstone Hill, Emu Plains
(near Penrith) (Liversidge 1894:233). This certainly seems
possible, as this material is among those present in the
exposed gravels adjacent to the Nepean at Emu Plains.
In a brief paper in 1928 W.W. Thorpe and M.S. Stanley
discussed Aboriginal axe manufacture and included a photo
of ten “pebble-axes from Emu Plains, Nepean Valley”
(Thorpe and Stanley 1928). No raw material source was
proposed in the paper but McCarthy (1978:50) referred to
this collection when writing of “the great axe factory there
on the Nepean”. The axe heads were donated to the
Australian Museum by local orchardists and are included in
my data. In 1948 McCarthy wrote of the “inexhaustible
supply of pebbles” in the Nepean beds for manufacturing
edge-ground artefacts. He also stated that “The most
favoured material …is hornfels, although a wide range of
other stones is present” (McCarthy 1948:23).
A ground-edge axe, plus a fragment of a second one,
were found in the 1960s during excavation of a rockshelter
on Gymea Bay (near Botany Bay) by a team led by Vincent
Megaw of the University of Sydney’s Archaeology
Department. The artefacts were identified as being made of
Cordierite Hornfels and the “nearest coastal locality” of
this rock was given as the Upper Shoalhaven River (Megaw
1966:33, Fig. 5). This area is over 150 kilometres south
west of Gymea Bay and is not what I would call coastal as
it is at least 50 kilometres from the sea. Other possibilities
were given as Bathurst or Marulan, both of which are also
over 150 kilometres away; the latter, in fact, is near the
Upper Shoalhaven. The rock type may be present in those
areas, but they are certainly not the “closest” known
locations - hornfels can be found around 50 kilometres
away, coastally at Bellambi Point near Wollongong and
inland in the Hawkesbury/Nepean3 gravels (see below Potential raw material locations).
David Branagan, a University of Sydney geologist,
collaborated with Megaw in a publication on the lithology
of artefacts excavated at Curracurrang, in Royal National
Park (Branagan and Megaw 1969). Seven axes were said to
have been thin-sectioned4. Of these, three were identified
43
as cordierite hornfels, two of igneous origin and one of
tinguaite, “an unusual [igneous] rock which is known to
outcrop only in the Minnamurra region, about 40 miles
south”. The other two igneous axes were thought likely to
have come from the Wollongong-Kiama region (Kiama is
on the coast about 70 km south), although Milton, even
further south, was also given as the possible source of one.
Again, the nearest source of hornfels is given as the Upper
Shoalhaven, but this time “perhaps more likely” sources are
even further away (around 250 km), on the south coast near
Moruya and Bodalla (Branagan and Megaw 1969:14 and
Table 5).
In 1974 Megaw edited a group of Sydney district
excavation reports. Again edge-ground hornfels artefacts
were potentially sourced to the far south coast. And this
time the Blue Mountains, “a dyke on the coast” and the
Shoalhaven area were added to the list of possible igneous
sources (Megaw 1974:5, 31). It is interesting that the
Megaw reports focus on the rock type and tend to ignore the
original form of the hatchet blanks when suggesting
sources, unlike the next reporter in this list.
Frank Dickson, from the University of New South
Wales, is perhaps the best known researcher of Sydney
hatchet heads. In 1972 he suggested that the flattish
waterworn pebbles of “indurated siltstone”, which were
used to make axes found at Kurnell on Botany Bay, might
have come from Thelma Head, some 30 km south. Other
axes from the same area were reported as having been made
of “basaltic” stone, which Dickson says may have been
obtained from a formation at the foot of Macquarie Pass
“about 100 km by road from Kurnell”. He also notes that
there is “a basalt reef below high tide level at Bellambi”
(Dickson 1972:206). Unfortunately, as it is unclear which
Kurnell hatchets are referred to and whether or not they are
in the Museum, it was not possible to verify the raw
material of those implements during my research, a
necessity before trying to tie them to a source.
In 1984 Jim Kohen and his co-writers felt sure enough
of their sourcing of artefactual stone from Shaws Creek II
rockshelter, on the western side of the Nepean River, to
merely mention in passing that there was a “convenient
supply of …basalt pebbles in the bed of the river and
suitable sandstone outcrops for grinding stone hatchet
heads” (Kohen et al. 1984:59). Isabel McBryde and Alan
Watchman, however, went through a comprehensive process
of elimination before concluding that the origin of a spotted
hornfels hatchet head, found in material recovered from the
1790 wreck of the “Sirius” off Norfolk Island, was likely to
have been the gravels of the Hawkesbury/Nepean
(McBryde and Watchman 1993). This artefact is now in the
Norfolk Island Museum and was unable to be accessed for
this research.
In recent times many archaeologists have found edgeground hatchets or fragments during field surveys and
excavations in the study area. The raw material is usually
identified in these reports as igneous, mainly basalt or
“fine-grained basic” (e.g. Kohen 1986:81). Similarly, when
speculating about hatchet raw materials and their potential
5
6
44
Pebble-shaped: Somewhat flattened ovoid to semi-rectangular pebbles
or cobbles. Their shape indicates water-rolling in river or ocean.
However, some volcanic outcrops contain exotic material picked up
during eruption.
sources, igneous types are often the only ones mentioned
(e.g. Attenbrow 1996:36; Smith 1988:16,17,28-29) - in the
past I have done it myself (e.g. Corkill 1990, 1995, 1999a).
As will become clear below, my current research indicates
that many of these identifications are likely to have been
incorrect and the speculations inadequate.
As mentioned above, some of the hatchet heads
described in reports are present in the Australian Museum
and were examined during this project. However, many are
in other collections or could not be found or accessed.
Potential raw material locations
Raw materials for edge-ground stone tools should be
“very tough, resistant to fracture and free of cracks and
other flaws. Fine to medium-grained rocks with strongly
interlocking textures or strong intergranular rocks are
preferred” (Domanski and Webb 2000:178-9, citing Fenton
1984:223, 231; see also Dickson 1981:27-33). These are
generally metamorphic and igneous rocks such as hornfels,
greenstone, basalt and dolerite (Dickson 1981:26;
Domanski and Webb 2000; Kamminga 1982:25; McBryde
1978:355-356).
Size and shape also seem likely to have been important
when selecting a hatchet “blank”. Preliminary examination
of the Museum’s Sydney hatchets indicated that most were
made from pebble-shaped5 cobbles (archaeologists often
call them pebble axes, but as the artefacts are usually longer
than the accepted pebble length of <64 mm [Whitten and
Brooks 1972:88], cobble is the correct term for the item
itself, as opposed to its shape). When attempting to source
these artefacts it is therefore obvious that one must find
locations where such cobbles are present (i.e. in current and
ancient river channels, beach gravel deposits and
conglomerates).
A minority of Museum hatchets were observed to be
more angular (as opposed to pebble-shaped). Although they
also might have been selected from gravel deposits and
knapped before grinding, their blanks could have been
directly quarried from in situ igneous or metamorphic
strata.
Potential sources were identified from geological maps
(but see Corkill 1999b for problems associated with their
use), from suggestions made by previous researchers (see
above) and during field investigations for this and previous
research (e.g. Corkill 1999a; Corkill in prep).
The relevant geological formations within the study area
consist of Quaternary and Tertiary channel deposits (which
occur most extensively in the western part of the study area)
and igneous outcrops such as dykes and diatremes (which
are more widely scattered but much less extensive) (refer to
Corkill 1999a: Chapter 6; Geological Survey of New South
Wales 1983, 1985, 1991; Herbert 1983; Jones and Clark
1991; Sherwin and Holmes 1986 for further information
about these formations). At a general level (i.e.
identification of a generic rock type such as hornfels,
quartzite or igneous, as opposed to a more detailed
mineralogical analysis aiming to characterize a particular
specimen) it appears that most of the raw materials suitable
for edge-ground hatchets (including igneous rocks) are
available as loose cobbles in the channel deposits. The
igneous outcrops generally contain only volcanic rock6,
which would usually have to be quarried directly from the
in situ strata.
In addition to the Sydney region channel deposits,
Tessa Corkill
Figure 5 River gravels at Emu Plains (EP/-) collection
area.
Figure 4 Raw material collection areas.
water-rolled cobbles of suitable materials are present in
coastal and inland regions surrounding the study area. For
example, at Bellambi Point, near Wollongong, to the south,
there are cobbles which were apparently washed up along
the coastline by a tsunami around AD 1480 (Assoc. Prof.
Ted Bryant, Geosciences, Uni of Wollongong, pers. comm;
see also Bryant 2001:67-71, 257-259). Cobbles of various
materials are also present in outcrops of the Munmorah
Conglomerates to the north, in a Narrabeen Group
conglomerate in the upper reaches of Mangrove Creek, also
to the north (Corkill 1999a:54) and in gravels along Coxs
River to the west.
Offshore, conglomerate bearing strata may have been
accessible somewhat closer to Sydney during times of lower
sea level but by the time the manufacture of edge-ground
hatchet heads in this area commenced, less than five
thousand years ago (Attenbrow 2002:155) such strata would
have been well underwater (Corkill 1999a: Figure 6.2).
Collecting raw materials
In order to examine and test potential raw materials for
comparison with hatchet materials a collection of more than
100 waterworn cobbles and other rocks, of suitable shapes
and sizes, from various locations was made (Fig. 4). The
collection consisted mainly of cobbles from a few of the
gravel exposures adjacent to the current channel of the
Hawkesbury/Nepean River in the western part of the study
area (Fig. 5 shows one of these exposures). A number of
cobbles were also collected from one of the many palaeochannel remnants, some distance from the current river, at
Oakville (OSR/- on Fig. 4) and others from beaches and
outcrops along the coast. The coastal collection resulted
from inspection of all beaches between Broken Bay and
Port Jackson (despite the fact that no gravels are mapped
along the coastline in the study area, it was thought that
some suitable material might be present, particularly from
volcanic intrusions or their margins; however very little
useful material was found).
Two areas beyond the main study area were also
sampled: Kulnura, north of Sydney (K/-), where basalt
outcrops are quarried for road base etc, and Bellambi Point,
south of Sydney (BPt/-), where wave washed cobbles are
present in abundance.
Sourcing methodology
Identification of rock type was determined in several
ways. Preliminary typology of collected raw material
samples was affected by eye in the field. All raw material
samples and hatchet heads were examined under a binocular
microscope at magnifications between 12x and 40x. In
addition, a number of raw material samples were subjected
to X-Ray Diffraction (XRD) analysis, in order to
characterize their composition and further assist in
identification of hatchet rock types, by microscopic
comparison with the analysed pieces. The hatchets
themselves were not subjected to XRD or any other type of
destructive analysis, due partly to the difficulty of obtaining
the necessary permits.
The rock types mainly fall into two basic groups metamorphic and igneous. After initial sorting and reevaluation eight categories were adopted for the hatchet
data base: hornfels, hornfels?, igneous, igneous?, quartzite,
quartzite?, metamorphic indeterminate and unknown.
Artefacts assigned to categories with question marks had
enough attributes to indicate they were likely to belong to
that class but could not be definitely identified. Many of the
Museum hatchets were easily identifiable as hornfels owing
to the presence of distinct spots (usually biotite or
cordierite), or vestiges of these spots that remained after
differential weathering.
Three metamorphic cobbles and two igneous rocks were
later ground in order to test their suitability for hatchet
making.
X-ray diffraction
Seven pieces from the raw material collection were
subjected to X-Ray Diffraction (XRD) analysis in order to
determine the crystalline structure of their minerals and
characterise the rock (Table 1) (see Renfrew and Bahn
1991:318-320 for examples of XRD use in archaeology).
Results of the analysis were mainly but not always
consistent with the microscopic examination of the rock
samples - for example BPt 1 was originally thought to be
igneous as it contained what looked like igneous crystals,
45
ID #
ROCK TYPE
MINERAL % (two highest only)
Chi2 error
DR 3*
YC 7
BPt 1*
BPt 10*
GB 6
K 1*
BH 2c
Spotted Hornfels
Hornfels
Sandstone■
Quartzite
Sandstone
Basalt
Teschenite
58% quartz; 29% muscovite
66% quartz; 16% bytownite
65% quartz; 35% labradorite
83% quartz; 7% labradorite
89% quartz; 4.1% kaolin
36% labradorite; 34% augite
34% labradorite; 27% analcime
3.02
1.37
2.12
5.73
1.61
1.67
2.94
Table 1 Characterisation of rock types by X-Ray Diffraction. For sample location see Figure 4. (*= rocks edge-ground by
researcher; ■ = slightly metamorphosed?).
Zones
1
(West)
2
(Centre)
3
(East)
Totals
A (North)
B (Centre)
C (South)
48
93
13
8
13
2
7
59
83
63
165
98
Totals
154
23
149
326
Table 2 Hatchet counts in each Zone.
Raw
Materials
Totals
West
Centre
East
1A+1B+1C 2A+2B+2C 3A+3B+3C
Hornfels
Hornfels ?
Igneous
Igneous?
Metamorphic
(indeterminate)
Quartzite
Quartzite?
Unknown
52%
7%
11%
2%
64%
9%
5%
1%
61%
4%
4%
0%
38%
5%
18%
3%
4%
8%
7%
9%
1%
6%
8%
6%
9%
0%
4%
17%
5%
11%
7%
11%
Totals
326
100%
154
100%
23
99%
149
98%
Table 3 Hatchet raw materials from west to east.
unlike most sandstones in the region; however, this results
from the separate origin of minerals in the sandstones of the
southern coastal areas. BPt 10 was characterized as a
quartzite but contains minerals which make it more suitable
for grinding than many other quartz-rich quartzites.
Teschenite (BH 2c) is a distinctive type of dolerite (a
volcanic igneous rock) which could have significance in
sourcing, a fact which is discussed below.
The hatchets
Three hundred and twenty six edge-ground hatchet
heads from the study area were included in analyses. They
comprised all those from the Museum’s main collection7
that were available at the time, plus a few from excavated
sites which are stored separately in the Museum.
7
The “main collection” consists largely of individual hatchet heads
stored together in geographically ordered cabinets, regardless of
collection date.
46
Spatial distribution - Overall
The hatchets were found within the nine Zones shown in
Table 2 and Figure 3. It can be seen that the greatest number
of hatchets were found in the west central Zone 1B (93
artefacts) and south east Zone 3C (83 artefacts). As the
latter Zone contains much less land than the former, there
are actually more hatchets per hectare. At first glance this
might appear to have some significance. However, as the
Museum collection results from many years of
unsystematic accumulation and is in no way likely to
represent the number of hatchets that were or are actually
present in the region as a whole, or in any of the Zones in
particular, there is nothing to be gained from comparing the
actual numbers in each Zone. In view of the fact that nearly
all hatchet grinding grooves are found in Hawkesbury
Sandstone (Attenbrow 2002:121), it is interesting to note
that Hawkesbury Sandstone Zones 2A, 3A and 2C have so
many fewer hatchets than the Zones with Wianamatta,
Tertiary and Quaternary geology. Zone 3C is also mainly
Hawkesbury Sandstone, but nearly all the hatchets were
found in the Quaternary sands around Botany Bay.
Spatial distribution - Raw materials
Tables 3 and 4 show raw material totals and percentages
from each Zone (Table 3 from west to east and Table 4 from
north to south). Overall it can be seen that hornfels (at 52%
of the total, or 59% if hornfels? is included) is the most
common material from which hatchet heads were made.
Quartzite and quartzite? make up 15% while igneous and
igneous? only account for 13%.
From west to east (Table 3) it can be seen that hornfels
declines, from 73% in the west to 65% in the central Zone,
to 43% in the east. Conversely, igneous increases from 6%
in the west and 4% in the centre to 21% in the east.
However, it is still only half the percentage of hornfels and
only just more than quartzite, which makes up 18% of the
total in the east.
From north to south (Table 4) there is also a decline in
hornfels percentages (from 78% to 60% to 44%) and an
increase in igneous (from 8% in the north and 7% in the
centre to 27% in the south). Similarly there is an increase in
quartzite (3% to 15% to 24%). Again, hornfels and the other
metamorphics are by far the most common in all three areas
(86% in the north, 79% in the centre and 68% in the south).
Spatial distribution - Blank types
As discussed above, the shape of the hatchet blank is an
important indicator of raw material source. The majority of
Sydney region hatchets are pebble-shaped (see Footnote 5);
Tessa Corkill
Raw
Materials
Totals
North
Centre
South
1A+2A+3A 1B+2B+3B 1C+2C+3C
Hornfels
52%
Hornfels ?
7%
Igneous
11%
Igneous?
2%
Metamorphic
(indeterminate) 4%
Quartzite
8%
Quartzite?
7%
Unknown
9%
68%
10%
6%
2%
55%
5%
6%
1%
36%
8%
22%
5%
5%
0%
3%
6%
4%
8%
7%
13%
2%
12%
10%
4%
Totals
63
100%
165
99%
98
99%
326
100%
Table 4 Hatchet raw materials from north to south.
West - P West - A
North
Centre
South
A
West
Centre
East
95%
91%
84%
5%
9%
16%
Totals
292
90%
34
10%
Table 5 Hatchet Blanks: West to East. (P = pebbleshaped; A = angular.)
P
A
North
Centre
South
89%
92%
86%
11%
8%
14%
Totals
292
90%
34
10%
Table 6 Hatchet Blanks: North to South. (P = pebbleshaped; A = angular.)
they are designated “P” in Tables 5-8. Most are unmodified
except for edge-grinding and, in some cases, minimal
flaking and/or pitting - the latter perhaps due to use as an
anvil. Some of the hatchets are more angular; these are
designated “A” in Tables 5-8. It was not possible to
ascertain their original (pre- selection) shape but they may
have been quarried from in situ strata of volcanic outcrops
or selected as angular pieces from gravel deposits. They
could also have originally been pebble-shaped cobbles that
were highly modified during manufacture.
Tables 5 and 6 show that 90% of the hatchet heads in the
study were fashioned from pebble-shaped blanks. The other
10% do not have enough diagnostic attributes (e.g. cortex)
to tell whether they were originally pebble-shaped cobbles
or pieces quarried from an in situ primary outcrop. There is
some difference from west to east (Table 5) with 95% of
those from the west having pebble-shaped blanks,
compared with only 84% of those in the east. Nevertheless,
the predominant form in the east is still a pebble-shaped
blank. The significance of this is discussed below. Table 6
shows that pebble-shaped blanks differ little from north to
south, with percentages from 86% to 92%.
8%
3%
8%
East - A
57%
85%
86%
43%
15%
14%
Table 7 Hatchet Blanks: Western and Eastern pebbleshaped (P) versus angular (A) blanks, from North
to South.
Raw Materials
P (pebble-shaped)
A (angular)
metamorphic
234
92%
36
84%
22
73%
19
8%
7
16%
8
27%
igneous
unknown
P
92%
97%
92%
East - P
Table 8 Blank Shape of Raw Materials.
In Table 7 it can be seen that pebble-shaped blanks
comprise between 92% and 97% of the totals from the
western zones. In the eastern zones they range from 57% in
the north to 86% in the south, but the difference between
north and south in the east seems likely to have a statistical
rather than a cultural cause (only a small number of hatchets
came from the north east zone - See Table 2).
Blank raw materials
Table 8 demonstrates that more metamorphic than
igneous blanks are pebble-shaped (92% versus 84%) but it
also shows that only 16% of the igneous hatchets do not
have a pebble-shaped preform. This is important in terms of
sourcing and is discussed below.
Hatchet grinding project
In order to test the grindability of raw materials five
rocks from the field collection were edge-ground (Table 9
gives relevant details). Three were pebble-shaped
metamorphic cobbles, similar to many of the hatchets in the
Museum collection. Two were irregularly-shaped pieces of
igneous material (pebble-shaped pieces of this material
were unable to be found).
The grindstone was a rectangular slab of mediumgrained Hawkesbury Sandstone placed on a slight slope,
with a small sprinkler trickling water over the surface
during the grinding process. Each of the three cobbles was
ground on both sides of one short end until a reasonably
sharp edge was produced. There was no attempt to achieve
a specific blade length or angle, which in any case would
depend on the size and shape of the original cobble.
Detailed descriptions and discussion of hatchet making are
to be found in publications by Frank Dickson (e.g. Dickson
1972; 1981).
Apart from some perfunctory wood chopping, no
attempt was made to use the finished artefacts as functional
hatchets. However it was noticeable that the artefact blade
which took the shortest time to grind (BPt 1) chipped
almost immediately it was used. This rock is a slightly
metamorphosed sandstone with what must obviously be a
soft mineral cementing the quartz grains. The other two
47
ID #
Raw Material
Meta/m or
Igneous
%
quartz
Blank
shape
Length
ground (mm)
Grinding
time (mins)
Notes
BPt 1
BPt 10
DR 3
K1
BH 2
Sandstone
Quartzite
Hornfels
Basalt
Teschenite
■
65
83
58
0
4
pebble
pebble
pebble
piece*
piece*
58
35
30
30
49
22
120
45
5
30
chipped easily when used
good edge but hard to grind
good edge
only small area ground
thick piece; not ground
to sharp edge
meta/m
meta/m
meta/m
ig
ig
Table 9 Edge-ground pieces (all were subjected to XRD). (■ = only slightly metamorphosed; *= irregularly shaped piece)
cobble artefacts took longer to grind - BPt 10, which
contained 83% quartz, took 120 minutes to achieve a
reasonable edge. When tested these two appeared to be
much tougher and less inclined to chip than BPt 1.
The two igneous pieces (K1 - basalt, and BH 2 teschenite, a variety of dolerite) were not ground to sharp
edges as the blank shapes were unsuitable. Each material
appeared likely to produce a good edge if an appropriately
shaped blank was available or flaked to shape, but, without
testing, it is not possible to say how easily damageable they
might be.
Variations in the grinding rate and potential for damage
of different rock types raise a number of socioeconomic and
technological issues (for example those concerning
selection criteria, the accessibility of raw materials and
availability of time) which await future investigation.
Conclusions
Data derived from analysis of 326 of the Australian
Museum’s collection of Aboriginal stone hatchet heads
from the Sydney region show that 90% were manufactured
from water-rolled cobbles. In the study area these are only
available in present day river channels and palaeo-channels
at least 20 kilometres, and mainly more than 50 kilometres,
from the sea. Along the coastline there are no documented
cobble beds or cobble bearing strata until one reaches the
Illawarra near Wollongong (around 50 km south of Sydney
CBD) or the Munmorah Lake area (around 80 km north). If
it is correct that cobbles on north Illawarra beaches result
from a tsunami throwing them up from offshore some 500
years ago (Bryant 2001:67-71, 257-259; see also Corkill
2003: Appendix 5), then Sydney manufacturers would
probably not have been able to obtain raw material from
these sources in earlier times.
Only 10% of the hatchets have blanks which may have
come from in situ strata such as volcanic intrusions.
However, even they may have originated in the gravels, as
angular cobbles, or larger pebble-shaped blanks of
particularly suitable material could have been knapped to
size and shape. This 10% includes only seven of the 43
indisputably igneous hatchets. Thus, even if volcanic
intrusions (such as those at Barrenjoey, Bondi or further
afield) were sometimes exploited for igneous raw materials,
it is obvious that material from the gravel beds was usually
preferred.
Overall, there is some decrease from west (95%) to
east (84%) of hatchets made from gravel-derived blanks,
but a large majority were still manufactured from such
blanks. Seventy eight percent of the hatchets were made
from metamorphic rocks, of which the most popular was
hornfels (59% of total). Only 13% were made from
48
igneous rocks. Even if all those in the “Unknown”
category (9%) are igneous they would still only make up
22% of the total. The use of hornfels declined from west
to east (from 73%-43%), whereas igneous increased (from
6%-21%). Both metamorphic and igneous rocks are
present in Sydney and Illawarra gravel beds. The
Munmorah conglomerates to the north also contain
metamorphic and probably igneous rocks (Corkill
1999a:53-54). And, of course, there are many gravel and
conglomerate outcrops further afield, in areas not
investigated for this project.
Perhaps the most significant findings of the research are
that the vast majority of Sydney region hatchets were made
from waterworn pebble-shaped metamorphic rocks. This is
the case in all parts of the region, even those areas where
such raw materials are not present, such as along the
coastline.
So where did the hatchet makers obtain their raw
materials? For the people of the western Sydney region the
most parsimonious answer would have to be the nearby
Hawkesbury/Nepean River and palaeo-channel gravel beds.
That they did so at “contact” is supported by the sparse
ethnohistorical reports.
But what about the people further away, particularly
those on the coast? Did they travel or trade to the west,
north or south, or all three? The fact that the “coastal”
Aborigines in Governor Phillip’s 1791 exploration party
between Rose Hill (Parramatta) and the Hawkesbury area
were able to converse with people whose language or
dialect differed from theirs (Phillip in Hunter
1793[1968]:513-523), even though they did not know them,
indicates that there was considerable social interaction
between coastal and inland groups. But whether they traded
or collected their own raw material is not recorded. There is
also evidence that people travelled up and down the
coastline, particularly to and from the north, for various
reasons including trade (Attenbrow 2002:122-124; Ross
1976:74-79). But again, collection or trade in hatchets or
blanks is not recorded.
This research makes it clear that most raw materials
were available within 50 kilometres of any point in the
Sydney region. So there was no physical need for the travel
or trade over hundreds of kilometres that was suggested by
some of the researchers referred to above. Which is not to
say it didn’t happen from time to time for reasons more
complex than pure propinquity.
More detailed research, probably requiring destructive
procedures (e.g. thin section or elemental analysis) would
be needed to determine a specific source for each of the
many hatchet heads in the Museum collection with absolute
certainty.
Tessa Corkill
And we are left with a plethora of questions. For
example:
1) How did the hatchets get to be where they were
found? Were they lost, discarded or what? And why?
2) Why have so few hatchets been found in Hawkesbury
Sandstone country, particularly since nearly all
grinding grooves are located in this terrain?
3) If people had easy access to inferior material but had
difficulty in getting good material, what choices did
they make and in what circumstances.
These and many other questions will have to await
future research.
Acknowledgements
I thank the Indigenous people of the Sydney region, both
those around today, for allowing me to research their
artefacts, and their ancestors who made them. I am also
grateful for the assistance of a number of people and
organizations during this project. In particular, at the
Australian Museum, Val Attenbrow, Leanne Brass and other
members of the Anthropology Section (for assistance in
locating the hatchets and archival material, use of equipment,
and helpful suggestions), Ross Pogson, Collection Manager,
Mineralogy Section (who undertook the majority of the work
in the XRD analysis) and the archaeological staff of
Australian Museum Business Services (who allowed and
helped me to use their MapInfo GIS program to produce
maps). Galston High School Science Department provided
the use their microscopes for examination of the raw material
collection - thanks especially to Bob Andrews and Eve
Kavanagh. I also thank Isabel McBryde for valuable
discussion about raw material sourcing, Ted Bryant,
Geosciences, University of Wollongong for information
about beach cobble deposits along the Wollongong coastline
and Ian Jack, of the University of Sydney, who was able to
pinpoint the location of the mysterious “Sterculia”. I am also
grateful to Val Attenbrow, Peter White, Kevin Tibbett and an
anonymous referee for their useful comments on earlier
drafts of this paper. And thanks, as always, to John Edgar, for
help and encouragement beyond the call of duty.
References
Attenbrow, V. 1996 Parramatta Park Plan of Management and
Interpretation Program. Aboriginal Sites - Stage II. Report to
Parramatta Park Trust, Paramatta, New South Wales.
Attenbrow, V. 2002 Sydney’s Aboriginal Past: Investigating the
Archaeological and Historical Records. Sydney: University
of New South Wales Press.
Beaglehole, J. C. (ed.) 1968 The Journals of Captain James Cook
- The Voyage of the Endeavour 1768-1771. Cambridge:
Cambridge University Press for the Hakluyt Society.
Bradley, W. 1792 A Voyage to New South Wales 1786-92. Facsimile
1969 ed. Sydney: Trustees of the Public Library of New
South Wales and Ure Smith.
Branagan, D. F. and Megaw, J. V. S. 1969 The lithology of a coastal
Aboriginal settlement at Curracurrang, N.S.W. Archaeology
and Physical Anthropology in Oceania 4(1):1-17.
Bryant, E. 2001 Tsunami: The Underrated Hazard. Cambridge:
Collins, D. 1798[1975] An Account of the English Colony in New
South Wales; with remarks on the Dispositions, Customs,
Manners of the Native Inhabitants of that Country. Facsimile
edition edited by Brian H. Fletcher. Vol. I. Sydney: A.H. and
A. W. Reed in association with the Royal Australian
Historical Society.
Corkill, T. 1990 Preliminary Survey for Aboriginal
Archaeological Sites along F2 - Castlereagh Freeway. Old
Windsor Road to Pennant Hills Road, Sydney, NSW. Report
to Manidis Roberts Pty Ltd for the RTA, Sydney, New South
Wales.
Corkill, T. 1992 Flat maps and steep slopes: How simple
measurements can misrepresent site densities. Australian
Archaeology 35:65.
Corkill, T. 1995 Port Jackson Archaeological Project:
Identification of Raw Materials for Stone Artefacts: John
Curtin Reserve, Northmead. Report to Dr Val Attenbrow,
Australian Museum, Sydney.
Corkill, T. 1999a Here and There: Links between Stone Sources
and Aboriginal Archaeological Sites in Sydney, Australia.
Unpublished MPhil thesis, University of Sydney.
Corkill, T. 1999b The use of geological maps in archaeological
research. In K. May, T. Denham and D. Campbell (eds)
Proceedings of the National Archaeology Students’
Conference 1998, pp. 53-55. Canberra: Department of
Archaeology and Anthropology, Australian National
University.
Corkill, T. 2003 Sydney Region Aboriginal Stone Hatchet Heads:
The Australian Museum Collection. Report to the Australian
Museum, Sydney.
Corkill, T. in prep From source to site: Links between stone
sources and Indigenous archaeological sites in Sydney,
Australia.
Cox, J. C. 1880 Stone axe heads from Castlereagh. Proceedings of
the Linnean Society of New South Wales 5:271-272.
Dickson, F. P. 1972 Ground edge axes. Mankind 8:206-211.
Dickson, F. P. 1981 Australian Stone Hatchets: A Study in Design
and Dynamics. Sydney: Academic Press.
Domanski, M. and Webb, J. A. 2000 Flaking properties, petrology
and use of Polish flint. Antiquity 74:822-832.
Edgeworth David, T. W. and Etheridge, R. J. 1889a On the
examination of an Aboriginal rock-shelter and kitchenmidden at North Harbour, Port Jackson. Records of the
Geological Survey of New South Wales 1(2):140-145 and
Plates XV-XXI.
Edgeworth David, T. W. and Etheridge, R. J. 1889b Report on the
discovery of human remains in the sand and pumice bed at
Long Bay, near Botany. Records of the Geological Survey of
New South Wales 1(1):9-15 and Plate I.
Fenton, M. B. 1984 The nature and sources of the manufacture of
Scottish Battle axes and axe-hammers. Proceedings of the
Prehistoric Society 50:217-243.
Flannery, T. (ed.) 1996 Watkin Tench 1788: A Narrative of the
Expedition to Botany Bay and a Complete Account of the
Settlement of Port Jackson. Melbourne: Text Publishing.
Geological Survey of NSW. 1983 Sydney. Department of Mineral
Resources. Edition 1. Map 9130. Scale 1:100,000.
Geological Survey of NSW. 1985 Wollongong - Port Hacking.
Department of Mineral Resources. Edition 1. Map 9030.
Scale 1:100,000.
Geological Survey of NSW. 1991 Penrith. Department of Minerals
and Energy. Edition 1. Map 9030. Scale 1:100,000.
Herbert, C. (ed.) 1983 Geology of the Sydney 1:100,000 Sheet
9130. Sydney: Geological Survey of New South Wales,
Department of Mineral Resources.
Hunter, J. 1793[1968] An Historical Journal of the Transactions at
Port Jackson and Norfolk Island. Australian Facsimiles
edition No.148, Libraries Board of South Australia ed.
London: John Stockdale.
Jones, D. C. and Clark, N. R. (eds) 1991 Geology of the Penrith
1:100,000 Sheet 9030. Sydney: New South Wales Geological
Survey.
Kamminga, J. 1982 Over the Edge: Functional Analysis of
Australian Stone Tools. Brisbane: Anthropology Museum,
University of Queensland. Occasional Papers in
Anthropology 12.
49
Kohen, J. L. 1986 Prehistoric Settlement in the Western
Cumberland Plain: Resources, Environment and Technology.
Unpublished PhD thesis, Macquarie University, Sydney.
Kohen, J. L., Stockton, E. D. and Williams, M. A. J. 1984 Shaws
Creek KII rockshelter: A prehistoric occupation site in the
Blue Mountains piedmont, eastern New South Wales.
Archaeology in Oceania 19(2):57-73.
Liversidge, A. 1894 Notes on some Australasian and other stone
implements. Journal of Proceedings of the Royal Society of
New South Wales 28:232-245.
McBryde, I. 1978. Wil-im-ee Moor-ring: Or, where do axes come
from? Mankind 10:354-382.
McBryde, I. and Watchman, A. 1993 ‘…lost in the Sirius…’? Consideration of the provenance of the hatchet head
recovered from the Sirius wreck site, Norfolk Island. In
J. Specht (ed.) F.D. McCarthy, Commemorative Papers
(Archaeology, Anthropology, Rock Art). Sydney: The
Australian Museum. Records of The Australian Museum:
Supplement 17:129-143.
McCarthy, F. D. 1948 The Lapstone Creek Excavation: Two
Culture Periods Revealed in Eastern New South Wales.
Sydney: The Australian Museum. Records of The Australian
Museum 22:1-34.
McCarthy, F. D. 1961 A Remarkable Ritual Gallery in Eastern
New South Wales. Sydney: The Australian Museum. Records
of The Australian Museum 25:115-120.
McCarthy, F. D. 1978 New light on the Lapstone Creek excavation.
Australian Archaeology 8:49-60.
Mathews, R. H. 1896 The rock pictures of the Australian
Aborigines. Royal Geographical Society of Australasia,
Queensland Proceedings 11:86105.
Megaw, J. V. S. 1966 The Excavation of an Aboriginal rock shelter
on Gymea Bay, Port Hacking, N.S.W. Archaeology and
Physical Anthropology in Oceania 1(1):23-50.
Megaw, J. V. S. (ed.) 1974 The Recent Archaeology of the Sydney
District: Excavations 1964-1974. Canberra: Australian
Institute of Aboriginal Studies.
Renfrew, C. and Bahn, P. 1991 Archaeology: Theories, Methods
and Practice. London: Thames and Hudson.
Ross, A. 1976. Inter-Tribal Contacts - What The First Fleet Saw.
Unpublished BA Hons thesis, Department of Anthropology,
University of Sydney.
Sherwin, L. and Holmes, G. G. (eds) 1986 Geology of the
Wollongong and Port Hacking 1:100,000 Sheets, 9029, 9129.
Sydney: Geological Survey of New South Wales,
Department of Mineral Resources.
Smith, L. 1988 Aboriginal Site Planning Study in the Sydney
Basin. Stage 1: The Cumberland Plain. Interim Report: Site
Survey and Site Analysis on the Northern Cumberland Plain.
Report prepared for The New South Wales National Parks
and Wildlife Service, Sydney.
Stanbury, P. and Clegg, J. 1990 A Field Guide to Aboriginal Rock
Engravings with Special Reference to those around Sydney.
Sydney: Sydney University Press.
Stockdale, J. 1789[1982] The Voyage of Governor Phillip to
Botany Bay, with an Account of the Establishment of the
Colonies of Port Jackson and Norfolk Island. Facsimile
Edition. Richmond, Victoria: Hutchinson.
Thorpe, W. W. and Stanley, M. S. 1928 Aboriginal axes. Australian
Museum Magazine 3(6):210-211.
White, J. 1790[1962] Journal of a Voyage to New South Wales.
Edited by Alec H. Chisholm, F.R.A.H.S. Sydney: Published
in Association with the Royal Australian Historical Society
by Angus and Robertson.
Whitten, D. G. A. and Brooks, J. R. V. 1972 The Penguin
Dictionary of Geology. Harmondsworth: Penguin Books.
Worgan, G. B. 1788[1978] Journal of a First Fleet Surgeon.
Sydney: The Library Council of New South Wales in
association with Library of Australian History. William
Dixson Foundation Publication 16.
DARWIN ARCHAEOLOGY:
Aboriginal, Asian and European Heritage
of
Australia’s Top End
Edited by
Patricia Bourke, Sally Brockwell and Clayton Fredericksen
Charles Darwin University Press
2005
ISBN 1-876248-98-X
RRP A$34.95
(paperback)
50
SHORT REPORTS
PURPLE HAZE: EVIDENCE FOR A LATER
DATE FOR SOLARIZED AMETHYST GLASS
Samantha Bolton
Archaeology, University of Western Australia, Perth, WA 6009,
Australia. Email: [email protected]
Introduction
On historical sites, artefacts with datable attributes
provide a terminus post or ante quem for a deposit. Glass is
common on 19th and 20th century sites, and is useful as the
period of manufacture can sometimes be determined. One
attribute used to date glass artefacts is colour. However, the
discovery of a piece of solarized amethyst glass with a post1934 trademark casts doubt on the commonly accepted date
of ca.1890-1916 for its manufacture.
The artefact in question was found at Woolgangie town
site, a rock catchment and railhead established in 1895.
Woolgangie is located approximately 450 km east of Perth,
Western Australia, along the transport route and settlement
corridor to the Coolgardie and Kalgoorlie goldfields.
The artefact was found in an assemblage of discarded
material ranging in manufacture from pre-1910 [indicated
by hole-in-cap cans (Rock 1984:105-106)] to post-1921,
[indicated by ceramics marked “MADE IN JAPAN” (Burke
and Smith 2004:371)]. The artefact is a partial base of a
glass vessel. It has a circular cross-section approximately 90
mm in diameter, and was formed using a two-piece mould,
with a post-bottomed base (Boow 1991:45). It was
machine-made and the side is embossed with a raised
diamond pattern. The vessel walls were rounded like a
standard bottle, with a flat panel on at least one side. The
fragment is solarized amethyst, and has an Australian Glass
Manufacturers (AGM) trademark on the base (Fig. 1) and
the characters: “S [undetermined] O 7”.
Figure 1 Australian Glass Manufacturers maker’s mark in
use 1934-1948 (reproduced with permission from Burke and
Smith 2004:370).
Dating the artefact
The attributes that can be used to date this artefact are
the type of mould, the colour and the maker’s mark.
However, they provide conflicting results.
Firstly, the two-piece mould with post-bottomed base is
either semi- or fully-automatically machine made,
manufactured after c.1900-1920 (Boow 1991:46). It has
been suggested that semi-automatic machine made bottles
were not made by AGM until 1918, although this date is not
definitive (Stuart 1993:18).
Secondly, the amethyst colour is produced by the
exposure of manganese dioxide in the glass to ultraviolet
light from the sun (White 1978:66). Manganese dioxide was
added during the manufacturing process to make the glass
clear by counteracting the colours produced by the raw
ingredients (Newman 1977:91,192). This technique was
commonly used at the end of the nineteenth century, as
concerns about the effect of direct light on food subsided
and consumers preferred to see the contents (Vader and
Murray 1975:10; Lockhart in press 2005). The cessation of
manganese as an ingredient in bottle glass is often thought
to have been around 1916, as a consequence of World War
I, and is discussed in further detail below (Vader and
Murray 1975:10; Newman 1970:74; Burke and Smith
2004:369; Casey and Lowe Associates 2000 Appendix 3.5).
Finally, manufacturer’s marks and trademarks are a
common, and often reliable, means of dating artefacts such
as glass and ceramics. The Australian Glass Manufacturers
used several different trademarks over time (Arnold
1985:27). The mark on this artefact (Fig. 1) was used
between 1934-1948 (Arnold 1985:27; Graham 1981:107).
Discussion
The method of manufacture and the trademark suggest
that the artefact was made after World War I, while the
colour dates it to before the war, resulting in a variation of
nearly twenty years. Therefore, it is necessary to re-evaluate
the commonly accepted dates for amethyst glass in
Australia.
The shortage of manganese previously imported from
Germany and the rising costs, are normally given as the
reasons that manganese dioxide stopped being used in glass
during World War I (Kendrick 1964; Miller and Pacey
1985:44). When supplies returned to normal after the war,
these factors no longer would have been an issue. Although
a decrease in supplies had an affect on the use of manganese
in glass, the decline was more closely related to the
introduction of semi and fully automatic machine-made
bottles. Manganese dioxide is more effective in the
manufacture of handblown glass, made using crucibles in
an oxidizing environment, than in machine-made bottles,
manufactured using open tanks and have less oxygen
(Lockhart in press 2005). Hence, the introduction of
automatic manufacturing machines in c.1900-1920 (Boow
1991) had a greater effect on the cessation of the use of
manganese dioxide than did the war. There is no clear cutoff point for the use of manganese, although there is
evidence that it was still in use, albeit on a very small scale,
51
Short Reports
in the United States in 1933 (Lockhart in press 2005) and in
Australia until the 1940s (Carney 2004 pers. com.).
The Australian Glass Manufacturers Company Limited
were incorporated in 1915, following the amalgamation of
the Melbourne Glass Bottle Works Company Limited and
Waterloo Glass Bottle Works Company Limited (Fountain
2000:92; Vader and Murray 1975:14). Therefore any bottles
with an AGM mark were made after 1915, and machinemade bottles after 1918 (Stuart 1993:18). This supports the
post-1934 date for the glass.
Conclusion
Dating solarized amethyst glass has recently come under
discussion (Lockhart in press 2005) and the discovery of
this artefact supports a later date for the use of manganese
dioxide. Another possible explanation is that the particular
AGM trademark was used before 1934, although to date
there is nothing to indicate that this is the case. It is more
likely that solarized amethyst coloured glass, whereas
commonly thought to be manufactured before 1916,
continued into the 1930s.
Acknowledgments
Thanks to Alistair Paterson, Shaun Mackey, Bill
Lockhart and Martin Carney for providing comments and
valuable insights on this paper.
References
Arnold, K. 1985 Collecting Australian Found Bottles: Glass Part
1. Maiden Gully, Victoria: Crown Castleton Publishers.
Boow, J. 1991 Early Australian Commercial Glass: Manufacturing
Processes. Sydney: Heritage Council of New South Wales.
Burke, H. and Smith, C. 2004 The Archaeologist’s Field
Handbook. Sydney: Allen and Unwin.
Carney, M. 2004 personal communication.
Casey and Lowe Associates 2000 Archaeological Investigation of
the Former CSR Site, Pyrmont (Jacksons Landing).
Unpublished report for Lend Lease Development, Sydney.
Fountain, H. 2000 Technology acquisition, firm capability and
sustainable competitive advantage: A case study of
Australian Glass Manufacturers Ltd 1915-39. Business
History 42(3):89-108.
Graham, M. 1981 Australian Glass of the 19th and Early 20th
Century. Sydney: David Ell Press.
Kendrick, G. 1964 The Antique Bottle Collector. Sparks, Nevada:
Western Printing and Publishing Co.
Lockhart, B. in press 2005 The color purple: Dating solarized
amethyst glass containers. Historical Archaeology.
Miller, G. L. and Pacey, A. 1985 Impact of mechanization in the
glass container industry: The Dominion Glass Company of
Montreal, a case study. Historical Archaeology 19:38-50.
Newman, H. 1977 An Illustrated Dictionary of Glass. London:
Thames and Hudson.
Newman, T. S. 1970 A dating key for post-eighteenth century
bottles. Historical Archaeology 4(1):70-75.
Rock, J. T. 1984 Cans in the countryside. Historical Archaeology
18(2):97-111.
Stuart, I. 1993 Bottles for jam? An example of recycling from a
post-contact archaeological site. Australian Archaeology
36:17-21.
Vader, J. and Murray, B. 1975 Antique Bottle Collecting in
Australia. Sydney: Ure Smith.
White, J. R. 1978 Bottle nomenclature: A glossary of landmark
terminology for the archaeologist. Historical Archaeology
12(1):58-67.
52
WALTER BURLEY GRIFFIN AND A MUSEUM
OF ARCHAEOLOGY AT THE HEART OF
AUSTRALIA’S CAPITAL
Sally Brockwell
School of Archaeology and Anthropology, Australian National
University, Canberra, ACT 0200, Australia. Email:
[email protected]
Christopher Chippindale
Cambridge University Museum of Archaeology and
Anthropology, Downing Street, Cambridge CB2 3DZ, England.
Email: [email protected]
An exhibition at the National Archives of Australia in
2002 put on view the remarkable drawings made by Marion
Mahony Griffin for her husband Walter Burley Griffin’s
designs, the designs that won the 1912 competition for
Australia’s new national capital in Canberra. Within those
designs is an archaeological element, not before noticed,
which we report in this note.
Griffin’s designs, though much compromised like all
grand plans for great cities, survive in the shape of Canberra
today. The fundamental axis, from Mount Ainslie through
the War Memorial to the Parliament, is unblemished. There
have been many losses. Instead of the extensive railway
system and its grand central station, there exists today only
a single-platform terminus with just two trains each way a
day to Sydney. Roads for Canberra’s cars now swallow up
Griffin’s park lands by the lake. Other things reported in the
exhibition show that there are unchanging elements in how
architectural and archaeological projects are actually
conducted. Mrs Griffin complained she had only five weeks
to do the drawings, when ‘It isn’t possible to do them in nine
weeks’. (The extra four weeks referred to the travelling time
by ship from San Francisco to Sydney). Completed
somehow just in time, they were taken from Griffin’s
Chicago office towards midnight, to catch the last train,
which would catch the last ship, which would reach
Australia just in time for the competition’s closing date
(Vernon 2002:13). Drawn on linen, and then lithographed
on to heavy cotton, they strikingly reflect conventions of
Japanese wood-block prints – the prints which the Griffins’
employer, Frank Lloyd Wright, collected.
Griffin was more than a designer charged with placing a
defined set of public buildings and ancillary settlements
into an ordered landscape. He largely decided what those
buildings and their surroundings should be, even down to
the species of street trees, as the fitting things that would
define the Australian capital, and thereby the nation. This is
where a place for archaeology came in. An element in
Griffin’s designs were the buildings of high culture, which
he concentrated into a group at the foot of Mount Ainslie,
extending across the flat land to its west. Here would be a
monumental railway terminus, on the European model, and
on the higher ground above it the cathedral, the military
college (the one element which was constructed) and the
opera house (not built, whilst the opera house in Sydney
was, and became famous). Then on the lower ground
beneath and to the west would be three cultural buildings, a
‘Gallery of Plastic Arts’, a ‘Plant Conservatory’ – and the
‘Museum of Archaeology’ (Fig. 1).
Reid’s (2002) account of Griffin’s plans for Canberra
says nothing as to why Canberra needed a Museum of
Short Reports
Mt Ainslie
Casino
Kings Ave Bridge
Cathedral College
Government Depts
Plastic Arts
Parliament House
Archaeology
Conservatory
Capitol
Figure 1 Drawing by Marion Mahony Griffin, 1912: The place of the Museum of Archaeology in the grand design for
Canberra. Detail from Section C-D Easterly Side of Land Axis Ainslie to Red Hill, 1912. Commonwealth of Australia Federal
Capital Competition. National Archives of Australia A710: 44-47.
Archaeology, what it would contain, or what would be the
nature and rôle of it or its neighbouring ‘Gallery of Plastic
Arts’. Still, it is true that a national Museum of Archaeology
– alongside national coats of arms, mottoes, and anthems –
was in that time one of the standard emblems by which
nations defined and identified themselves (as they still so
do today). The drawing of the museum building façade –
heavy, flat, low – has a hint of Mesoamerica to it, reminder
of the fashion for Mayan inspiration which ran through
American architecture at this time (cf. Ingle 1984:13-24). It
is similar to the Carrie Eliza Getty Tomb in Chicago,
designed by Griffin’s philosophical mentor, the legendary
Louis Sullivan, which may have inspired the architecture of
this building. The Mayan influence is also reflected in the
design of Griffin’s most important feature in the new city of
Canberra; the so-called ‘Capitol’ building that was to stand
where the New Parliament House stands today. Griffin’s
vision for this structure referenced Meso-America (or
perhaps Indo-China) with its ziggaraut cap (Vernon
2002:8). One source for the Griffins’ interest in MesoAmerica likely was the replicas of Mayan and Aztec
architecture on display at Chicago’s 1893 World’s
Columbian Exposition. The University of Chicago
launching a number of archaeological expeditions to
Central and South America in the early 20th century also
reflected the widespread interest in this theme at the time.
Reid (2002) reports no record of what the Australian
National Museum of Archaeology might contain. One fears
53
Short Reports
it might have come to hold Greek painted pots and plaster
casts of statues of Roman emperors – emblems of a national
identity created by Australia’s immigrants from Europe, a
national identity whose deep base in the distant past would
be expressed archaeologically by the artefacts of ancient
Europe. If a Classical-centred collection had been acquired
from what was available in the 1920s or later, when the
supply of artefacts of first-rate quality and secure
provenance was much reduced, it could well have been a
second-division affair.
Neither the Museum of Archaeology nor the Gallery of
Plastic Arts ever happened. Whatever specific form the
Griffin vision of a national Museum of Archaeology took
(and whether archaeology would or would not have been of
Australia), the idea fell away along with other aspects of his
grand design for Canberra. When archaeology came to find
its place in an ambitious new museum in Canberra, it took
a very different form in a very different climate and context,
that of a museum defined by its scope being Australia
(archaeology included).
It was not until March 2001 that Canberra and
Australia’s national capital acquired any substantial
archaeological display – in the new National Museum of
Australia. This is not where Griffin’s museum would be, but
on the Acton Peninsula on the other side of the city’s central
lake, named after Burley Griffin, formerly the site of the
Royal Canberra Hospital, and just below the Australian
National University. The university was also a part of
Griffin’s larger vision and built where he planned it. The
former Institute of Anatomy (now Screensound Australia,
housing the National Screen and Sound Archive) was
constructed in 1930 before the Griffins left Australia
(Walter Burley Griffin left in 1935 for India where he died
in 1938. Mrs Griffin subsequently returned to the United
States). It housed a large number of Aboriginal and other
artefacts that have now been relocated to the new National
Museum. What is not known is what became of the
Aboriginal artefacts that were unearthed ca. 1925 while Old
Parliament House was under construction (C. Vernon pers.
comm.).
The National Museum’s design, by Ashton Raggatt
MacDonald and Robert Peck von Hartel Trethowan, reflects
a self-vision of Australia wholly transformed from that of
Griffin’s time. Its construction followed a long period of
doubt in the late 20th century. First, a national museum was
contemplated but not acted on. Then it was created as an
institution but without a physical building in the capital.
Finally, the physical museum was built, opening on the
centenary of Australian federation as a single nation-state.
As nation-states change and seem to weaken as entities in a
changing world, it is less clear that a national museum is a
defining necessity, and more doubt as to what on earth it
might contain. To the specific doubts about Canberra, as a
new city for the Australian capital, and about what the
Australian national identity actually is, are added a broader
loss of confidence in the idea of a national museum.
There is nothing in the new National Museum’s displays
about ancient Europe, and its exhibits include a very
significant element of Aboriginal history. It contains
important major exhibits on the history of ideas in
archaeology, for example, in the Tangled Destinies gallery.
A collection of 80,000 stone tools is one of its significant
collections, and a small portion of these artefacts is well
presented in the permanent display. Both design and
54
exhibits are emblems of a national identity to which
Australia’s Indigenous people are fundamental. Alongside
the many erosions of Griffin’s vision in the Canberra of
today, there remains a modest place for archaeology in the
kind of story the Australian nation chooses to tell of itself.
Acknowledgements
We thank Jay Arthur (National Archives of Australia),
John Mulvaney, Mike Smith (National Museum of
Australia), Christopher Vernon (University of Western
Australia) and an anonymous reviewer for their helpful
comments on this paper. We would also like to thank the
National Archives of Australia for granting permission to
reproduce the image shown as Figure 1.
References
Ingle, M. 1984 The Mayan Revival Style: Art Deco Mayan
Fantasy. Salt Lake City, Utah: Peregrine Smith Books,
Gibbs M. Smith Inc.
Vernon, C. 2002 A Vision Splendid: How the Griffins Imagined
Australia’s Capital. Canberra: National Archives of
Australia.
Reid, P. 2002 Canberra Following Griffin: A Design History of
Australia’s National Capital. Canberra: National Archives of
Australia.
LATE HOLOCENE OCCUPATION AT
BUNNENGALLA 1, MUSSELBROOK CREEK,
NORTHWEST QUEENSLAND
Michael Slack, Richard Fullagar, Andrew
Border, Jackson Diamond and Judith Field
Department of Archaeology, University of Sydney, Sydney,
NSW 2006, Australia. Email: [email protected]
Introduction
Bunnengalla 1 is a sandstone rockshelter fronting onto a
permanent waterhole on Musselbrook Creek located on the
northern margin of Boodjamulla National Park in northwest
Queensland. In July 2004 excavation of the rockshelter
revealed a rich 1.5 m sequence of human occupation. A
preliminary report of the radiocarbon ages and stratigraphic
sequence is presented here. The site provides a record of
Late Holocene occupation from at least 6000 years BP with
a considerable increase in occupation debris from 1300
years BP that is coincident with the amelioration of the
ENSO dominated climate and increased precipitation in
northern Australia.
The Bunnengalla 1 rockshelter is located on Bowthorn
pastoral station, adjacent to Boodjamulla National Park
(formerly Lawn Hill) in northwest Queensland, about 200
km south of the Gulf of Carpentaria. It is approximately 5 km
west of the transition from the Gulf of Carpentaria savanna
plain, within a sandstone gorge system on the eastern margin
of the Barkly Tableland on Musselbrook Creek (refer to Fig.
1). Musselbrook Creek commences on the Barkly Tableland
about 75 km west of the Bunnengalla 1 site and flows into the
Gregory River approximately 100 km to the east. During the
dry season (March – November), it is reduced to a series of
large waterholes along its course. Bunnengalla 1 is adjacent
to one of these waterholes, which is 3 km in length.
Bunnengalla 1 occurs within the traditional country of
the Waanyi People. It is specifically related to the Wogaia
Short Reports
clan, of which Bunnengalla was a traditional owner. The
Musselbrook area is of great spiritual importance to the
Waanyi, although the Bunnengalla 1 site is interpreted by
Waanyi elders as domestic camp site. The site consists of an
extensive shell midden at the base of sandstone escarpment
that forms a small sandstone rockshelter (Fig. 2). A surface
scatter of mussel shell, flaked stone artefacts (including
tulas, scrapers, and points) and a large sandstone
grindstone and a muller were identified in a 50 m2 area
under the escarpment overhang. Rock art in the shelter
consists of a partial bichrome red and yellow rainbow-like
motif; well known from other sites in the region (see
Border 1988, 1989; Slack 2002; Walsh 1985); and a series
of yellow painted emu tracks, and three hand prints.
In July 2004 a 1 m x 1 m test pit was excavated within
the Bunnengalla 1 shelter (Fig. 3a, Fig. 3b). Excavation
followed stratigraphic changes in 5 cm spits to a depth of
99 cm (spit 19). Below this level excavation continued in a
50 cm x 50 cm area in the northern corner of the square to
a depth of 135 cm where steeply sloping sandstone was
encountered. Two stratigraphic units were noted
commencing under surface sediment and an ashy shell
horizon 27 cm below the surface. A compact red/brown
sediment (unit A) occurred in the southwest side of the test
pit, and a loose brown/grey sediment (unit B) in the
northeast corner. All sediments were dry sieved through 4
mm and 2 mm mesh with some preliminary sorting
completed on site.
Radiocarbon determinations
Six radiocarbon determinations were obtained on
charcoal and freshwater mussel shell (including a paired
sample), recovered in situ and from column samples, from
spits 2, 4, 12, and 23. All charcoal samples were pretreated
with an acid/alkali/acid wash. Shell material was surface
etched to remove post depositional calcite. The radiocarbon
dates on paired shell and charcoal samples from unit A
(OZH 613 and OZH 612 respectively) overlap at two
standard deviations. The results of AMS dating are
presented in Table 1 and the age-depth curve for unit B
sediments in Figure 4. The age – depth curve indicates
steady sediment deposition over the last 1300 years in the
northeastern side of the excavation. Sediment accumulation
rates approach 1 cm every 100 years for the entire unit B
Figure 1 Location of the Bunnengalla 1. The map shows Musselbrook Creek flowing east through the center with dotted
lines indicating ‘wet’ season streams. The sandstone escarpment is shaded grey, Late Pleistocene alluvial soils
are unshaded.
55
Short Reports
sequence. The dates from the unit A sediments show a
slower sedimentation rate beginning much earlier than unit
B. Increased seasonal rainfall at Bunnengalla 1 appears to
have scoured the older unit A sediments in the northeastern
side of the excavation, with subsequent accumulation of late
Holocene unit B sediments.
The Bunnengalla deposit contained an abundance of
flaked stone, faunal remains, shell and charcoal. The
highest densities of finds were recorded in unit B sediments
(spits 1-17), which has accumulated over the last 900 years.
Finds recovered below this depth to 135 cm were not well
preserved, particularly the shell which was soft and friable.
The dominant mussel species was Alathyria pertexta and
Velesunio sp. was also identified. The faunal remains
consist of both riverine and terrestrial species, including
small and medium mammals (mostly macropods, and sugar
Figure 2 Photograph of Musselbrook Creek showing location of the Bunnengalla 1 site.
Figure 3a Section drawing of Bunnengalla 1 Test pit 1 showing unit A and unit B sediments.
56
Short Reports
Figure 3b Site plan of Bunnengalla 1 rockshelter showing
location of test pit, extent of surface midden and
surface lithic finds.
glider), turtle, mollusca, crustacea, and fish (perch and
catfish). Stone raw materials include chert (from river
cobbles) and silcrete. The latter is probably sourced in the
lateritic duracrusts associated with nearby mesas on the
savanna plain. Only a very small amount of flaked stone has
been retouched, however two backed artefacts and a tula
have been identified. High concentrations of mussel shell,
crayfish, perch and small mammal remains, mostly small to
medium macropods, imply an economy centered around
locally available resources. The unit A sediments contained
very few faunal remains but more lithic material than the
younger sediments of unit B.
The Bunnengalla 1 site has yielded a sequence
consistent with those reported for Lawn Hill (Hiscock
1988:80). Over the last 3000 years there has been high rates
of sediment accumulation at raised levees associated with
freshwater creeks within sandstone gorge systems. The
rapid build up of sediment during the late Holocene is most
likely associated with increased local precipitation after the
extreme El Niño/Southern Oscillation (ENSO) cycles of the
mid to late Holocene in northern Australia had concluded at
about 3500 years BP (Hope and Golson 1995; Kershaw
1995; McCarthy and Head 2001). Increased local rainfall in
this area has probably led to seasonal inundation of the
Musselbrook Creek since at least 1000 years BP (see
Schulmeister and Lees 1992). It is likely that flooding has
resulted in rapid sediment deposition in levees along the
creeks in all local sandstone gorges, such as at Bunnengalla
Sample
Unit A
Mussel Shell
Charcoal
Charcoal
Unit B
Charcoal
Charcoal
Charcoal
Figure 4 Age – depth curve for unit B sediments, testpit 1
at Bunnengalla 1, northwest Queensland,
showing steady accumulation of sediments over
the last 1300 years.
1. We argue that a reconfiguration of Aboriginal settlement
of these areas followed the amelioration of ENSO
conditions after 3500 years BP as larger permanent water
supplies became available. This phase is represented at
Bunnengalla 1 by the unit B sediments. Further analysis of
the Bunnengalla 1 sequence and other sites in similar
environments will provide a high resolution picture of
occupation patterns and resource use during the mid to late
Holocene period.
Acknowledgements
The project was funded by an Australian Research
Council Linkage Grant (LP0211430) to J. Field, A. Border
and M. Archer and an Australian Institute of Aboriginal and
Torres Strait Islander Studies (AIATSIS) grant to M. Slack.
It was undertaken in collaboration with the Waanyi people.
Fieldwork was facilitated with the support of the Queensland
Parks and Wildlife Service and the Queensland EPA.
References
Border, A. 1988 An Archaeological Survey in the Lawn Hill and
Musselbrook Creek Area, Northwestern Queensland
Highlands, Unpublished Report, Queensland Department of
Community Services, Brisbane.
Border, A. 1989 An Archaeological Survey on Lawn Hill and
Bowthorn Stations, Northwestern Queensland. Unpublished
Report to the Heritage Section, Queensland Department of
Environment and Heritage, Brisbane.
Spit
Depth below
surface (cm)
Sample ID
Lab Number
d13C
AMS
Radiocarbon Age
4
4
12
15.0
15.0
56.5
BUNN1TP14CS1A
BUNN1TP12 4A
BUNN1TP11212A
OZH613
OZH612
OZH614
-10.6 ‰
-27.0 ‰
-10.0 ‰
2820 ± 50 BP
2970 ± 60 BP
6120 ± 60 BP
2
11
23
9.6
51.5
119.0
BUNN1TP12
BUNN1TP111B
BUNN1TP123B
Beta 194878
Beta 194876
Beta 194877
-24.0 ‰
-26.2 ‰
-26.2 ‰
200 ± 50 BP
660 ± 40 BP
1340 ± 40 BP
Table 1 AMS radiocarbon determinations for spits 2, 4, 11,12 and 23 from units A and B at Bunnengalla 1 shelter,
Musselbrook, northwest Queensland.
57
Short Reports
Hiscock, P. D. 1988 Prehistoric Settlement Patterns and Artefact
Manufacture at Lawn Hill, Northwest Queensland.
Unpublished PhD thesis, University of Queensland,
Brisbane.
Hope, G. and Golson, J. 1995 Late Quaternary change in the
mountains of New Guinea Antiquity 69:818-830.
Kershaw, A. P. 1995 Environmental change in Greater Australia.
Antiquity 69:656-676.
McCarthy, L. and Head, L. 2001 Holocene variability in semi-arid
vegetation: New evidence from Leporillus middens from the
Flinders Ranges, South Australia. The Holocene 11:681689.
Schulmeister, J. and Lees, B. 1992 Morphology and
chronostratigraphy of a coastal dunefield; Groote Eylandt,
northern Australia. Geomorphology 5:521-534.
Slack, M. J. 2002 Aboriginal Responses to European contact in
northwestern Queensland. Unpublished MA thesis,
Australian National University, Canberra.
Walsh, G. L. 1985 The Archaeological Significance of Lawn Hill
Gorge. Unpublished Report for Queensland Parks and
Wildlife Service, Brisbane.
one piece of modified spinifex resin were located on the
rock-shelter floor.
Six thin timber pieces appear to have been heat
straightened and are partially burnt; four of the six have
pointed ends. The spinifex resin has four elongate linear
marks on its surface similar to finger marks. The pointed
timber pieces, modified spinifex resin and the stone artefact
are consistent with ethnographic accounts describing spear
manufacture and resultant debris (Hayden 1981:76, Plate
16). Furthermore the pointed timber pieces indicate that
composite spears may have been made at this site (Bindon
and Lofgren 1982:116-123; Davidson 1935).
A piece of straightened timber and a fragment of
modified spinifex resin were submitted for radiometric
analysis. The results suggest that this site was used from
around 1000 BP to 350 BP (see Table 1). A number of items
were salvaged from the site, conserved and mounted in a
display case, which is to be returned to the relevant
communities.
A NOTE ON RADIOCARBON DATES FROM
THE PARABURDOO, MOUNT BROCKMAN
AND YANDICOOGINA AREAS OF THE
HAMERSLEY PLATEAU, PILBARA,
WESTERN AUSTRALIA
CME-A-30
This site is a large rock-shelter near the top of the
Channar Ranges. A dead tree trunk (Acacia sp.), trimmed of
peripheral branches was located in the shelter’s rear. The
trunk is adjacent to two rock-shelves/niches 1.5 m above the
shelter floor and appears to have been deliberately
positioned as a ladder. One wood sample from the tree trunk
was submitted for radiometric analysis, which resulted in an
age estimation of 440 ± 70 BP. (Table 1).
Bruce Veitch and Fiona Hook
Archae-Aus Pty Ltd, PO Box 177, South Fremantle, WA 6162,
Australia.
Elizabeth Bradshaw
Principal Advisor Cultural Heritage, Community Relations,
Rio Tinto Ltd., 55 Collins Street, Melbourne, VIC 3000,
Australia.
Introduction
Excavations have been carried out at a number of rockshelters in the Pilbara region of Western Australia as part of
mitigative salvage work and research supported by
Hamersley Iron Pty Ltd (Fig. 1). The excavations
commenced in 1998 and are ongoing. At all times, the work
was conducted with participation of and supervision from
the relevant Aboriginal people. These people are the
Innawonga in the Paraburdoo area, the Gurama in the
Mount Brockman area and the Bandjima and Niabarli in
the Yandicoogina area. Their contributions to the research
are gratefully acknowledged.
Thus far, 24 radiocarbon dates have been obtained from
10 rock-shelters (Table 1) in the iron rich gorges and gullies
of the Hamersley Plateau. The dates obtained range from
the Pleistocene through to the late Holocene. Most dates,
however, are confined to the last 2000 years. Although
some of the work is ongoing, this short report has been
prepared to provide access to the radiocarbon dates and a
brief description of each site beyond the grey literature of
consulting reports. A more detailed article will be
forthcoming.
Paraburdoo
CME-A-18
This site is a large rock-shelter near the top of the
Channar Ranges. Large quantities of burnt timber, some of
which has been straightened, one flaked stone artefact and
58
Yirra
CME-A-31
The site is a large rock-shelter in a low gully near a large
riverbed in the Channar Ranges. A 1 m by 1 m test-pit was
excavated at this site with bedrock encountered at 79 cm.
Flaked stone artefacts, burnt plant remains and hearth
features were noted throughout the deposit (Veitch et al.
2004) (Table 1).
Preliminary results obtained indicate several artefact
and charcoal concentrations. Of significant interest is the
lowest concentration in the depth range of 43.13 to 51.50
cm below the surface, which is bracketed by the dates of
16,950 ± 90 BP and 19,270 ± 140 BP (from depth range of
38.63 and 55.80 cm below surface respectively). This
concentration is consistent with the use of the Hamersley
Plateau as a refugium during the Last Glacial Maximum
(LGM) (Hiscock 1988; Smith 1988; Thorley 1998; Veth
1989, 1993). Above this concentration, there are two others
that may relate to post- glacial amelioration and later
Holocene developments (e.g. Smith 1988; Veitch et al.
2003; Veth 1989, 1993). These alternatives will be more
fully developed in a later publication.
There is an obvious problem with the anomalously
young age returned for Wk-12536. This result is believed to
be owing to limited bioturbation by ants (eg. Hussey 2002).
This matter will be taken up in more detail in a future article
dealing with aspects of bioturbation. Given that all other
dates are in sequence, it is concluded that the level of
bioturbation in this deposit has been minimal. Additionally,
Hamersley Iron has undertaken to support further
excavations at this site and to avoid it.
ERP-04
Site ERP-04 is a small rock-shelter located near the
Short Reports
Figure 1 Map showing location of the Pilbara region of Western Australia, and sites mentioned in text.
summit of the Eastern Ranges. A total of nine flaked stone
artefacts and one ironstone millstone are located on the
shelter floor. A 50 cm x 50 cm test-pit was excavated to
bedrock reaching a depth of 18.6 cm. The excavated deposit
contained one hearth feature and four flaked stone artefacts.
A charcoal sample from the hearth feature was submitted
for radiometric analysis, which returned an estimation of
2000 ± 50 BP (Table 1).
ERP-11B(ii)
This site is a large rock-shelter within the Eastern
Ranges. A scatter of 61 artefacts was located on the rockshelter surface. A 1.0 m x 0.50 m test-pit was excavated to
bedrock reaching a depth of 20.1 cm. A total of 100 flaked
stone artefacts were recovered. Only scattered charcoal was
present within the deposit. Two samples of charcoal were
submitted for radiocarbon determinations, which suggest
very recent use (i.e. from around 370 ± 50 BP and 350 ± 50
BP, Table 1). A further excavation measuring 2.5 m by 0.5
m was conducted in October 2002. The analysis of the more
recently excavated material has yet to be completed.
ERP-15
This site is a medium sized rock-shelter among a series
of steep gullies in the Eastern Ranges. The shelter contains
a small surface assemblage of flaked stone artefacts and
manuports. Two 50 cm x 50 cm test-pits were excavated to
bedrock reaching a depth of 15.8 cm and 19.3 cm
respectively. No flaked stone artefacts were recovered from
either test-pit but one charcoal concentration, interpreted as
a hearth, was encountered in test-pit 1 at a depth of 3.0 cm.
A charcoal sample from this hearth returned an age
estimation of 950 ± 50 BP (Table 1).
ERP-22
ERP-22 is a small rock-shelter containing a hearth on the
surface of the shelter floor. No artefacts of any kind were
noted on the surface of the deposit. The site was excavated
to determine if it constituted a site as defined under the WA
Aboriginal Heritage Act 1972. Three 50 cm x 50 cm test-pits
were excavated in the rock-shelter. With the exception of the
hearth, no cultural features or artefacts were encountered.
Charcoal was recovered from the upper layers of test-pit 1,
where the hearth feature was located. One sample of
charcoal from test-pit 1 was submitted for radiometric
analysis and returned an age estimation of 560 ± 50 BP.
(Table 1).
ERP-26
This site is a medium sized rock-shelter near the base of
a shallow north/south oriented gully. A single quartzite
muller was recorded within the shelter, on a rock shelf at the
rear centre of the shelter. In addition two manuports, both
large flat ironstone slabs interpreted as anvils, were
recorded on the shelter floor.
Initially, a single 50 cm x 50 cm test-pit was excavated
to a depth of 24 cm. This test-pit was not excavated to
bedrock because a relatively large number of flaked stone
artefacts and a hearth feature were encountered. These were
considered sufficient to evaluate the archaeological
potential of the site, as specified by the Excavation Permit.
A total of 28 flaked stone artefacts were recovered. A
distinct hearth feature was also encountered. A charcoal
sample from this feature was submitted for radiometric
analysis and resulted in an age estimation of 1840 ± 50 BP
(Table 1).
Hamersley Iron has undertaken to avoid this site. At the
59
Short Reports
Site Field
Code
Depth
Below
Surface
(cm)
Conventional
Age (BP)
Lab
Number
Material
Dated
Feature
Report
Unpublished
Area
CME-A-18
CME-A-18
CME-A-30
CME-A-31
CME-A-31
CME-A-31
CME-A-31
CME-A-31
Surface
Surface
Surface
4.5
10.7
22.1
24.00-29.00
29.00-34.00
390 ± 50
1000 ± 60
440 ± 70
1307 ± 40
1610 ± 50
3230 ± 50
3702 ± 56
1057 ± 45
Wk-8002
Wk-8001
Wk-8955
Wk12534
Wk-9147
Wk-8953
Wk-12535
Wk-12536
Spear Point
Spinifex ball
Tree ladder
no feature
Hearth
Hearth
No feature
No feature
(Veitch et al. 2001)
(Veitch et al. in prep)
P
P
P
P
P
P
P
P
CME-A-31
CME-A-31
CME-A-31
CME-A-31
ERP-04
ERP-11b(ii)
ERP-11b(ii)
29.00-34.00
34.50-36.00
43.1
55.1
3.9-7.5
4.0
6.
10,628 ± 74
13,375 ± 94
16,950 ± 90
19,270 ± 140
2000 ± 50
350 ± 50
370 ± 50
Wk-13779
Wk-12537
Wk-9148
Wk-8954
Wk-6325
Wk-6322
Wk-6326
Wood
Spinifex Resin
Wood
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
(Corymbia sp)
3 mm sieve
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
Charcoal, in situ
(Hook et al. 1998)
(Hook et al. 1998)
P
P
P
P
P
P
P
ERP-15
ERP-22
ERP-26
BM99-10
3.1
7.5
10.0
Surface
950 ± 50
560 ± 50
1840 ± 50
540 ± 50
Wk-6323
Wk-6324
Wk-6327
Wk-7099
Charcoal, in situ
Charcoal, in situ
Charcoal
Spinifex Resin
No feature
Hearth
Hearth
Hearth
Hearth
Hearth
Scattered
Charcoal
Hearth
Hearth
Hearth
Spinifex ball
P
P
P
B
BM99-10
3.0
470 ± 50
Wk-8000
Charcoal, in situ
Hearth
Y97-28
Surface
810 ± 80
Wk-5605
Shell (Melo sp.)
Y
Y97-28
14.7
modern
Wk-5603
Charcoal, in situ
(Hook et al. 2000)
Y
Y97-28
21.5
modern
Wk-5604
Charcoal, in situ
(Hook et al. 2000)
Y
Y97-28
14.3
380 ± 80
Wk-5627
Charcoal, in situ
Water carrier
fragment
Scattered
charcoal
Scattered
charcoal
Scattered
charcoal
(Hook et al. 1998)
(Veitch and
Di Lello 2000)
(Veitch and
Di Lello 2000)
(Hook et al. 2000)
(Hook et al. 2000)
Y
B
P = Paraburdoo; B = Mt Brockman; Y = Yandicoogina
Table 1 Radiocarbon dates from sites in Paraburdoo, Mt Brockman and Yandicoogina, Western Australia.
request of the Aboriginal community the excavation was
expanded and continued to bedrock in October 2002. The
deposit reached a depth of 65 cm and was found to contain
one hearth and flaked stone artefacts. The analysis of the
more recently excavated material has yet to be completed.
Little charcoal was preserved beneath the hearth and
consequently no further radiometric determinations have
been obtained.
Mount Brockman
BM99-10
This site is a small rock-shelter, the surface of which
contained a considerable number of flaked stone artefacts
and two balls of modified spinifex resin. A 1.0 m x 0.50 m
area was excavated to bedrock to a depth of 8 cm.
Relatively large amounts of charcoal were recovered; one
sample from a hearth feature was submitted for radiometric
analysis which resulted in an age estimation of 470 ± 50 BP
(Table 1). A total of 138 flaked stone artefacts were
identified. A radiometric date of 540 ± 50 BP was also
obtained from one of the spinifex resin balls (Table 1).
60
Yandicoogina
Y97-28
This site is a medium sized rock-shelter which had a
large surface scatter of flaked stone artefacts, a high
proportion of grinding material and a fragment of baler
shell (Melo sp.). The presence of a large amount of grinding
material in this rock-shelter indicated its use for either the
storing of grindstones or for intensive seed grinding.
Two test-pits, measuring 1 m x 1 m and 0.5 m x 0.5 m
respectively, were excavated to bedrock reaching depths of
45 cm and 17 cm. Both squares contained abundant flaked
stone artefacts and scattered charcoal but no hearth
features. Test-pit 1 contained 272 flaked stone artefacts and
test pit 2 contained 73 artefacts. Four samples of charcoal
and shell were submitted for radiometric analysis. The two
dates on charcoal from test-pit 1 returned modern
estimations. The charcoal sample from test-pit 2 resulted in
an age estimation of 380 ± 80 BP and the surface baler shell
sample an estimation of 810 ± 80 BP. When compared to the
dates from test-pit 2 and the surface, the estimations from
test-pit 1 appear anomalous. Notwithstanding this problem
Short Reports
with test-pit 1, the dates from test-pit 2 and the surface
suggests relatively recent use of the shelter (Table 1).
Discussion
The age estimations from Yirra (CME-A-31) suggest
relatively intense occupation over the LGM, and during
subsequent periods. It is not clear at present whether the data
from this site demonstrate relatively continuous use from the
end of the LGM to the later Holocene, or reflect sample size
phenomena (cf. Jerardino and Yates 1997; Lyman 1987;
O’Neil 1993). It is hoped that further excavation, dating and
analysis at this site will clarify these matters.
In addition there is an abundance of material from other
sites covering the period from approximately 2000 BP to the
last few hundred years. Further detail and analysis on these
excavations will be provided in a more comprehensive
article to follow.
References
Bindon, P. and Lofgren, M. 1982 Walled rock shelters and a
cached spear in the Pilbara region, Western Australia.
Records of the Western Australian Museum 10:111-126.
Davidson, D. S. 1935 Australian spear-traits and their derivations.
Journal of the Polynesian Society 34:41-72, 143-162.
Hayden, B. 1981 Palaeolithic Reflections. Lithic Technology and
Ethnographic Excavations among Australian Aborigines.
Canberra: Australian Institute of Aboriginal Studies. AIAS
New Series 5.
Hiscock, P. 1988 Prehistoric Settlement and Artefact Manufacture
at Lawn Hill, North-east Queensland. Unpublished PhD
Thesis, University of Queensland, Brisbane.
Hook, F., Jackson, G., Martin, C. and Veitch, B. 1998 The report
of the test excavation of 11 Aboriginal archaeological sites,
in the Eastern Ranges project area, Paraburdoo, Pilbara,
Western Australia. Hamersley Iron Pty Ltd and Archae-Aus.
Hook, F., Veitch, B. and Martin, C. 2000 A report of an Aboriginal
archaeological salvage program of sites located on the HI
Yandi railway route and mine area, Hamersley Plateau,
Western Australia. Volume One. Hamersley Iron Pty Ltd and
Archae-Aus.
Hussey, B. M. J. 2002 Wattle I plant for wildlife. Conservation
Science W. Aust 4:62-71.
Jerardino, A. and Yates, R. 1997 Excavations at Mike Taylor’s
midden: A summary report and implications for a recharacterisation of megamiddens. South African
Archaeological Bulletin 52:43-51.
Lyman, R. L. 1987 On the analysis of vertebrate mortality
profiles: Sample size, mortality type, and hunting pressure.
O’Neil, D. H. 1993 Excavation sample size: A cautionary tale.
American Antiquity 58: 523-529.
Smith, M. A. 1988 The Pattern and Timing of Prehistoric
Settlement in Central Australia. Unpublished PhD thesis,
University of New England, Armidale.
Thorley, P. B. 1998 Pleistocene settlement in the Australian arid
zone: Occupation of an inland riverine landscape in the
central Australian ranges. Antiquity 72:34-45.
Veitch, B. and Di Lello, A. 2000 The report of the test excavation
of five Aboriginal archaeological sites in the proposed pit 4
and 6 extension areas situated within the Brockman Mine,
Pilbara, Western Australia. Hamersley Iron Pty Ltd and
Archae-Aus.
Veitch, B., Hook, F. and Fry, R. 2001 A second addendum report
to: Hook, F. and Veitch, B. 1999. The report of an
Aboriginal Heritage Assessment of the proposed Channar
Mine Extension Areas, near Paraburdoo, Western Australia
- Salvage Program. Hamersley Iron Pty Ltd and ArchaeAus.
Veitch, B., Hook, F. and Martin, C. 2004. A third Addendum to
Hook, F. and Veitch, B. 1999. The report of an Aboriginal
Heritage Assessment of the Proposed Channar Mine
Extension Areas, near Paraburdoo, Western Australia November 2000 Salvage Excavations. Archae-Aus.
Veitch, B., Hook, F. and Martin, C. in prep A fourth Addendum to
Hook, F. and Veitch, B. 1999. The report of an Aboriginal
Heritage Assessment of the Proposed Channar Mine
Extension Areas, near Paraburdoo, Western Australia –
November 2000 Salvage Excavations. Archae-Aus.
Veitch, B., Hook, F. Warren, L. and Spooner, N. 2003 Relocation
and Preliminary Optical Dating of the Gurdadguji Stone
Arrangements in the Pilbara region of Western Australia.
Paper presented to the Fifth World Archaeological Congress,
Washington D.C.
Veth, P. M. 1989 The Prehistory of the Sandy Deserts: Spatial and
Temporal Variation in Settlement and Subsistence Behaviour
within the Arid Zone of Australia. Unpublished PhD thesis,
University of Western Australia, Perth.
Veth, P. M. 1993 Islands in the Interior: The Dynamics of
Prehistoric Adaptations within the Arid Zone of Australia.
Ann Arbor: University of Michigan. International
Monographs in Prehistory, Archaeological Series 3.
INDIGENOUS ARCHAEOLOGICAL SITES
AND THE BLACK SWAMP FOSSIL BED:
ROCKY RIVER PRECINCT, FLINDERS
CHASE NATIONAL PARK, KANGAROO
ISLAND, SOUTH AUSTRALIA
Keryn Walshe
School of Humanities, Flinders University, Adelaide, SA
5001, Australia. Email: [email protected]
Introduction
Fossil evidence for extinct megafauna at Black Swamp,
Rocky River in the Flinders Chase National Park,
Kangaroo Island was first noted in 1908 by C.J. May, then
caretaker of the Rocky River Reserve (Tindale et al. 1935).
Formal palaeontological investigation of the fossil area was
initiated during a visit to the former Flinders Chase Flora
and Fauna Reserve by Norman Tindale in late 1934
(Tindale et al. 1935; Tindale 1937a, 1937b). Further
interest was not rekindled until the late 1970’s (Hope et al.
1977). Since 1995 however, palaeontological investigations
have been intensely focused on the Black Swamp fossil site
(Wells et al.1997; Thammakhantry 1998; Dalgairns 1999).
Archaeological evidence from Kangaroo Island, South
Australia generally suggests occupation by Indigenous
people between about 16,000 and 4000 years ago (Lampert
1981, 1983). Archaeological sites and objects have been
recorded at Black Swamp, Rocky River and the intriguing
question of coalescence between megafuana and
Indigenous people was initially posited by Tindale in the
1930’s (Tindale 1937a, 1937b). This same question has
‘shadowed’ later palaeontological investigations at Black
Swamp, Rocky River but irrefutable physical evidence
remains as elusive here as it does for the vast majority of
megafaunal sites on the mainland.
Archaeological finds on Kangaroo Island have been
reported since 1903 (Howchin 1903) but investigations
were concentrated principally on the eastern part of the
island. It was not until the 1980’s that archaeological
investigations were undertaken, albeit brief in time at Black
61
Swamp (Draper 1991, 1992). Recently, as part of an
ongoing and joint palaeontological and archaeological
research program in Flinders Chase National Park, formal
surveys have taken place around Black Swamp.
Previous investigations into Indigenous archaeology at
Black Swamp
Kangaroo Island was deemed unoccupied by Indigenous
people following Matthew Flinders’ 1802 report on the
seeming absence of human habitation. It was thus of some
surprise to the scientific community when early and
significant archaeological investigations on the Island
identified evidence for extensive occupation (Howchin
1903; Tindale and Maegraith 1931; Cooper 1943, 1960).
The majority of early archaeological investigations were
concentrated at the eastern portion of the island. It was the
investigation of megafaunal fossil bones at Black Swamp
that prompted preliminary archaeological investigation of
the western area in 1934. The fossil excavation by Tindale,
Fenner and Hall (1935) failed however, to yield
archaeological material (Tindale 1937). A single
hammerstone from Rocky River was reported by Tindale
(1937:43) as having been collected somewhere south of the
fossil bed by a former caretaker. Tindale (1937) notes that
ploughing on the neighboring flats had revealed bones, but
not implements.
In 1988 road works at the entrance to the Rocky River
precinct revealed five stone tool surface scatters on a
deflation surface opposite ‘May’s Cottage’ and
approximately 500 m east of the known Black Swamp fossil
bed (Fig. 1). Five one meter square trenches of variable
depths (0.9 m to 1.15 m) were excavated into this deflation
surface by archaeologist Neale Draper. Draper reported that
the excavations yielded 289 (mostly quartz) tools and some
charcoal samples (1991). Two of the charcoal samples were
submitted for radiocarbon dating and returned dates of 2340
± 130 BP (SUA2835, Draper 1991:684) and 400 ± 50 BP
(Beta 30176, Draper 1991:684).
The radiocarbon dates reported by Draper (1991) are
significantly more recent than those obtained by
archaeologist Ron Lampert from Seton Rockshelter and
numerous open sites on Kangaroo Island. Lampert (1981,
1983) had previously placed Kangaroo Island occupation
between about 16,000 and 4000 years ago. Draper also took
the opportunity in 1988 to excavate a former borrow pit,
thought to be the location of Tindale’s 1934 excavation for
megafaunal material. The face of this borrow pit was first
cleaned back along a 3.5 m section then excavated back 25
cm but did not recover any animal bones (Draper 1989,
1991). Two quartz blades or flakes and one ‘simple,
quartzite cobble chopper’ were however recovered (Draper
1989, 1991:690). Charcoal recovered from just below the
cobble chopper, at a depth of about 55 cm, yielded a date of
1280 ± 140 BP (SUA2836). Charcoal associated with the
quartz material yielded a date of 1380 ± 80 BP (Beta
30175). As Draper (1991) points out, this particular
excavation did not reveal a direct association between
megafaunal material and stone tools, but again suggests a
more recent occupation date than found elsewhere on the
island. Draper also excavated a 1x1 m trench intended to
reveal more information about the composition of the
calcareous dunes (Draper 1989). The trench reached 60 cm
and yielded two quartz artefacts before a layer of hard
limestone was encountered (Draper 1989).
62
Figure 1 Upper photo is view looking to the west showing
turnoff to the May Homestead and fossil area,
approximately 500 m away. Blowout in dune just
visible opposite T junction. Lower photo, closer
view showing blowout in dune opposite the road
junction and area of stone tool scatter, originally
excavated by Draper (1991).
In 1996-7, further archaeological survey and excavation
was carried out in the same vicinity and on a high dune
above Black Creek Swamp under supervision by then postgraduate candidate Heather Builth, in turn supervised by
Neale Draper. No formal field report has been presented
from this work but it is understood that no samples for
dating were obtained and no collections were made.
The area of surface scatters reported by Draper (1991)
was revisited in 1998 by Marin and Hodgson (1998) as part
of a broader cultural heritage survey of select areas within
the Flinders Chase National Park. Marin and Hodgson
(1998) were unable to re-locate the scatters identified by
Draper some ten years earlier but they recorded three
previously unidentified, low density artefact scatters around
the visitor centre infrastructure. Two of the three recently
recorded sites consisted entirely of a few pieces of quartz
whilst the remaining scatter consisted of one quartz and two
chert tools. These sites were all recorded on higher ground
around the southern side of Black Swamp, between
approximately 120 and 400 m from the known fossil site
(Marin and Hodgson 1998).
Inspection of the same area in 2000 and 2001 by this
author revealed stone tools scattered over the deflation
surface originally reported by Draper (1991) but failed to
Short Reports
identify the three discrete, low density scatters reported by
Marin and Hodgson (1998). The exposed, de-vegetated
surfaces around the southern margins of the swamp allow
significant water runoff and it is evident that discrete, low
density quartz scatters are subject to significant and rapid
degradation. The question of loss arises and the resultant
impact (if any) on site recording. Sensibly, if the swamp
margins were densely covered with site material then
presumably small, discrete sites would be regularly
identified. This is simply not the case. No artefact scatters
or other site material has been reported directly on or along
the margins of Black Creek Swamp proper.
A feature “…consisting of a series of fire blackened
cobble-like stones of up to 100 mm in length, set in the
ground, in an approximate circular shape with charcoal
near some of the stone and appearing to extrude from the
soil” (Marin 1999:4) was located approximately 6 km
south of the Rocky River Precinct. This feature was
initially considered to be an Aboriginal hearth and was
excavated in 1999. However, Marin (1999) reported that
after excavation, the feature was considered highly
unlikely to be of Aboriginal origin. Charcoal samples were
collected for dating, but no further information has
become public.
Discussion
The archaeological finds to date at Black Swamp,
Rocky River consist mostly of quartz artefact scatters, with
some quartzite and chert tools and a few isolated pebble
choppers near to but not in association with the megafaunal
fossil bed.
As part of an integrated palaeontological and
archaeological investigation at Rocky River, Flinders
Chase, a comprehensive surface survey of the swamp and
its surroundings was carried out at intervals, between 2001
and 2003. This survey recorded two pebble choppers and
numerous outcrops of ‘vein’ quartz. As described by Cooper
(1960) vein quartz is highly abundant on the island and
offers a serviceable raw material source. However, the
challenge for site recording is distinguishing between
modified and unmodified quartz fragments. The recently
found pebble choppers are similar to those described by
Tindale (1937) and highly characteristic of those collected
extensively by Cooper (1960). The small grain size of the
pebble gives a smooth finish and high density. The closest
source of such fine grained, high density pebbles is yet to be
identified. Inspection of all accessible river mouths and
beaches both east and west of Rocky River have so far
failed to locate a source of such fine grained pebbles.
‘Pebble surveys’ were recently extended along almost the
entire western and northern coastlines of Kangaroo Island,
by following deep cut river channels into the coast and then
along adjacent coastal margins but these too, failed to locate
that particular fine grained, dark and green hued stone.
The charcoal dated to about 2340 years old was found in
association with quartz artefacts and taken from a depth of
approximately 90 cm, just above a sterile clay unit. As
regards the more recent sample however, Draper (1991:684)
states that it “… came from charcoal collected from
excavation unit #5, 55-75 cm below the surface (not
associated with archaeological material) ….which may be
too recent to be considered a finite dating determination”.
Draper (1991) also notes freshwater mussel shell and
mammal bone including kangaroo and possibly ring-tail
possum remains in association with the same charcoal
sample. Fifteen ring tail possums were introduced onto
Kangaroo Island in 1926 (Robinson and Armstrong
1999:188) suggesting either a relict possum population or
modern contamination of the archaeological deposit.
The Rocky River Precinct artefact scatters as described
by Draper (1991) are significantly disturbed due long term
ploughing, borrow pit activity, road works, natural erosion
processes and prior collecting. Draper (1991:692) reported
that “…The upper half of test pit 2 …is disturbed and
contains pieces of metal in the three upper units…”. The
dates reported by Draper in 1991 need to carefully
evaluated against the disturbed contexts from which they
were derived and in light of revelations on problems in
earlier radio carbon dating outcomes (Roberts et al. 1994).
Conclusion
Archaeological surveys between 1934 and 2003 at
Black Swamp, Rocky River have identified a number of
stone (mostly quartz) tool scatters and a few isolated pebble
choppers. Excavations between 1934 and 1988 have
revealed further stone tools and a pebble chopper to a depth
of about half a meter. Charcoal samples have returned
relatively recent dates for occupation of this part of the
island. Excavation has also revealed highly disturbed
contexts and has raised issues of validity for these same
dates. So far no physical evidence for coalescence between
ancient megafauna and Indigenous people has been found
in either archaeological or palaeontological excavation.
Absence at this stage should not however be read as
improbable. The age of the fossil bed is yet to be
definitively stated and dates for Indigenous occupation of
Kangaroo Island generally are yet to be verified. The
context of the cultural and biological landscape is still
poorly understood and interpretative work is preliminary.
Of primary concern to the archaeological program at this
point is the cultural and environmental context of visitation
to and departure from the island by Indigenous people. The
question of Indigenous interaction with megafaunal animals
is indisputably intriguing, but of lesser interest at this stage
than the primary issues of understanding a complex cultural
landscape and its connection to a larger, similarly complex
land mass. If visitation and/or occupation of the island were
restricted to times of low sea level (during the last glacial
maximum) then there is also Indigenous response to
regional climatic change awaiting interpretation.
References
Cooper, H. M. 1943 Large stone implements from South Australia.
Records of the South Australia Museum 7:343-369
Cooper, H. M. 1960 The archaeology of Kangaroo Island, South
Australia. Records of the South Australia Museum 13:481503.
Dalgairns, S. N. 1999 A paleoenvironmental and taphonomic
review of the late
Pleistocene swamp site at Rocky River, Kangaroo Island, South
Australia. Unpublished Honours Thesis, School of
Biological Sciences, Flinders University, Adelaide, South
Australia.
Draper, N. 1989 Report on Archaeological Fieldwork, Kangaroo
Island. Unpublished Report to Conservation and Land
Management Division, South Australia, Adelaide.
Draper, N. 1991 Cape Du Couedic Rockshelter and the Aboriginal
Archaeology of Kangaroo Island, South Australia. PhD
thesis, Department of Anthropology, University New
Mexico, Albuquerque.
63
Gunumbah: Archaelogical and Aboriginal meanings
Draper, N. 1992 The history of Aboriginal land use on Kangaroo
Island. In A. C. Robinson and D. M. Armstrong (eds)
Biological Survey of Kangaroo Island, South Australia 1989
and 1990, pp. 33-46. Adelaide: Heritage and Biodiversity
Section, Department for Environment, Heritage and
Aboriginal Affairs.
Hope, G. S., Clark, R. L. and Hope, J. H. (1977) Report on a
stratigraphic investigation of a fossil bone deposit and its
relationship to Black Creek Swamp at Rocky River, Flinders
Chase National Park, Kangaroo Island South Australia.
Unpublished Report to National Parks and Wildlife Service,
Adelaide.
Howchin, W. 1903 Aboriginal occupation of Kangaroo Island.
Transcripts of the Royal Society of South Australia 27(1):90.
Lampert, R. J. 1972 A carbon date for the Aboriginal occupation
of Kangaroo Island, South Australia. Mankind 8:223-229.
Lampert, R. J. 1975 Preliminary report on some waisted blades
found on Kangaroo Island, South Australia. Australian
Archaeology 2:45-48.
Lampert, R. J. 1979 Aborigines. In M. J. Tyler, C. R. Twidale, and
J. K. Ling (eds) Natural History of Kangaroo Island, pp. 8190. Adelaide: Royal Society of South Australia.
Lampert, R. J. 1981 The Great Kartan Mystery. Canberra:
Research School of Pacific Studies, Australian National
University. Terra Australis 5.
Lampert, R. J. 1983 Kangaroo Island 18 ± 2 KA. In J.M.A.
Chappell and A. Grinrod (eds) CLIMANZ. A Symposium of
Results and Discussions Concerned with Late Quaternary
Climatic History of Australia, New Zealand and
Surrounding Seas, pp. 63. Canberra: Research School of
Pacific Studies, Australian National University.
Marin, P. and Hodgson, G. 1998 Cultural Heritage Survey:
Flinders Chase National Park, Kangaroo Island. Covering
Admirals Arch, Cape Du Couedic, Rocky River.
South Australia, Adelaide.
64
Marin, P. 1999 Rocky River Hearth, Initial Excavation Report,
Flinders Chase National Park, Kangaroo Island.
South Australia, Adelaide.
Roberts, R. G., Jones, R. and Smith, M. A. (1994) Beyond the
radiocarbon barrier in Australian prehistory. Antiquity
68:611-616.
Robinson, A. C. and Armstrong, D. M. (eds) 1999 A Biological
Survey of Kangaroo Island South Australia, 1989 and 1990.
Adelaide: Heritage and Biodiversity Section, Department
for Environment, Heritage and Aboriginal Affairs.
Thammakhantry, I. 1998 Faunal analysis of the megafauna deposit
at Rocky River, Kangaroo Island, South Australia.
Unpublished Palaeontology Report, School of Biological
Sciences, Flinders University, Adelaide.
Tindale, N. B. 1937 Relationship of the extinct Kangaroo Island
culture with cultures of Australia, Tasmania and Malaya.
Records of the South Australia Museum 30:39-60.
Tindale, N. B. and Maegraith, B. G. 1931 Traces of an extinct
Aboriginal population on Kangaroo Island, South Australia.
Records of the South Australia Museum 3:275-290.
Tindale, N. B., Fenner, F. J. and Hall, F. J. (1935) Mammal bone
beds of probable Pleistocene age, Rocky River, Kangaroo
Island. Transactions of the Royal Society of South Australia
59: 103-106.
Wells, R., Prideaux, G., McDowell, M. and Hall, L. (1997) A late
Pleistocene deposit at
Rocky River, Kangaroo Island, South Australia. CAVEPS
Abstract, Perth.
Wells, R. and Walshe, K. 2001 An Integrated Research Plan for
Reconstruction of the Palaeontological and Archaeological
History of the Rocky River Precinct including Black Creek
Swamp, Flinders Chase, Kangaroo Island. Unpublished
Report to National Parks and Wildlife Service, South
Australia, Adelaide.
OBITUARIES
Dr Bruce Veitch
30 June 1957 – 10 March 2005
Bruce Veitch passed away in Perth on 10 March 2005
after a short battle with motor neurone disease. Bruce was
married to archaeologist Fiona Hook and had a young son
Conall. Bruce was a co-director of the cultural heritage
company Archae-Aus Pty Ltd with Fiona.
Bruce has made a major impact on the practice and
ethics of archaeological work in Western Australia. From
his pioneering work on the Mitchell Plateau for his
doctorate, to his collaborative cultural heritage work with
Fiona in the Pilbara and elsewhere, he was known for his
energy, persistence and honesty. He mobilised and
published consultancy work, collaborated closely with the
traditional owners whose sites he was working on and
worked strategically with major industry players – such as
Hamersley/Rio and BHP – as well as colleagues in the
Department of Indigenous Affairs and in the archaeological
profession. Bruce was committed to mentoring graduates
and was endlessly supportive and generous with his time,
skills and knowledge. This obituary tracks some of Bruce’s
more significant achievements through time.
Bruce completed his BA (Hons) at the University of
New England in 1985, examining ethnohistorical sources
and archaeological imprints of the pre-contact exploitation
of bracken fern. In his later postgraduate life for his
doctorate he carried out ethno-archaeological survey and
excavation programs on the Mitchell Plateau of the very
remote north-west Kimberley coast of Western Australia.
His analysis of rockshelter deposits and mounded middens,
in particular, generated discourses about the likely prime
movers for economic and demographic change being
embedded in either social process or changing
environmental landscapes. The work of Dr Harry Lourandos
was pivotal in these analyses and debates. Part of his PhD
research specifically focussed on a technological analysis of
flaked stone from three rockshelters from the Mitchell
Plateau (Ngurini, Wundalal and Bangorono). As stated by
Bruce, the object of the exercise was to: (a) identify the time
when points were first produced; (b) identify and quantify
the nature of change associated with the appearance of
points both on the Mitchell Plateau and in northern
Australia; and, (c) identify changes in lithic procurement
strategies, and by inference, changes in relative levels of
logistic and residential mobility from the time points
appeared both on the Mitchell Plateau and in northern
Australia. The other major contribution of his thesis
involved the analysis of three large shell mounds on the
Mitchell Plateau (Goala, Wundadjingangnari and Idayu).
Here Bruce challenged prevailing models relating Anadara
shell mound formation in northern Australia to
environmental change, instead linking the appearance of
Anadara and Tapes to regional population growth, reduced
mobility, a broadening of the resource base and wider
structural changes in Aboriginal territorial arrangements.
Bruce was awarded his PhD from UWA in 2000.
Most of Bruce’s cultural heritage and collaborative
research work over the last decade was in the Pilbara region
– where mitigation projects included recovering and dating
stone arrangements, linear middens and rockshelter
habitation sites (see major publications below).
Bruce was always field-active (a cruel irony given his
disabling condition during the last 6 months of his life). In
1982 he participated in excavations with Professor Graham
Connah at Bagots Hill historic site, NSW, and with
Professor Mike Morwood at the Rocky Scrub Creek site, in
south-east Queensland. In 1984 he participated in surveys
with Dr Luke Godwin within the Apsley Gorge of northwest NSW and then with Dr Dan Gillespie and Ms Hillary
Sullivan on the rock art assemblages of Kakadu National
Park. In 1985 he acted as an excavation supervisor (with
Professor Graham Connah and A/Professor Judy
Birmingham) for a joint University of Sydney and
University and New England project at Regentville, NSW.
During the next year his field efforts accelerated and Bruce
spent a month with Dr Moya Smith engaged in
anthropological study of Bardi fishing technology at Cape
Levique, WA, and with myself (Peter Veth) for three months
carrying out the first field season of archaeological survey
and excavation in the Great and Little Sandy Deserts of
Western Australia.
By 1987 Bruce was establishing the base for his doctoral
research on the Mitchell Plateau, negotiating with
Wunambal people at Mowanjum (near Derby) and
Kalumburu. Enrolled in a doctorate at UWA Bruce
65
Obituary
What is patently clear from this précis is that Bruce was
engaged in an extraordinarily broad range of archaeological
endeavours across Australia – all of which were carried out
closely with custodial and traditional owner support and
participation and which were supervised and written up to a
satisfactory conclusion. In addition to these productions,
and his peer-reviewed papers and chapters (some of which
are listed below) Bruce presented some 15 papers on all
aspects of his research and consultancy activities at both
domestic and international conferences.
Bruce’s dedication to his friends and the profession will
make him sorely missed. The loss to his family is
immeasurable. As the numerous mourners at his funeral
filed past Bruce’s coffin, his iconic and severely battered
Akubra placed jauntily at one end, Votives – in the form of
Western Australian (South-West, Pilbara and Kimberley)
shellfish – were symbolically offered, in recognition of a
truly admirable person and career.
Peter Veth and Sean Ulm
Major Publications
subsequently carried out 8 months of Mitchell Plateau
fieldwork; funded by AIATSIS. During the following years,
while working (usually part-time) on his PhD, Bruce tutored
at UWA (1989), carried out surveys for the Northern
Territory Museum and Art Gallery (1990), worked as a
Heritage Assessment Officer for the Department of
Aboriginal Sites (1992) and then as Manager of the Port
Hedland Department of Aboriginal Sites Office (1993).
Bruce self-employed consulting career began in earnest
between 1993-1994.
Between 1993 and 1996 Bruce worked as a Senior
Archaeologist for the company Anthropos Australia Pty Ltd
engaging in studies in the southern Lake Eyre region, the
Little Sandy Desert, the WA Goldfields and on the arid
north-west coastline near Onslow.
In 1997 he established the company Archae-Aus Pty Ltd
with Fiona Hook and Gavin Jackson. As their Senior
Archaeologist, he worked in the WA Goldfields, Western
Desert, the Pilbara uplands, the Burrup Peninsula, northwest Queensland and arid South Australia. In 1998 he
completed his first native title report – destined for eventual
litigation – for an area of the WA Goldfields and in 1999
was the Expert Witness for the Karajarri Native Title Claim
for the Kimberly Land Council. This claim saw native title
awarded by the Federal Court in 2004 He was also the
Expert Witness for the Wanjina/Wunggurr-Wilinggin
Native Title Claim, again for the Kimberly Land Council,
which was successfully determined in 2004 .
In 2003 Bruce oversaw the archaeological
salvage/excavation programme of the Stone Arrangements
Relocation and Dating Program – for BHP Billiton Iron
Ore, Marditja Bunjima and the Innawonga, Bunjima
Nyapialri Aboriginal Communities. This ambitious project
saw the survey, excavation and dating with relevant
traditional owners (via hundreds of OSL dates) of stone
arrangements scheduled for impact. In 2004, when already
ill, Bruce participated in the Indigenous, maritime and
historical archaeological field reconnaissance, of Barrow
Island with colleagues from both the company, the WA
Maritime Museum and UWA.
66
Veitch, B. 1985 Burning Bracken Fern: A Contribution to the
Ecology of the Aborigines of Southeast Australia.
Unpublished B.A. (Hons) thesis, University of New
England, Armidale.
Veitch, B. 1994 Hearth stones in the mound: One variable that may
aid in the differentiation between shell mounds and
megapode incubation nests. In M. Sullivan, S. Brockwell
and A. Webb (eds) Archaeology in the North: Proceedings of
the 1993 Australian Archaeological Association Conference,
pp. 167-175. Darwin: North Australia Research Unit.
Veitch, B. 1996 Evidence for mid-Holocene change in the Mitchell
Plateau, northwest Kimberley, Western Australia. In P. Veth
and P. Hiscock (eds) Archaeology of Northern Australia, pp.
66-89. St Lucia: Anthropology Museum, University of
Queensland. Tempus 4.
Veitch, B. 1999a Shell middens on the Mitchell Plateau: A
reflection of a wider phenomenon? In J. Hall and I. J.
McNiven (eds) Australian Coastal Archaeology, pp. 51-64.
Canberra: ANH Publications, Department of Archaeology
and Natural History, Research School of Pacific and Asian
Studies, Australian National University.
Veitch, B 1999b What Happened in the Mid Holocene?:
Archaeological Evidence for Change from the Mitchell
Plateau, Northwest Kimberley, Western Australia.
Unpublished PhD thesis, Centre for Archaeology,
Department of Anthropology, University of Western
Australia, Perth.
Veitch, B. 2002 Aspects of the use and fire management of
Bracken Fern (Pteridium esculentum). In D. Georghui (ed.)
Fire in Archaeology: Papers from a Session held at the
European Association of Archaeologists, Sixth Annual
Meeting in Lisbon 2000, pp. 45-54. Oxford: Archaeopress.
BAR International Series 1089.
Veitch, B., Hook, F. and Bradshaw, E. 2005 A note on radiocarbon
dates from the Paraburdoo, Mount Brockman and
Yandicoogina areas of the Hamersley Plateau, Pilbara,
Western Australia. Australian Archaeology 60: 58-61.
BACKFILL
The Australian Archaeology
Electronic Archive
Michael Haslam
School of Social Science, The University of Queensland,
St Lucia, QLD 4072, Australia.
Abstract
This report outlines the process and reasoning behind
the creation of the Australian Archaeology digital archive.
The archive, covering Volumes 1-57 of Australian
Archaeology (1974-2003), was completed in late 2004 and
distributed on a single DVD. Every page (totalling more
than 6500 pages) of the journal’s first 30 years was
converted to Adobe PDF format, and collated into a
searchable index. Future plans for the archive include
internet access through the Australian Archaeological
Association (AAA) website.
Introduction
Electronic texts are gaining prominence both within and
outside of academia, an inevitable result of the widespread
impact of personal computers, and the ease of
dissemination of digital texts worldwide through the
internet and email. Here I report on the conversion of the
first 30 years of the journal Australian Archaeology (AA)
into a digital format, and describe the process of
digitisation from its initial conception. The resultant AA
digital archive covers Volumes 1-57 of the journal,
published between November 1974 and December 2003.
This paper acts both as a report of the Electronic Archive
committee to Australian Archaeological Association
members, and as an example highlighting the issues
surrounding journal digitisation.
Digital texts
The convenience of an electronic archive is readily
apparent. There will always be a need for hard-copy print
material, if only for the ‘feel’ of holding a physical copy of
a journal, however despite possible disadvantages caused
by changing technologies there are good reasons (including
space, searching, access and permanence) for considering
the electronic version to be a more flexible and useful
product. It was the recognition of these benefits, as
elaborated below, which led to the AA Electronic Archive
project.
The first important benefit of digitisation is saving
space. Every journal requires an ever-increasing amount of
shelf space to store, whereas physically the electronic
version takes up no space at all. Additionally, the dramatic
reductions in the cost of computer storage devices over the
past few years has meant that it is now practical to store
several gigabytes (Gb) of information on a computer harddrive. Even without access to internal computer storage, the
rising use of DVD media means it is also possible to store
decades of digitised journal issues on a single disc,
requiring only the equivalent shelf space of one small book.
While these advantages most obviously apply to
institutions with large print-journal collections, individuals
with restricted space also benefit.
Aside from all other issues, the main contributors a
journal’s usefulness are accessibility and searchability. If
you cannot readily find information or obtain a copy of a
desired article, then the journal is not fulfilling its function.
That said, searching a print journal without electronic aid
can best be achieved with a photographic memory and a lot
of spare time. While many databases currently allow
searching for title, author and other bibliographic
information, relevant articles must still then be retrieved
and read in hard-copy. Problems may also be encountered in
that while most databases will index article information,
few also index editorials, letters and other material. Under
these circumstances, the ability of an electronic version to
index every word on every page in a searchable format
offers a more efficient solution. Again, once a relevant
article is found, a digital text is simple to both access and
distribute to others through email or the internet, reducing
reliance on library holdings and opening hours. Of course,
if necessary, an electronic article may also be printed in
hard-copy.
The final major advantage of digitisation is the creation
of a complete, permanent and accessible archive of the
journal. Few subscribers own the entire back-catalogue of
any print journal, making older published research
inaccessible in many cases. Important early material may be
ignored, particularly by newer researchers without access to
the holdings of large libraries. Traditionally, reprinting was
the only way to make such material widely available,
although expense and the difficulties of finding early
material to print from could be prohibitive. Reprinting
expenses are avoided using digital texts, as the reproduction
of the entire print history of a journal may cost only a few
dollars per copy. Provided care is taken in capturing the
print material in a stable and enduring electronic form,
digital archiving offers the opportunity to access all back
issues simply and simultaneously worldwide.
Digitising Australian Archaeology
The first serious discussion of converting the backcatalogue of Australian Archaeology into a digital format
was conducted on the AUSARCH-L email discussion list
during 2000. At that time, there was considerable interest
internationally among academics from many disciplines in
creating electronic archives, resulting in part from the
promise shown by early online archiving services such as
JSTOR (www.jstor.org) and institutional efforts such as the
National Library of Australia’s PADI initiative
(www.nla.gov.au/padi). The AUSARCH-L discussion was
summarised by then Australian Archaeological Association
(AAA) Secretary Louis Warren at the AAA Annual General
Meeting held at Beechworth, Victoria, in December 2000.
The consensus at the time was that the hard-copy journal
should be retained, with various access policies proposed
for an electronic version, including AAA-members-only
access or limited (abstracts only) web-based access. A
variety of recommendations were made by Louis at the
67
Backfill
Beechworth meeting, including that the AAA Executive
thoroughly investigate the costs of producing an electronic
archive, and “that AAA consider making all back issues of
AA available for sale on a special issue CD in pdf format”
(AAA 2001). Despite further debate, in the end no motions
were passed, and the matter was not investigated further.
The Jindabyne meeting of AAA in December 2003
provided the next major impetus towards the digitising
process, in part as 2003 represented 30 years of publication
of Australian Archaeology. A motion that “a subcommittee
be established to examine options for creating a 30th
anniversary electronic archive of AA” (AAA 2004) was
passed unanimously, and I was co-opted to chair the
committee. Luke Kirkwood and Sean Ulm volunteered to
assist with the project. We set a target to produce a
searchable digital archive of every AA issue from 19742003 in time for the 2004 AAA Annual Conference. The
main considerations were the actual collection and
conversion of print copies of the journal, the file format and
size of the final product, and the medium in which to
disseminate the archive.
Sourcing a comparatively clean copy of each of the 57
volumes required was expected to be a major hurdle.
Fortunately, Ian Lilley possessed just such a collection, and
was willing to have these guillotined for the greater good.
The success of the electronic archive owes a great deal to
Ian’s generosity, and to Sean Ulm and Jill Reid, who also
provided clean copies when necessary. With the journals in
hand, a commercial scanning company (The Data Image
Group in Brisbane) was employed to convert all issues
(totaling almost 6100 pages) first into multi-page Tagged
Image File Format (TIFF) files, then into Adobe Portable
Document Format (PDF) files. This conversion process
follows that of other digital distribution projects such the
US Library of Congress Digital Interlibrary Loan scheme
(www.loc.gov/rr/loan/illscanhome.html).
Considerable testing was conducted by the Electronic
Archive committee to determine the most appropriate file
types and scanning resolutions to use. TIFF was chosen as
it does not involve compression of the original page image
file, providing an archive-quality reproduction of each
journal volume. There is currently a debate among
archivists and librarians as to the longevity of digital media,
and the ability to read older digital storage formats. TIFF is
one of the standards recommended by many of the
participants in the debate as a reliable archive format, and
having the TIFF version will allow AAA to generate the AA
archive in other formats as they become available. The rate
of change of digital media and file formats is one of the
chief disadvantages in depending on an electronic archive,
and we have tried to ensure that we have the most reliable
product possible. AAA should also aim to keep at least one
pristine set of print copies of the journal in addition to any
digital initiatives.
Adobe PDF was chosen as the distribution format for
the archive, to take advantage of the widespread use and
cross-platform compatibility of Adobe Reader and Adobe
Acrobat software. Conversion of the TIFF files to PDF first
involved Optical Character Recognition (OCR) of each
page, to allow full-text searching of the PDF files. OCR
degrades image resolution, and a compromise was reached
between high-resolution page images and keeping the
overall file sizes manageable. The end result was that text
pages were scanned as black and white images at 400dpi,
68
and pages with graphics (maps, photos, illustrations, etc.) as
greyscale at 300dpi. The OCR version of the text is tied to
the PDF file, giving an exact image of the page as printed,
with all words both searchable and selectable (and therefore
able to be copied). The total cost for scanning was
$2931.04.
Each digitised volume of Australian Archaeology was
supplied by the scanning company as a single PDF file,
which meant that it was then necessary to divide these
volumes into individual articles and other contributions.
Every page was checked for scan quality and correct
orientation and over 100 pages were rescanned to ensure an
acceptable standard. A series of decisions were made
regarding just what should constitute a separate file (an easy
decision with articles and thesis abstracts, less so with
letters to the editors, corrections, advertisements and many
of the items in the 1970s newsletter version of the journal).
The front cover, editorial and any advertising material
related to the journal were kept as one file and named
‘Frontmatter’, following JSTOR conventions, and notices at
the back of the journal were kept as ‘Backmatter’ when
splitting them up was not deemed necessary. Duplication of
pages was required when articles, abstracts and other
material did not start on a new journal page, and the
resulting inclusion of the one page in both files brought the
final tally to more than 6500 pages, with over 1350 separate
files.
File naming conventions were developed based on the
year of publication, volume number, start page and end
page (in this order) of the included text. This gave 13-digit
file names, of the format YYYYVVVSSSEEE.pdf, where
Y=year, V=volume, S=start page and E=end page. This
format has the benefit of being unique to each file, except
in rare cases when two articles begin and end on the one
page, in which case the letter a or b was added to the 13
digit filename. In addition, the naming conventions order
all files in correct publication/chronological order when
sorted by filename during browsing or when viewing
search results. Each journal issue was placed in a separate
folder, to facilitate browsing by filename when a
researcher already has the reference for an article and does
not need to use the search function. A searchable index
was then created which referenced every word in every
volume, using Adobe Acrobat’s Catalog function. The total
file size of the index and all journal volumes was 1.9Gb.
Having completed the indexed archive, consideration
turned to the appearance of each article within Adobe
program, and the inclusion of metadata. Title and author
information was entered for every document, with the
year of publication added before the title to help the
researcher choose an appropriate file from the search
results. We took into consideration the reduced screen
size of notebook computers by limiting extra information
in the document title, to ensure that the contents of each
file could still be assessed when viewed even with a small
search results window. For clarity, however, we did add a
one word descriptor to the title of documents such as
advertisements and obituaries to act as a guide to file
content. Each document was designed to open in full page
view within Adobe Reader/Acrobat, which avoids
confusion when an article or other item begins halfway
down a page. Along with several other minor
modifications, these solutions make the archive both
comprehensive and user-friendly.
Backfill
Dissemination
The final stage of archive development was producing
an effective means of accessing and distributing the
collection. The initial aim of distribution on compact disc
(CD) was not feasible, as the large file size would require
three CDs, each of which would need to be searched
independently whenever a researcher wanted to find a
particular reference. The long-term goal is distribution via
the AAA website, however again this requires a large
amount of server space which was not available at the time.
The compromise decision was to distribute the archive on a
single DVD (which holds 4.7Gb), allowing for a search of
the complete archive at one time. We realised that many of
our members would not have computer DVD drives,
however these are becoming increasingly common and our
decision reflected a belief that present ease of use
outweighed any limits this may place on short-term
dissemination.
In order to make the collection as user-friendly as
possible, we designed a menu to automatically run on
Windows computers, which allows the user to select either
the searchable index or an option to browse the files. Apple
Macintosh users click an icon to search. Both platforms can
access a comprehensive help file which details searching
procedures and many of the file conventions outlined above.
The help file contains the system requirements for reading
the DVD, which is necessary both for current users and to
ensure future access as digital formats change. The archive
was tested on a variety of platforms and operating systems,
and with a number of versions of the Adobe Acrobat and
Reader software package. The latest version of Adobe
Reader was included on the DVD to ensure compatibility
for all users. The final product, with a cover designed to
match that of the print version of the journal, was produced
by Media Technology of Brisbane, with the initial run of
100 DVDs costing $682.00.
The DVD archive was launched at the AAA Annual
Conference held in Armidale, New South Wales in
December 2004. Within two days, enough discs were sold
(at $50.00 each) to recover all costs for the project. It is
expected that as new members continue to join AAA and
DVD drives become commonplace in computers that the
archive will continue to generate funds for the Association,
particularly as the only costs associated with production at
this point are pressing new discs. The $50.00 price of each
copy of the archive equates to 1.3 cents per page (the
equivalent of spending less than $4.00 on a 300-page book),
representing significant value for money for AAA members
and others interested in the history and current directions of
archaeology in Australia.
Conclusion
The Australian Archaeology Electronic Archive project
takes advantage of each of the main benefits of digital texts:
saving space, providing ease of access and searching to all
back issues, and permanently archiving the journal. Future
issues of AA will be archived directly from the original
digital files as sent to the printer, providing pristine copy,
avoiding the need to scan print issues, and saving on digital
conversion costs. Making the archive accessible via the
AAA website, using a ‘moving wall’ policy which
maintains the value of the print issues as they are released,
is the next step in gaining a wider audience and increasing
accessibility to the journal.
Acknowledgements
As members of the Electronic Archive subcommittee,
the assistance of Luke Kirkwood and Sean Ulm was
fundamental to the success of the project. Ian Lilley’s
journals and help were invaluable. Judy Field, Joe Dortch,
Michael Slack and Amy Roberts (the AAA Executive)
provided support for the project, as did AA Editors Donald
Pate and Pam Smith. Emma Oliver and Cameo Dalley
assisted with data entry and quality control. Thanks also to
Nathan Woolford, Jill Reid, Alison Crowther, Cath Latham,
Valerie Morley, Jenna Lamb and Duncan Lord.
Reference
Australian Archaeological Association 2001 Minutes of the 2000
Annual General Meeting of the Australian Archaeology
Association. Australian Archaeology 53:59-62.
Australian Archaeological Association 2004 Minutes of the 2003
Annual General Meeting of the Australian Archaeological
Association Inc. Australian Archaeology 58:53-62.
2004 AAA Conference Dinner Awards
Rhys Jones Medal for Outstanding
Contribution to Australian Archaeology
Emeritus Professor John Mulvaney AO CMG
(1925-)
The Rhys Jones medal is Australian Archaeology’s
highest honour, and the presentation to Professor Mulvaney
is an acknowledgement of his outstanding contribution to
Australian archaeology, to AAA, the academic discipline,
and to increasing public awareness of the discipline and the
importance of Australia’s cultural heritage.
John Mulvaney was as the first university-trained
prehistorian to make Australia his subject, and he has been
justly described as the ‘Father of Australian Archaeology’.
John was born in 1925 in Yarram in south Gippsland.
His father was a teacher and before the Second World War
the family moved around country Victoria, to towns such as
Rainbow in the Mallee where his father had been promoted
to headmaster of Rainbow Higher Education School (and
where John was house captain), and eventually to Frankston
near where John was born.
In 1943, when John was 18, he joined the RAAF and
was sent to Canada for training. The following year he was
posted to England. Fortunately for Australian archaeology,
he survived the war. During his military service in England
he toured the countryside and visited megalithic standing
stones called ‘the Consuls’. It was this particular visit that
sparked his interest in prehistory.
On his return to Australia, John enrolled at the
University of Melbourne and studied Roman History under
John O’Brian – Roman History was a special subject and
the class was only six students. In 1949 he was appointed
tutor in ancient history by Max Crawford the history
professor at the University of Melbourne. He immediately
enrolled in an MA and submitted his thesis 12 months later.
The topic of his thesis was ‘State and Society in Britain at
the time of Roman conquest’. This was a turning point of
sorts, because John’s study on ancient Britain had
convinced him that Australia must have a significant
archaeological record as well. While it seems self evident
today, this was not at all the case in 1950.
69
BackfillBook Reviews
John next applied for an Australian National University
post-graduate scholarship. In his application he argued that
it was essential for him to train as an archaeologist and that
he would have to do undergraduate studies in prehistory at
Cambridge University. Fortunately, the ANU committee
was persuaded, and in 1951 John became an undergraduate
student again, this time at Clare College. At Cambridge
John studied stone tools and to gain essential field
experience he participated in a number of archaeological
excavations in Britain and Ireland. He also toured
archaeological sites in Germany and Denmark.
Shortly after his return to Australia, John began
lecturing in ancient history at the University of Melbourne,
with his former teacher and mentor John O’Brian. One of
the courses he taught was the history of archaeology.
By the mid 1950s John had begun his journey into
Australian prehistory by excavating a limestone rockshelter
at Fromm’s Landing on the Murray River. His labours
continued into the early 1960s and included his discovery of
what are still the oldest recorded dingo remains in Australia,
and evidence of a massive flood of the Murray thousands of
years ago. His report on Fromm’s Landing was published a
year after he completed his last season of fieldwork – John
has always set a standard in the speed in which he brings his
research findings into print. John’s second excavation was
also a limestone rockshelter, this time at Glen Aire on Cape
Otway. It was Isabel McBryde’s first fieldwork experience
– Isabel McBryde was awarded the Rhys Jones medal last
year for her own distinguished contribution to Australian
archaeology. John’s third excavation, at Kenniff Cave
between 1960 and 1963, pushed back the antiquity of
Aboriginal occupation of Australia many thousands of years
into the Pleistocene era.
In 1965 John was called to the Australian National
University and within a few short years published his book
Prehistory of Australia. This book has now seen three
editions (the most recent with Jo Kamminga as co-author in
1999) and was last reprinted only a few months ago.
In 1969, John went with Jim Bowler and Rhys Jones to
Lake Mungo to investigate Jim’s discovery of Pleistoceneage artefacts and human remains that were later to be
known as ‘Mungo lady’. As with much of John’s work, this
expedition is now history. He returned with Jim in 1973 to
direct the largest dig ever at Mungo, which revealed a
hearth dated to about 31,000 years.
In 1971 John was appointed to the Foundation Chair in
Prehistory in the Arts Faculty at the Australian National
University and in the following year introduced Prehistory
1 as an undergraduate subject. He also turned his attention
to public issues. He was involved in organizing the first
meeting of the Australian Archaeological Association,
which will have its 30th anniversary next year.
For many years he was a Commissioner of the
Australian Heritage Commission, involved in the
formulation of the Burra Charter, and the chief Australian
delegate to the inaugural UNESCO meeting in Paris, held to
determine the criteria for World Heritage listing. He was
instrumental in nominating the Willandra Lakes and
Kakadu National Park to the World Heritage list. He served
a total of 18 years on the executive of the (then) Australian
Institute of Aboriginal Studies, including a term as
Chairperson, and served on the Committee of Inquiry on
Museums in 1974-75 which preceded the establishment of
the National Museum of Australia.
70
John was a leading light in bridging the gap between the
public and academia. He actively campaigned on pubic
issues, not the least the struggle to save the Franklin River
and its Aboriginal heritage, and support for Dawn Casey
during her tenure as Director of the National Museum of
Australia. In fact, my own introduction to Australian
archaeology was at a public meeting at Rockdale Town Hall
in 1983 when John came to talk about the archaeology of
the Franklin River during the lead up to the election which
saw Bob Hawke become Prime Minister. In this brief
citation we can only offer glimpses of his many
contributions to public debates and to the welfare of the
nation. His role as a public intellectual during his long
career has been detailed in the book ‘Prehistory to Politics.
John Mulvaney, The Humanities and the Public Intellectual’
edited by Tim Bonahady and Tom Griffiths.
After his formal ‘retirement’ from the ANU in 1985
John said he was leaving the discipline in the hands of
younger generations. However, he has maintained an
enormous productivity to the benefit of Australian history
and prehistory and the study of the humanities in general.
He served for many years as Secretary of the Australian
Academy of the Humanities, an honorary and unsalaried
position which was in every sense was a full-time
appointment, and testimony to his considerable energy.
He has continued to write, coauthor and edit books,
including, in 1992 ‘Commandant of Solitude (The journals
of Captain Collet Barker)’, in 1997 ‘My Dear Spencer. The
letters of F.J. Gillen to Baldwin Spencer’, the third revised
and enlarged edition of ‘ Prehistory of Australia’, and as
recently as this year ‘Paddy Cahill of Oenpelli’.
The Rhys Jones Medal is not John Mulvaney’s first
award for his distinguished and lasting contributions to
Aboriginal studies and the discipline of prehistory. In 1970,
John was awarded a PhD by Cambridge University; in 1982
a CMG (Companion in The Most Distinguished Order of St.
Michael and St. George); in 1991 an Order of Australia
(Australia’s highest Order), and in 1999 the Graham Clark
Medal by the British Academy.
In awarding the Rhys Jones Medal to John Mulvaney,
the AAA acknowledges his pioneering spirit, his
distinguished and sustained achievements in Australian
prehistory, his fostering of the discipline in Australia and
mentoring of so many young archaeologists, including
those who themselves have attained distinction, and more,
his inspiration, dedication, integrity, and exceptional
professionalism.
Jo Kamminga and Judith Field
Life Membership for Outstanding
Contribution to the Australian Archaeological
Association: Luke Kirkwood
Luke Kirkwood’s journey in archaeology began with
winning a high school Earthwatch competition to excavate
with Bruno David in Cape York Peninsula followed by an
honours degree at the University of Queensland. Luke’s
major contribution to the Association has been to commit
months of voluntary time as AAA Webmaster to build two
complete versions of the website over the last 5 years (and
all this while completing a PhD!). Luke’s work has made the
website the major public face of the Association (with over
Backfill
50,000 visits per month and growing) and provides services
to members and access to a wide range of resources for the
broader archaeological and non-archaeological community.
These resources are not readily available elsewhere and are
vital to the Association’s core mission to promote the
advancement of Australian archaeology. Luke’s other
contributions to the Association include: coordinating IT
for the 2001 Hervey Bay conference; co-editing the 2001
conference proceedings (published as Tempus 7); building a
new user-friendly AAA membership database; assisting
with digitising AA1974-2003 and production of the DVD;
creating the National Archaeology Week website, including
working on the popular Meet the Archaeologist page;
publishing research in AA; and active participation at AAA
annual conferences. Luke’s work has almost singlehandedly brought the Association into the 21st century and
the website stands as a substantial and enduring
contribution.
Sean Ulm
Paper and Poster Prizes
Best Overall Paper Prize: Ken Mulvaney
Best Student Paper Prize: Sarah Martin
Best Overall Poster Prize: Maria Cotter
Best Student Poster Prize: Oliver Brown
AAA Photo Competition Prizes
Archaeological Site Images: Matt Schlitz
Archaeological Fieldwork or Lab Work in Progress:
Matt Schlitz
Artefact Images: Kerrie Grant
Electronic copies of the winning photos can be viewed at the
AAA website: http://www.australianarchaeologicalassociation.
com.au/awards/
WHY ARE SO FEW AUSTRALIAN
ARCHAEOLOGISTS QUATERNARY
SCIENTISTS?
The biennial meeting of the Australasian Quaternary
Association (AQUA) took place on 6 - 10 December 2004,
at Cradle Mountain in Tasmania. As at previous AQUA
meetings I have attended, the absence of other
archaeologists was regretted, although Richard Cosgrove
did deliver one of the keynote addresses; a masterly
overview of the Tasmanian archaeological record, placing it
in its climatic context.
Why more members of the Australian Archaeological
Association (AAA) do not also join AQUA I do not know,
as there are obvious synergies between the two associations.
Membership is still cheaper than AAA, even though AQUA
has just doubled its fees (the new rates are given below). It
includes automatic attachment to the AQUA electronic
mailing list and two issues of Quaternary Australasia a year.
QA carries fewer articles, more news and views, than
Australian Archaeology, it is true; but the number and range
of journals to which Quaternary scientists can submit
papers is greater than those available to Australian
archaeologists. AQUA conference fees are also lower than
AAA's and offer great value for money; camping is always
an option at meetings usually taken up by many members.
AQUA conferences last longer than AAA meetings and
usually include free or low-cost field trips, often rendered
permanently useful by the provision of field guides. AQUA
also includes the email addresses of all attendees in the
abstract volume, helpful information that AAA might
consider offering – even if only as printout available at the
conference. Like AAA, AQUA offers student travel grants
and cash prizes for student posters and oral presentations, of
which there are usually many, proof of the health of
Quaternary Science in Australasia.
AAA
AQUA
Membership fee
$50
$40
Student /unwaged
$30
$25
Conference fee
$160 (3 days) $250 (5 days)
Student /unwaged
$98
$120
Dinner
$45
$32
The pleasant atmosphere at AQUA meetings is also a
distinct drawcard. Discussion at AQUA conferences is
always friendly, constructive and helpful, particularly of
student presentations.
It is true that some of the sessions at the recent AQUA
meeting contained few papers of interest to those
archaeologists who work exclusively in the recent past. For
example: Defining the Quaternary - the lower boundary has
been moved back to 2.6 Ma (again!), although northern
hemisphere glaciation actually began about 3.2 Ma – which
neatly disposes of the Pliocene, if one equates the
Pleistocene with the Quaternary; or Emerging Chronologies
of Glaciation - chiefly of interest to Tasmanians and New
Zealanders, many of whom attended.
On the other hand, the two sessions on Water Pasts –
Water Futures included data that will be useful to my own
research in the semi-arid zone of WA. Some of the papers in
the fluvial session were clearly of relevance to
archaeologists working in southeastern Australia, although
they focused on the later Pleistocene (0.7 – 0.125 Ma);
those in the lacustrine to loess session included a paper on
the origins of salinas in Western Australia – probably during
the Tertiary, and several that discussed proxy climatic
records for the last glacial-interglacial cycle (MIS 4-1), the
period during which people first reached Australia.
The papers in the two sessions on Biotic Responses to
Climate Change and that on Speleothems and Corals also
contained proxy climatic information useful to
archaeologists, including high resolution records from
southeastern Queensland, Tasmania and western Victoria.
El Niño events and the nature of glacial terminations were
discussed; as was the evidence for the Younger Dryas (11 –
10 ka) in Australia - the jury seems still to be out on whether
this late glacial cold flip was felt here.
Finally, a few papers discussed Marine Archives,
including isotopic data from which sea level can be
inferred. These papers focussed on the southern oceans and
were perhaps more relevant to African than to Australian
archaeologists, but anyone interested in human evolution
71
Backfill
would have found them useful. These papers would have
provided a greater understanding to archaeologists of the
often unrecognised complexity of sea level curves including
factors such as isostasy, the rate of sea level rise and
tectonism.
The conference abstracts may still be available on the
AQUA website, if people want to access them. Alternatively, I
imagine copies could be obtained by email from
[email protected] or [email protected].
All told, AQUA 2004 lived up to the high standards I
have come to expect, based on previous meetings I have
attended, not least because of the strong New Zealand
presence: 10% of attendees and presentations. It augered
well for the Australasian contribution to INQUA, to be held
in Cairns in mid-2007. The next AQUA meeting should
take place in late 2006, although the place and date have yet
to be determined, other meetings have to be worked around.
May I suggest that if AQUA and AAA want to form a
synergistic relationship, something I obviously feel would
benefit both, that AQUA tries to schedule its next meeting
to run back-to-back with AAA, at a neighbouring locale.
Then, people who come from far afield might be able to fit
both meetings into one airfare.
In the meantime, I would encourage archaeologists to
join AQUA. You would be made most welcome. Contact
Janelle Stevenson for membership forms, or visit the AQUA
website (www.aqua.org.au). Quaternary scientists often
attend AAA conferences; there were several at Armidale. It
is time archaeologists returned the compliment and attended
AQUA meetings.
Esmée Webb
Centre for Human Genetics, Edith Cowan University
HANDBOOK OF STABLE ISOTOPE ANALYTICAL
TECHNIQUES
Volume 1
Edited by
P.A. de Groot
Elsevier
2004
ISBN 0-4445-1114-8
RRP US$150.00
(hardcover)
72
THESIS ABSTRACTS
AN ARCHAEOLOGICAL ANALYSIS OF GENDER
ROLES IN ANCIENT NON-LITERATE CULTURES
OF EURASIA
Mike Adamson
Ascription of sex to inhumed remains on the principle
basis of grave-goods, as distinct from anthropometric data,
can be a vague process due to incipient gender bias in
interpretation. Cross-matching of athropometrics with
grave goods can sometimes generate results that appear
ambiguous or paradoxical as they may not accord with
preconceived relationships between gender roles and sex.
This reduces confidence in the demography of various
archaeologically-revealed cultures, especially those of Iron
Age Europe, which were erected on the basis of what we
may now see as potentially flawed analysis.
Comparative and contrasting analyses are made of
contemporary and related cultures to investigate gender role
assumptions on a wide basis. Regarding non-literate
cultures, archaeologists have limited means to interpret the
relationships between sex and gender-roles, and these
methods are explored. The traditional outlook is assessed
for functional bias in light of its origins and perpetuation,
and a new synthesis is proposed for ongoing analysis. This
synthesis includes strict application of refined anthropometric methodology and the resolution of paradox by
adoption of a revised underlying hypothesis.
A correlation is observed between use of the horse and
a significant blurring of gender role stereotypes, occurring
in nomadic cultures whose legacy persists to the present
day. This is examined in light of the proposed new synthesis
for a consequential or coincidental relationship, the former
being apparent. It is found that gender role bias has played
an uncomfortably large part in Iron Age scholarship, and
that outdated sociocultural assumptions continue to foster
an unsupportable view of elements of world history.
Degree and University: MA, Department of Archaeology,
Flinders University
Date Submitted: July 2004
Copies Held: Flinders University Library, Department of
Archaeology, Flinders University
Current Affiliation: Department of Archaeology, Flinders
University
CONTROL AND POWER IN AUSTRALIAN
COMMUNITY ARCHAEOLOGY: CASE STUDIES
FROM WAANYI COUNTRY,
NORTHWEST QUEENSLAND
Cameo Dalley
In this thesis I undertake one of the first critical
examinations of Australian community archaeology,
focussing on issues of control and power which are central
to the community archaeology approach. In the past,
archaeologists have driven archaeological inquiry, creating
a situation where Indigenous communities rarely benefited
directly from research. In response to this, community
archaeology attempts to reorient the archaeological process
to better suit the needs of Indigenous communities, thus
moving towards greater Indigenous control.
I argue that understanding relationships between
archaeologists and Indigenous communities also requires
consideration of the institutional frameworks under which
research is undertaken and the relative access that
archaeologists and Indigenous communities have to ‘power
resources’. In order to understand how community
archaeology compares to more conventional approaches to
heritage investigation, I compare a community archaeology
project to a cultural heritage assessment and a processual
research project all conducted in Waanyi country, northwest
Queensland.
I conceptualise my findings as being along a
continuum of control. While the community archaeology
project achieved more control for the Waanyi than the other
projects, I also recognise the significant contributions that
the other projects make towards achieving Indigenous
control. Findings suggest that the issues surrounding
community archaeology will continue to play a defining
role in guiding Australian archaeology in the future. I
conclude that archaeologists must continue to engage with
issues of Indigenous control and that community
archaeology is a suitable way to achieve this.
Degree and University: BA (Hons), School of Social
Science, University of Queensland
Date Submitted: October 2004
Copies Held: Fryer Library, University of Queensland;
School of Social Science, University of Queensland
Current Affiliation: School of Social Science, University
of Queensland
73
Debitage
RECOGNISING PHYSICAL CHILD ABUSE
IN ANTIQUITY: A PALAEOPATHOLOGICAL
APPROACH
Stefani Eagle
A study was conducted to establish the optimal means
of inferring physical abuse from immature skeletal remains
in past populations. Skeletal trauma commonly associated
with physical abuse was examined in light of the sociocultural and archaeological context of subadult remains.
The varied perceptions of child maltreatment and abuse
were explored in agricultural Indian settings to illustrate the
complexity of deconstructing societal views concerning
child abuse. The quantitative and qualitative analysis of
subadult skeletal trauma and cultural perspectives of child
abuse determined that the most useful means of interpreting
skeletal trauma was through a palaeopathological guide.
The research emphasises the significance of applying
multidisciplinary strategies for a balanced and consistent
interpretation of trauma in skeletal remains recovered from
archaeological excavations. An holistic construal of skeletal
pathologies in ancient remains is important for
understanding human actions of the past.
Date Submitted: November 2004
Copies Held: School of Social Science, University of
Queensland
of Queensland
THE HISTORICAL ARCHAEOLOGY OF
TASMANIAN BASED WHALING IN SOUTH
AUSTRALIAN WATERS 1820-1850
Kylli Firth
This thesis provides a comparative historical and
archaeological analysis of the shore-based and pelagic
(deep-sea) whaling industry of Tasmania and South
Australia. This was an important maritime industry which,
more often than not, has been dismissed within
contemporary Australian historical writings. It is argued
that Launceston and Hobart Town whalers, who plied their
trade during a relatively short but vital period of economic
growth in colonial history, were familiar with the spoils of
whaling, not only in their own coastal and oceanic waters,
but also in those of South Australia.
The identities of this industry, both owners and
workers, are examined. They were often well acquainted,
either through business or through rivalry, and were
entrepreneurs with a common goal. The whaling vessels
were owned, captained and regularly exchanged within the
same small group of men. The coastal whaling voyages and
shore-based whaling establishments set up by these men
opened up a significant number of frontier settlements
along the South Australian coastline.
Historical documentary records combined with
maritime archaeological evidence are employed to examine
the nature and extent of the role played by the Tasmanian
entrepreneurs in the development of the South Australian
74
whaling industry. A database is developed that provides a
summary of whaling vessel voyages, dates and destinations
from both primary and secondary source material. This
information is sufficient to determine a trajectory of events,
and provides a direct correlation between vessels, owners
and workers, and the establishment of several early shorebased whaling sites in South Australia.
Both maritime and historical archaeology are
integrated to determine that a shore-based whaling station
site existed at Fisherman’s Point in Spalding Cove, South
Australia. It is confirmed that the whaling station was
owned and operated by Henry Reed of Launceston during
1831 and 1832. Furthermore, the precise location of the site
is determined from the documentary and archaeological
evidence.
Degree and University: MA, Department of Archaeology,
Flinders University
Date Submitted: January 2005
Copies Held: Flinders University Library, Department of
Current Affiliation: Department of Archaeology, Flinders
University
OUT OF THE BOX: POPULAR NOTIONS
OF ARCHAEOLOGY IN DOCUMENTARY
PROGRAMS ON AUSTRALIAN TELEVISION
Stephen Nichols
In this thesis I investigate the relationships between
mass media and popular notions of archaeology in
Australia, and consider the implications of these
relationships for the public outreach strategies of Australian
archaeologists. First, I review the limited survey data
available regarding public opinions of archaeology in
Australia, together with the results from more extensive
surveys conducted in North America. These surveys suggest
that popular notions of archaeology are characterised by a
variety of misconceptions and stereotypes that are not only
incongruous with the ethical goals of the profession, but
which may also inhibit the wider acceptance of
archaeological perspectives in contemporary social and
political discourses. Second, I develop a theoretical model
of mass media that articulates the nature of the relationships
between producers of mass media and their audiences. This
model predicts that widespread popular notions of
archaeology are likely to be reflected in the texts of
mainstream mass media. Third, I present the results of a
content analysis study undertaken in relation to
archaeological documentary programs screened on
Australian television, demonstrating that a number of
misconceptions about archaeology are deeply entrenched
within contemporary Australian society. Finally, I identify
a number of pathways along which archaeologists might
seek to engage mass media as part of a broader
‘popularisation’ approach to public outreach in Australian
archaeology.
Queensland
Debitage
of Queensland
THE BLOKEMUSEUM: MOTOR MUSEUMS AND
THEIR VISITORS
Rob Pilgrim
Motor vehicles have been on the roads of Western
nations for over a century and in that time they have
changed the world in which they operate to the point that
today's society could not exist in its current form without
them. The motor vehicle has altered daily life beyond the
comprehension of those who lived in the pre-automobile
age. In that same 100 years, museums too have changed
radically in the way in which they collect, interpret and
exhibit objects. They have gone from being places of private
pleasure for a select few to being places of public recreation
and education.
The development of the museum has paralleled the
rise of the automobile. The first motor museum came into
existence in 1912, less than two decades after the advent of
the motor vehicle and, since that first emergence, motor
museum numbers have fluctuated, but generally increased.
With the centenary of the coming of the motor car, however,
there has been a sudden increase in the number of motor
museums in existence. Although there is no central register
or list of motor museums, and many museums are private
entities that are publicised by word of mouth, there are
probably well over a thousand in the Western world.
This thesis uses the National Motor Museum at
Birdwood in South Australia as a lens through which to
examine motor museums generally through a face to face
survey of visitors to that museum. The aim of that survey
was to ascertain what it is that visitors expect from their
visit. A further postal survey of motor museums in English
speaking countries, examines what staff in those museums
see to be the aims of the visitor and also the way in which
the museums strive to meet those aims. As essential
elements of the thesis, the nature of car collecting and how
this influences the vehicles collected by, and exhibited by,
the motor museum; as well as how those collected vehicles
are interpreted, are examined. In addition, the way in which
collection policies and goals vary from museum to museum
is addressed. The desire of many visitors to see 'the real
thing', the authentic vehicle, is also considered as is the
status and use of simulacra and replicas in the motor
museum. The ambition of many museums, in response to
that perceived visitor objective, to fill the exhibition halls
with numbers of vehicles that have been restored, to a point
beyond their original, as manufactured condition is also
evaluated. The nostalgic goals of many visitors, as well as
the ways in which the museums strive to meet those goals is
assessed. Finally the thesis examines the intersection of the
gendered object, the automobile, and the gendered space,
the museum, examining the ways in which the female
motoring experience is interpreted in the motor museum
and suggesting ways in which motor museums might
change to better include visitors of all types.
Degree and University: PhD, Department of Archaeology,
Flinders University
Date Submitted: August 2004
Copies Held: Flinders University Library; Department of
Current Affiliation: Senior Curator, National Motor
Museum, Birdwood, South Australia
BONES, BONES, BONES – WHAT SECRETS DO
THEY KEEP? EXAMINING THE FEASIBILITY OF
USING TRACE ELEMENTS AND RARE EARTH
ELEMENTS TO DETERMINE GEOGRAPHICAL
DIFFERENCES OF ARCHAEOLOGICAL REMAINS
Ian Scott
The provenancing of unprovenanced human remains is
an issue of increasing concern to anthropologists,
archaeologists and indigenous communities. A range of
techniques is currently employed to determine the origin of
unprovenanced human remains. However, the techniques
currently available are either too broad or too specific in the
scale of data resolution. There is a need for a method that
will help place the remains in a more specific geographical
area than is already possible.
This pilot study examines the feasibility of using trace
elements and rare earth elements to determine geographical
difference of archaeological remains. Non-human bone
material from two sites, Platypus Rockshelter in southeast
Queensland and Grinding Groove Cave in Central
Queensland are used as case studies. As this is a pilot study
limited by the availability of resources, it does not produce
a set of elements that are unique to the sites. However, this
study demonstrates that it is feasible to separate
archaeological remains from sites in different geographical
areas with the use of trace elements and rare earth elements.
Queensland
of Queensland
CORRECTIONS TO AA 59
The review of Keith Windschuttle’s book ‘The
fabrication of Aboriginal history, Volume 1, Van Diemen’s
Land 1803-1847’ was correctly attributed to Sandra
Bowdler on the Contents page but incorrectly attributed to
Darren Griffin on page 69 of the Book Review section.
The references for Shane Burke’s Short Report (pp.
62-63) were omitted. Graphic Print Group have provided an
adhesive-backed replacement page 64 containing the
missing references that can be inserted in AA 59. This page
has been included with the volume 60 mailing.
The Editors apologise for these errors.
75
BOOK REVIEWS
THE INCAS, by Terence N. D’Altroy. Maiden,
Massachusetts/Oxford/Carlton, Victoria: Blackwell
Publishing Ltd (2003) xv+391 pages. ISBN 1-40511676-5 (paperback). Price A$69.95.
David Bulbeck
The author Terence D’Altroy belongs to the Realpolitik
school of anthropological archaeology, and this perspective
comprehensively informs his representation of the Inca
empire, through his selection of which historical events to
relate and which socio-political aspects to emphasize. The
result is a wide-ranging and sophisticated description of the
Incas which is, however, readily accessible to the general
reader and specialist alike through a neatly organized
chapter structure and avoidance of unnecessary jargon.
Still, the book would disappoint readers who had been
hoping for a romanticized account. Carving out a vast
empire in a matter of decades, and consolidating imperial
rule over a geographically and ethnically diverse realm,
were hardly pretty affairs, and D’Altroy’s work gives
political intrigue and military might equal billing with the
marvellous accomplishments of the Incas in record keeping,
road works, and integrating the technological and
agricultural skills previously developed by the Incas’
subject societies.
Chapter one introduces the available sources on the
Incas, and discusses the author’s intention to combine the
historical and archaeological evidence to a degree not
hitherto achieved. Chapter two briefly describes the central
Andean belt and its coastal and jungle fringes in terms of
physiography, geography and society, including a few pages
devoted to pre-Inca prehistory. The Killke antecedents of
the Incas, in the Cuzco basin (chapter three), appear quite
unremarkable in this context, and D’Altroy wisely avoids
looking for root causes to explain the Inca expansion. The
expansion of the empire is recounted in chapter four, and
the political organization of the empire, both in its
consolidated regions and in the borderlands where Inca
armies continued their conquests at a retarded rate, is the
concern of chapter five. Chapter six covers Cuzco and the
sacred Urubamba Valley, and the question of whether the
imperial expansion was fuelled by the prerogative to
accumulate further estates for deceased rulers with each
passing emperor.
Chapter seven describes the Cuzco-centred state
ideology, with the ruler as the sun god’s living
representative, as well as the belief systems of the subject
societies and how they fitted uneasily with the Incaimposed religion. In chapter eight we learn how people
made a living, and how the Incas controlled the central
Andean surplus both to feed their armies and to host
sumptuary feasts (associated with state construction works
and Inca festivals). The following chapters provide a more
detailed summary of the organization of the empire in terms
of its army, provincial rule, food production and storage of
surplus, and the superlative accomplishments of Andean
and coastal Peruvian societies in textile production, metal
76
work, ceramics and masonry. The book finishes with a
succinct account of the Spanish invasion at a time of civil
war, the Inca resistance even after the Spanish had occupied
Cuzco and established a new capital at Lima, the tragic
depopulation (through disease and harsh Spanish rule) in
the aftermath of the invasion, and the maintenance of preSpanish traditions amongst many Peruvian and Bolivian
communities to this day.
Each chapter contains sufficient background
information to allow a reader who wishes to learn about a
particular topic to dip into the relevant pages. The book also
works as an integrated whole with later chapters providing
the detail on topics raised in earlier passages. D’Altroy,
whose background is archaeology, combines history and
archaeology as well as can be achieved. Even when the
primary evidence is one or the other, it is set in the context
of its complement; for instance, knowledge of the initial
imperial expansion may rely on early colonial records, but
archaeological evidence is the critical source on the size and
organization of the societies which were conquered.
Technically, the writing style is informal to the point of
seeming almost whimsical on occasions, but always crystal
clear, and the illustrations are nicely prepared, even if they
sometimes require good eyesight or spectacles to appreciate
the detail. The final pages of the book include a useful
glossary and a large bibliography which, it should be noted,
gives more space to many of D’Altroy’s colleagues than to
his own publications.
Overall, this is a very successful book on the Incas
which is destined to replace standard academic overviews
(e.g., Anne Kendall’s Life of the Incas). It is written in layers
of meaning, so that first-year archaeology students, and
other readers seeking a bird’s-eye view, can scan it for a
quick appreciation, while more specialist readers can also
extract useful nuggets for their purposes from the detail.
That said, it may be inferred that D’Altroy’s book would
also serve as an excellent later-year textbook as well as a
useful ‘Inca thesaurus’ to adorn the bookshelves of
postgraduate students and academics.
COLOURING THE PAST: THE SIGNIFICANCE
OF COLOUR IN ARCHAEOLOGICAL
RESEARCH, edited by Andrew Jones and Gavin
MacGregor. Oxford: Berg (2002), xv +250 pages; 50
illus. ISBN 1-8597-3542-8 (hardcover) Price £45.00;
ISBN 1-8597-3547-9 (paperback). Price £14.99.
Noelene Cole
‘Why has it taken so long for archaeology to undertake
a critical treatment of colour?’ The editors of Colouring the
Past find the answer to this question in various (post-1980)
developments in Anglo-American archaeology: a new
emphasis on the experiential nature of material culture,
increased awareness of the senses in archaeological inquiry
and a heightened interest in representation and visual
communication. This may bemuse a few rock art
Book Reviews
researchers who have long been interested in the
archaeology of visual communication, but the premise is
valid for mainstream archaeology.
The genesis of Colouring the Past was the 1999
European Archaeological Association conference. The
book’s contents (12 chapters by 13 authors) deal with colour
in funerary practice, stone monuments, stone and metal
artefacts and wall paintings. The editors describe the
temporal scope as ‘outside the traditional purview of art
historical analysis’. However, as the focus is mainly
European Neolithic and Bronze Age societies, it is
surprising that rock art of these contexts is unrepresented.
On the other hand, it is refreshing to find an emphasis on the
subtle colour symbolism of soils, pebbles, rocks and
building stones.
In the introduction, Jones and MacGregor aim to
develop an approach to the ‘deep history’ of colour in the
context of materiality. They introduce debates in cognitive
psychology on the relativity or universality of colour, in
particular relating to the Berlin and Kay model and its use
of linguistics and the Munsell Colour Chart. Jones and
MacGregor dispute the diachronic conclusions of Berlin
and Kay and the narrow approach to colour classification in
the Munsell scheme. However, they approve the latter’s use
as a ‘site’ allowing researchers to discuss colour with the
same terms of reference, its operating principle being that,
physiologically speaking, ‘humans in different cultural
settings perceive colour in similar ways’. I am puzzled
therefore that Munsell values were not employed in various
tables (e.g. Figs. 1.2, 2.5, Table 9.1) to attempt to objectify
colour terms such as ‘buff’, ‘brown-black’, ‘brick red’ etc.
In Chapter 2 Chapman provides a more detailed critique of
the ‘falsely diachronic’ Berlin and Kay model, noting that
colour pathways are not characterised by developing colour
complexity. Chapman provides an overview of alternative
approaches in anthropology and cognitive linguistics to the
integration of colour meaning into colour studies.
Chapters 2 to 9 present case studies of foregrounded
colours and their roles in communicating cultural messages
in various social contexts. Most authors attempt to go
beyond Western concepts and terminology to explore colour
in its synchronic and diachronic contexts. Chapter 1 by
Boric introduces the colourful Danube Gorges, where, he
argues, selected colours and designs were associated with
apotropaism or ‘enchantment’ in Mesolithic Neolithic
times. Keates (Chapter 5) concludes that in North Italian
society luminosity and colour of copper artefacts were
potent carriers of symbolic information. MacGregor
(Chapter 7) identifies colour and texture in the recumbent
stone circle tradition of northeast Scotland as expressions of
social identities.
Several papers (including the epilogue) demonstrate the
value of researching the innate attributes of colour in the
archaeological analysis of stone. The selective use of white
quartz pebbles in Neolithic monuments on the Isle of Man
is explained by Darville (Chapter 3) as a reflection of sacred
geography – continuing white symbolism bridged the
ideological gap between Christianity and earlier belief
systems. Cooney (Chapter 4) examines symbolic
associations of stone in axeheads of the Neolithic period as
a manifestation of a long tradition of colour symbolism in
Ireland.
Mortuary practice is a rich source of data on the
archaeology of colour. In one of the more succinct accounts
of the volume, Owoc (Chapter 6) demonstrates the
metaphorical power of colour as expressed in the selection
of soils in Bronze Age funerary practice. This is a model
study of the contrived appearance of a feature in a context
– the meaningful, deliberate and contextual construction of
colour through site design and use. It shows how the
addition of sequential embellishments and new mounds to a
funerary site (and the meaningful incorporation of features
induced by natural weathering) involve changes in colour,
texture, location, depth and consistency, which are imbued
with symbolic meaning. I am confident that this
explanation of site design has wider application, as in the
study of accumulated superimpositions in rock art. Andrew
Jones (Chapter 8) explores complex colour biographies of
funerary artefacts and produces an alternative explanation
of their significance. Tairov and Bushmakin (Chapter 9)
conduct a standard mineralogical analysis of cached
powders from burial mounds of South Urals and North
Kazakhstan. This provides useful data on paint use and
exchange systems, but I suspect does not constitute a ‘deep’
study of colour.
I especially enjoyed Allison’s imaginative way of
communicating the psychological impact of colourful wall
paintings in a redecorated Pompeian house (Chapter 10).
Allison shows that the selection of specific, well
documented colours and their careful arrangement in light
and architectural space point to a household of some wealth
and prestige. Saunders (Chapter 11) explores the same
attributes (light and colour) in Mesoamerican contexts. In a
sense the synthesised approach of this paper (and the
selective widening of the geographic scope to include
America) interrupts the organisational flow of the book. At
this point a study dealing with innate or consciously applied
colour in rock art would have complemented the preceding
chapter (wall paintings of Pompeii).
The final chapter (Epilogue by Scarre) provides a
review of colour studies which could have led to a useful
statement on the future direction of colour research instead
of another discussion of the salient features of stone in
prehistoric monuments. However Scarre’s conclusions
provide important guiding principles for colour studies in
archaeology, for example:
• the need to recognise the full materiality of the
artefacts concerned;
• colour may not be the most salient feature of materials
or artefacts.
In dealing with abstractions such as the creative
imagination, it is a challenge for archaeology to balance
what the editors describe as ‘the objective practice of data
recording and the hermeneutics of interpretation’. Although
I found a little too much emphasis on symbolic clichés (e.g.
red symbolises blood), and too little on taphonomy (loss
and/or changes in colour through various taphonomic
processes), recurring issues are of global interest: colour as
a temporal and spatial component of the natural
environment; culturally specific colour terminology and
selectivity as a source of insights into the processes of
symbolisations and categorisations; the universality of the
restricted colour palette; the introduction of novel materials
as a source of new colour perceptions and selection; the
importance of attributes (other than hue) such as texture,
luminosity, hardness, brightness, darkness and light; colour
perceptions in the use of stone (e.g. quartz) and metals (e.g.
copper); the meaningful, deliberate and contextual
77
Book Reviews
construction of colour and other qualities, as in technical
transformations to achieve lustre.
Overall, this volume makes an important contribution to
archaeology. Hopefully it will stimulate others to explore
the varied and complex ways in which past societies
perceived, selected, transformed and used colours to
transcend materiality. But the main contribution of
Colouring the Past is methodological – it has much to offer
archaeologists as an incentive to adopt integrated, cognitive
approaches to the analysis of material culture. The work of
Taçon (including his contribution to a series of short papers
on colour in the Cambridge Archaeological Journal 1999)
is widely cited in this volume, but it appears that Australian
archaeological studies which focus on the deep, innate
qualities of colour, texture, light etc. are few. Colouring the
Past confirms that the study of colour, ‘this compelling
attribute’ can be undertaken archaeologically, across a wide
range of temporal, spatial and material contexts.
A PACIFIC ODYSSEY: ARCHAEOLOGY AND
ANTHROPOLOGY IN THE WESTERN PACIFIC.
PAPERS IN HONOUR OF JIM SPECHT, edited by V.
Attenbrow and R. Fullagar. Sydney: Australian Museum.
Records of the Australian Museum, Supplement 29
(2004) 186 pages. ISBN 0-9750-4763-9 (paperback).
Price A$60.00.
Alison Crowther
The latest festschrift to honour the formal retirement of
yet another of Australia’s pioneering archaeologists brings
together 19 papers by 26 of Jim Specht’s friends and
colleagues. The ‘Specht-schrift’ stems from a one-day
conference organised by the Australian Museum in 2000
entitled, “A Pacific Odyssey: recent archaeological
discoveries, on the occasion of Jim Specht’s retirement”,
which aimed to present and discuss the results of important
recent discoveries in Pacific archaeology (p. v). Although
the lag between the conference and eventual publication of
the volume suggests that many of these discoveries may no
longer be so recent, the case studies highlight the range of
archaeological and anthropological research underway in
the Western Pacific.
Edited by Attenbrow and Fullagar, the volume
appropriately begins with a tribute to Specht’s career
(Taçon et al.) and a compilation showcasing his extensive
publication history (Kahn). Both highlight the depth and
breadth of Specht’s work in the Western Pacific and the
pioneering nature of much of his research, particularly in
the New Britain region - he is, after all, credited as being
the first archaeologist of the Bismarck Archipelago. The
remaining papers are organised alphabetically, by author,
owing to the lack of thematic grouping according to
geography, subject or object (p. v).
A strength of the volume lies in those papers that bring
new data to bear on old sites or old problems. Athens and
Ward, for example, tackle several issues associated with the
settlement history of Guam via palaeoenvironmental
analysis, while Denham evaluates the argument for
agriculture during Phase 1 at Kuk Swamp from new lines
of multi-disciplinary evidence. Phytoliths in sediments
excavated some 20 years ago from the Reber-Rakival
Lapita site provide Lentfer and Green with a means to
78
reconstruct past vegetation change. These and other papers
(e.g. Pavlides and Wilson) are a reminder that refining past
models and revisiting old sites are as important as finding
new ones.
Specht’s first assignment after joining the ANU in 1965
was to follow up the discovery of Lapita pottery on Watom
Island (Taçon et al.). The scope of current Lapita research is
indicated in papers by Torrence, who argues that a preLapita stemmed tool found by Specht in West New Britain
is evidence for the in situ development of a prestige
economy in the Bismarck Archipelago; and Summerhayes,
who presents recent data on sourcing obsidian from Lapita
sites in Anir. In separate papers, Spriggs and Lilley address
the issues of continuity and connection in the post-Lapita
sequences. Spriggs’ review of the debate surrounding
similarities in post-Lapita pottery is particularly useful after
Bedford and Clark (2001) threw a spanner in the works at
one of the last Lapita conferences. Although Spriggs
somewhat conservatively agrees with their claims, arguing
that “both Bedford and Clark are overstating their case, but
perhaps not by much” (p. 142), he argues that better dated
and described assemblages are the key to resolving this
debate. The possibilities for post-Lapita cultural continuity
discussed by Spriggs contrasts with those raised in Lilley’s
paper on the Vitiaz Strait region. Here the archaeological
sequences are characterised by periods of movement and
abandonment associated with highly disruptive local
volcanic activity, rather than continuous occupation and
pottery production.
Another theme that emerges from the volume is that of
repatriation. As Taçon et al. (p. 5) note, the Australian
Museum has been recognised as a world leader in the return
of cultural property, and Specht has been at the forefront of
this movement. Papers by Bonshek and Bolton build on
Specht’s repatriation efforts by presenting case studies (both
successful and failed) on attempts to return items to
Vanuatu and the Solomon Islands. On the flip-side of the
coin, Knowles and Gosden review a century of collecting in
New Britain, a place where Specht himself spent many
years as an archaeological ‘collector’. Each of these papers
examines the many complex relationships involved in the
collection/repatriation process and the role of museums in
mediating these social processes.
The rest of the papers vary in content from subsistence
studies (Galipaud and Swadling), to the analysis of
monuments (Smith), and the integration of oral tradition in
archaeological research (Sheppard et al.). I found Sand’s
paper on unravelling the ‘mystery’ of Walpole Island to be
particularly valuable as it presents material not previously
published in English. Sheppard et al. also present a highly
informative and interesting review of the use of oral
traditions in archaeology. They use their case study in
Roviana Lagoon to demonstrate that, when applied
critically and in tandem with archaeological research, oral
histories can provide a powerful explanatory tool for
understanding the past.
The value of “A Pacific Odyssey” lies not in lengthy
reviews but in the presentation of succinct case studies that
illustrate the innovative approaches of today’s Pacific
researchers. Given the variety of themes, issues and
methodologies presented, readers should have no difficulty
finding something of interest in this volume, even if their
own work lies outside the Pacific. That the editors have
relied on the volume’s geographical focus and occasional
Book Reviews
links to Specht’s career to thread the papers together is
probably its main fault, and only a minor one at that. I think
a final summation to draw the papers together and position
them within the current state of research would have been
useful for demonstrating their value beyond just being a
tribute. They do, after all, report some major contributions
to our understanding of aspects of Pacific prehistory.
Published as part of the Records of the Australian
Museum, Supplements series, the volume’s production is
professional, typographical errors are minimal (although
not absent, for example Torrence’s quote of Kirch on p. 170)
and its cost is comparable with similar archaeological
series, such as Terra Australis. Those who have entered the
digital age can even access a copy of each paper on the
Australian Museum website in Portable Document Format
(PDF), enhancing the volume’s accessibility and portability.
Overall, an excellent resource, a must have for anyone
keeping up with advances in Pacific archaeology, and an
excellent tribute to Specht’s brilliant career.
References
Bedford, S. and G. Clark 2001 The rise and rise of the incised and
applied relief tradition: A review and assessment. In G. R.
Clark, A. J. Anderson and T. Vunidilo (eds) The Archaeology
of Lapita Dispersal in Oceania: Papers from the Fourth
Lapita Conference, June 2000, Canberra, Australia, pp 6174. Canberra: Pandanus Books. Terra Australia 17.
ARCHAEOLOGY AND COLONIALISM:
CULTURAL CONTACT FROM 5000 BC TO THE
PRESENT, Chris Gosden. Cambridge University Press
(2004), 200 pages, 18 line diagrams, 4 half-tones and 5
tables. ISBN 0-521-78795-5 (paperback). Price: $49.95.
THE ARCHAEOLOGY OF THE COLONIZED,
Michael Given. London and New York: Routledge
(2004), 200 pages, 21 line diagrams and 19 black and
white photographs. ISBN 0-4153-6992-4 (paperback).
Price £18.99.
REVIEW ESSAY: ARCHAEOLOGIES OF
COLONIALISM
Rodney Harrison
Colonialism and culture contact have become hot topics
in archaeology. These two new books demonstrate the
breadth of the field, and distinguish themselves by dealing
with issues of colonialism, as distinct from culture contact
(see Silliman 2005), head-on. In doing so they set a series
of agendas for both archaeology and broader studies of
colonialism in the new Millennium.
Gosden’s Archaeology and Colonialism is broad in
scope, taking what is essentially a ‘top down’ approach to
the topic of colonialism. Drawing on the work of World
Systems theorists, in particular Wallerstein (1974, 1980)
and Frank and Gills (2000), Gosden suggests that it is
possible to see forms of colonialism in the archaeological
record from as early as 5000 BC. His most significant
contribution is the development of a model of colonialism
which manifests itself in three different forms, each one
then forming the basis for a global comparative discussion
in subsequent chapters, with a concluding chapter
focussing on ‘power’. ‘Colonialism within a shared cultural
milieu’ is the term he uses to describe the earliest (and
perhaps most controversial in terms of their acceptance as
colonial contexts) colonial forms, such as those that existed
from Mesopotamia to the Greeks, and amongst the Aztecs,
Incas, Chinese and Tongans. This form of colonialism is
characterised by colonial relations between state and nonstate polities, where power is manifest in forms that operate
within norms of social behaviour and where the limits of
colonisation are controlled not by military power, but by the
area over which a particular culture is shared or spread.
Colonial ‘middle grounds’ are exemplified by the
peripheries of the Greek colonies and the Roman Empire,
and early modern contacts with indigenous peoples in
North America, Africa, India and the Pacific. These forms
of colonialism are characterised by experimentation and
creativity, and consist of accommodation on the part of both
indigenes and colonists and the development of regularised
social relations. ‘Terra nullius’ colonialism, as evidenced by
the major settler societies of Australia, New Zealand, North
America and Russia from the mid-Eighteenth century, but
also by the Mongols and Spanish in Peru and Mexico, is
characterised by extreme violence, mass appropriation of
land, and the spread of disease which enables the
destruction of existing forms of social relations. Active
resistance to colonial forms which allow indigenous
cultural continuity often exists in such circumstances. Terra
nullius colonialism is differentiated from the other forms by
the existence of relatively fixed categories of difference,
whereas in middle ground colonialism new categories of
difference are often created by cross-cultural encounters
and in the case of colonialism in a shared cultural milieu, no
such categories of difference may exist.
Gosden’s work is informed by a judicious reading of
post-colonial theory, particularly in terms of understanding
the intellectual traditions that underlie the concept of
colonialism, and how nineteenth century views of
colonialism influence strongly the way in which it has been
studied by academics. He is, like other archaeologists
currently working in the field of culture contact (e.g. papers
in Torrence and Clarke 2000), strongly influenced by the
work of Nicholas Thomas (1991, 1994), who’s ideas about
the way in which colonialism is mediated by way of
material things echo in Gosden’s concluding statement that:
Colonialism is not many things, but just one.
Colonialism is a process by which things shape
people, rather than the reverse. Colonialism exists
where material culture moves people, both culturally
and physically, leading them to expand
geographically, to accept new material forms and to
set up power structures around a desire for material
culture (2004: 153).
As in his other recent book Collecting Colonialism:
Material Culture and Colonial Change (Gosden and
Knowles 2001), by focussing on the material dimensions of
colonialism, Gosden is authoring a specific role for
archaeology and material culture studies in understanding
the range of forms in which colonialism is manifest in both
modern and ancient human societies. Gosden’s grasp of his
topic is noteworthy, and the book is written in the sort of
clear and engaging style that would make it equally
accessible to both students and academics. This
impressively broad look at the many and diverse forms of
79
Book Reviews
colonialism in the post 5000 BC world not only
distinguishes itself by developing a convincing argument
regarding the various forms of colonialism, but also sets a
new agenda for global studies of colonialism, and
archaeology’s place in such an endeavour.
In contrast with Archaeology and Colonialism, Given’s
interest in The Archaeology of the Colonized is with the
lived experience of colonialism for those on the receiving
end of colonial regimes, a ‘bottom up’ approach to
understanding the materiality of the colonial project.
Given’s focus is narrower than Gosden, both geographically
and in terms of the forms of colonialism with which he
deals. Most examples are drawn from Turkey and Cyprus,
although he also cites material from Old Kingdom Egypt,
Nazi Germany and nineteenth century Scotland. He is most
interested in those particular forms of colonialism which
express themselves through taxation and the material
manifestations of Empire. This is not necessarily a bad
thing, and the material certainly acts in ways which are
complimentary to Gosden’s book, by giving greater focus to
a series of more detailed case studies, rather than
Archaeology and Colonialism’s broad perspective.
Like Gosden (e.g. 2004: 5; see also Silliman 2005),
Given sees colonialism as distinct from other forms of
cultural contacts in terms of the unequal power relations
that characterise colonial systems. Unlike Gosden, Given
makes a fairly convincing stab at reviving the sort of
‘Resistance’ model outlined in the influential Domination
and Resistance (Miller et al 1998), and other major works
such as McGuire and Paynter’s edited volume The
Archaeology of Inequality (1991). While Gosden, like other
recent authors on the archaeology of culture-contact (eg
papers in Torrence and Clarke 2000), is critical of the merits
of this approach in a study of colonialism, suggesting it
prejudices the terms of the encounter and may only be
relevant to some models of colonialism (2004: 22), it seems
appropriate to Given’s aim to focus on the ways in which
archaeology can contribute to studies of colonialism by
identifying specific instances of social impacts brought
about by particular colonial regimes:
The most direct involvement of ordinary people with
imperial rule is when their hard-won food is
removed from in front of them and taken right out of
their family, their community, and often their
country… this is colonialism, as experienced by the
great majority of people who live under it. Tribute
begins at the threshing floor (2004: 3).
Given’s study is framed in terms of landscape
archaeology, drawing, amongst other studies, on the work
undertaken as part of the Sydney Cyprus Survey Project (eg
Given and Knapp 2003). He employs small vignettes and
descriptive passages as narrative devices at various points
throughout the text to address issues of gaps in the record,
‘to imagine a lost perspective, form new questions and
stimulate new thought’ (2004: 23). This approach is
effective in drawing out the agency of individuals in the
colonial contexts discussed, where actions of resistance
might otherwise be extremely difficult to read (see also
Paterson’s (2003) discussion of the ‘texture of agency’ in a
central Australian context).
The Archaeology of the Colonized’s chapters focus on
the archaeology of taxation (Chapter 3), the role of census,
survey and mapping in the colonial project (an issue which
80
has recently been discussed in relation to the archaeology of
colonialism in Australia by Byrne 2003)(Chapter 5),
imperial landscapes (Chapter 4), forced labour (Chapter 6),
the resistance of taxation (Chapter 7), and the ways in which
resistance might be ‘read’ in the archaeological landscape
(Chapter 8). This book connects with postcolonial theory
through its focus on agency, and provides important insights
into the ways in which local people engage with systems of
centralised control, and how these relationships might be
understood through the archaeological record. Its real
strength is its engagement with landscape archaeology, and
the thorough grounding of the book in a series of appealing
(at times even surprising) and inventive case studies.
Given’s writing style is innovative and engaging, and the
topic one which is both interesting and important.
I think these two books would act in compliment as part
of an undergraduate course on historical archaeology or the
archaeology of colonialism and culture contact, as each
offers a uniquely different perspective. While Gosden’s
book is broad and comparative, and develops important
insights on this basis of this perspective, Given’s addresses
what could be read as a gap in Archaeology and
Colonialism (but certainly not in Gosden’s other work such
as Collecting Colonialism) in drawing out the experience of
colonialism for ordinary people who exist within colonial
regimes. Along with a suite of other recent books on the
archaeology of colonialism (Lyons and Papadopoulos
2002), culture contact (Fagan 1998; Torrence and Clarke
2000; Murray 2004) or the historical archaeology of
indigenous peoples (cf. Murray 1996; Harrison and
Williamson 2002) with relevance to Australia, these
volumes contribute to a renewed interest from within
archaeology on issues of contemporary relevance to both
indigenous and non-indigenous peoples in settler societies,
and to the development of a truly global perspective on
world history. As Gosden notes, ‘colonialism is the major
cultural fact of the last 500 years, and to some extent of the
last 5000 years, although it is said we now live in a postcolonial world… we are still wrestling with the economic,
intellectual and social consequences… by looking at the
varying forms power can take we can learn much about the
past and unlearn much about the present’ (2004: 6, original
emphasis). These important books merit a place on the
shelves of all scholars interested in the role of archaeology
in developing research which has relevance to the
experiences of ordinary people in both the modern and
ancient worlds, and in teaching students the ways in which
archaeology can contribute to contemporary debates
regarding modernisation, postcolonial theory, globalization
and the oppression of the modern nation-state.
References
Byrne, D. 2003 Nervous landscapes: Race and space in Australia.
Journal of Social Archaeology 3(2):169-193.
Fagan, B. 1998 Clash of Cultures (2nd ed.) Walnut Creek,
California: Alta Mira Press.
Frank, A. G. and Gills, B. K. 2000 The five thousand year world
system in theory and praxis. In R. A. Denemark, J.
Friedman, B.K. Gills and G. Modelski (eds) World System
History: The Social Science of Long- term Change, pp. 3-23.
London: Routledge.
Given, M. and Knapp, A. B. (eds) 2003 The Sydney Cyprus Survey
Project: Social Approaches to Regional Archaeological
Survey. Los Angeles: University of California Cotsen
Institute of Archaeology. Monumenta Archaeologica 21.
Book Reviews
Gosden, C. and C. Knowles 2001 Collecting Colonialism:
Material Culture and Colonial Change. Oxford: Berg.
Harrison, R. and C. Williamson (eds) 2002 After Captain Cook:
The Archaeology of the Recent Indigenous Past in Australia.
Sydney: Archaeological Computing Laboratory, University
of Sydney. Sydney University Archaeology Methods Series 8.
Lyons, C. L. and Papadopoulos, J. K. (eds) 2002 The Archaeology
of Colonialism. Los Angeles: Getty Research Institute.
McGuire, R. H. and Paynter, R. (eds) 1991 The Archaeology of
Inequality. Cambridge, Massachusetts: Blackwell.
Miller, D., Rowlands, M. and Tilley, C. (eds) 1989 Domination and
Resistance. London: Unwin Hyman.
Murray, T. 1996 Contact archaeology: Shared histories? Shared
identities? In S. Hunt and J. Lydon (eds) SITES. Nailing the
Debate: Interpretation in Museums, pp. 199-213. Sydney:
Historic Houses Trust of New South Wales.
Murray, T. (ed) 2004 The Archaeology of Contact in Settler
Societies. Cambridge: Cambridge University Press.
Paterson, A. 2003 The texture of Agency: an example of culturecontact in central Australia. Archaeology in Oceania
38(2):52-65.
Silliman, S. W. 2005 (in press) Culture contact or colonialism?
Challenges in the archaeology of native North America.
American Antiquity 70(1).
Thomas, N. 1991 Entangled Objects: Exchange, Material Culture
and Colonialism in the Pacific. Cambridge, Massachusetts:
Harvard University Press.
Thomas, N. 1994 Colonialism’s Culture: Anthropology, Travel and
Government. Cambridge: Polity Press.
Torrence, R. and Clarke, A. (eds) 2000 The Archaeology of
Difference: Negotiating Cross-cultural Engagements in
Oceania: London: Routledge.
Wallerstein, I. 1974 The Modern World System, I. New York:
Academic Press.
Wallerstein, I. 1980 The Modern World System, II. New York:
Academic Press.
HUNTER-GATHERERS IN HISTORY,
ARCHAEOLOGY AND ANTHROPOLOGY, edited
by Alan Barnard. Oxford and New York: Berg (2004).
x+278 pages, 2 figures and 2 tables. ISBN 1-85973825-7 (paperback). Price US$26.95.
Ian McNiven
This volume is based on papers presented in the
‘Hunting and Gathering as a Theme in the History of
Anthropology’ session at the Ninth International
Conference on Hunting and Gathering Societies held at
Edinburgh in September 2003. The editor Alan Barnard is
well placed to assemble this collection following the
success of his earlier book History and Theory in
Anthropology (Cambridge University Press, 2000). On the
back cover, Tim Ingold writes ‘Alan Barnard has assembled
some outstanding contributions, providing a benchmark
assessment of the past achievements and future prospects
of hunter-gatherer research’. The key word here is ‘some’,
for as with most edited volumes, we are presented with an
eclectic group of papers that vary in quality. Yet the
thematic diversity of the papers mirrors nicely the
multiplicity of directions hunter-gatherer studies have
taken over the last 20 years. Whereas the landmark 1966
Man the Hunter conference at the University of Chicago
attempted to bring about an Anglophone channeling of
hunter-gatherer studies, the ensuing rise of 4th world
peoples’ activism, globalization, postmodernism and
postcolonialism have seen this authority challenged and
fragment. Barnard has done well to capture some of the
diversity and challenges confronting modern huntergatherer studies.
Barnard divided the 17 chapters into three parts: (1.)
Early visions of hunter-gatherer society and their influence,
(2.) Local traditions in hunter-gatherer research, and (3.)
Reinterpretations in archaeology, anthropology and the
history of the disciplines. Part 1 presents four papers
exploring the origins of the concept of ‘hunter-gatherers’
amongst various scholarly traditions. Pluciennik and
Barnard provide comparative historical overviews of
European and Asian approaches to classifying and ranking
societies. While both papers are excellent, the opportunity
was missed to explore in depth the colonial fabric of such
constructions. The hypothetical concept of hunter-gatherers
was developed within the broader theoretical framework of
social evolutionism as a philosophy of colonialism. Huntergatherers were invented as a category of humanity to
position many of the world’s Indigenous peoples as an
anachronistic form of ‘primordial man’ (‘savages’) who
beseeched conquest and extermination. It is no coincidence
that the two key periods of Western scholarship on
development of the hypothetical concept of huntergatherers – Ancient Greece and Rome, and 18th and 19th
century western Europe, correspond to the two key periods
of European intercontinental colonialism (for an extended
discussion of this issue, see McNiven & Russell
Appropriated Pasts: Indigenous Peoples and the Colonial
Culture of Archaeology, AltaMira Press forthcoming). All
hunter-gatherer specialists are aware of the basics of this
complex and often sordid history. Perhaps this is why
hunter-gatherer studies continue to be re-invented in an
ever-ending attempt to shed the ghosts of social
evolutionism. But for hunter-gatherer studies, social
evolutionism it not so much a ghost but a shadow whose
form can be changed but never eradicated.
Part 2 features six papers that discuss German, Russian,
Japanese and Indian anthropological traditions of huntergatherer studies. Peter Schweitzer in his paper on Germanlanguage debates of the late 19th and early 20th centuries
speaks for the entire group when he states that the aim of his
paper is to ‘counter… English-language bias when
discussing the history of hunter-gatherer studies’.
Schweitzer brings to our attention Ernst Grosse who in the
1890s not only wrote major critiques of Herbert Spencer
and Lewis Henry Morgan, but also distinguished ‘lower
hunters’ from ‘higher hunters’ in a schema that ‘reappeared’
in the late 20th century with immediate/delayed return and
simple/complex hunter-gatherers. Artemova and Sirina in
their respective papers bring out the role of state politics and
ideology in Russian/Soviet traditions of hunter-gatherer
studies. Their papers bring home the need for similar
reflexive appraisals of the political and ideological
backdrops to the development of the hunter-gatherer
concept by Enlightenment scholars of western Europe
(including Great Britain). Papers by Ichikawa and Sugawara
provide informative overviews of Japanese scholarship in
central Africa (‘pygmies’) and southern Africa (San
peoples). Pappu’s paper provides a detailed and at times
critical overview of the use of ethnographic analogy in
Indian hunter-gatherer archaeology.
81
Book Reviews
Part 3 presents the most challenging and engaging
papers in the volume. Lane and Schadla-Hall discuss the
intriguing question of why during four decades of reexamination and reinterpretation of the famous Mesolithic
site of Star Carr that the research questions have not moved
on from issues of subsistence and settlement to embrace
new approaches to hunter-gatherers provided by
anthropologists. No book on hunter-gatherers can neglect
optimal foraging theory and Sheehan broadens the
discussion to consider issues of analogy and the empirical
limitations of applying OFT to the past. Suzman and
Widlok discuss issues of historicity and analogy
respectively with regards to Kalahari peoples. Suzman’s
essentialist argument for the ahistorical nature of some
Kalahari peoples is somewhat naïve as it fails to engage
with literature on the mnemonic role of landscape and
places in the emic construction of Indigenous histories.
For me, papers by Yengoyan, Myers, and Pinkoski and
Asch on the theoretical and legal legacies of Julian
Steward’s research on the Great Basin Shoshone and Paiute
are the highlights of the volume. Steward’s paradigm of
cultural ecology transformed hunter-gatherer studies and
was foundational to the New Archaeology of the 1960s and
70s. But as Myers notes, Steward’s ‘culture core’ concept
dismissed religion, ritual and social structure as
epiphenomenal and marginal to understanding the so-called
core issue of subsistence and environmental adaptation.
Such a view alienated Native Americans and, as Yengoyan
states bluntly, ‘created a new discourse in which
evolutionary and economic models have reduced human
actors to disemboweled humans who no longer have
cultural anchors’. Pinkoski and Asch document the
insidious impact of Steward’s approach as the US
Department of Justice used his theoretical approach to
undermine Northern Paiute land claims in the 1940s. This
case study is mandatory reading for all those interested in
the link between anthropological constructs and native title.
But these three papers are not the only reason I recommend
this book to all those interested in hunter-gatherer studies.
The volume contains a stimulating mix of papers that not
only de-centre Anglophone academic traditions but also
challenge us to think more broadly in our formulations of
those peoples past, present and future who we categorise as
hunter-gatherers.
CONSTRUCTING FRAMES OF REFERENCE: AN
ANALYTICAL METHOD FOR ARCHAEOLOGICAL
THEORY BUILDING USING ETHNOGRAPHIC AND
ENVIRONMENTAL DATA SETS, by Lewis R. Binford.
Berkeley: University of California Press (2001) xx + 563
pp. ISBN 0-5202-2393-4 (hardcover) Price US$75.00.
F. Donald Pate
Lewis Binford has devoted his professional career to the
development of a rigorous scientific framework for the
reconstruction of past human lifeways. During the past 40
years, Binford has made significant contributions to the
foundations of contemporary anthropological archaeology.
Binford’s major influence on the development of scientific
archaeology or the “New Archaeology” is reflected in his
seminal publications that emerged in the 1960s – 1980s
(Binford 1962, 1965, 1968, 1972, 1978, 1980, 1981, 1983,
82
1989; Binford and Binford 1968). The “New Archaeology”
and subsequent post-processual approaches resulted in
critical assessments of archaeological theories and methods
employed to make inferences about past human behaviour.
The epistemological dynamics associated with scientific
archaeology have had a major impact on the ongoing
evolution of archaeological theory and method.
This volume represents a landmark synthesis of
Binford’s approaches to the scientific analysis of past
human behaviours. “The primary problem that this book
addresses is the development of a method for productively
using ethnographic data to serve archaeological goals” (p.
3). Binford provides a detailed analysis of ethnographic and
environmental data from 339 historically known huntergatherer societies and argues that behavioural variability is
influenced significantly by a small number of
environmental and demographic variables. With a focus on
hunter-gatherer dynamics and the emergence of early
farming societies, the book is an essential reference for all
prehistorians. Binford’s critical assessment of the
employment of ethnographic data in studies of past huntergatherer societies makes this book particularly valuable to
Australian archaeologists.
Binford’s ‘frames of reference’ are methodological
devices for structuring data in order to make them useful to
archaeologists (p. 3). Two major frames of reference are
developed in the book. The first documents the primary
variables conditioning habitat variability and provides a
means to relate archaeological ‘facts’ to various
environmental variables. Thus, this frame of reference
addresses the documentation and identification of huntergatherer adaptive responses to habitat variability. In the
second frame of reference, variability documented among
ethnographically known hunter-gatherers is related to
archaeological remains. This frame of reference facilitates
the development of models relating archaeological remains
to past hunter-gatherer behaviour.
The book is divided into four parts. Part I Exploring
Prior Knowledge and Belief consists of three chapters that
summarize prior conceptions about hunter-gatherers.
Chapter 1 “Founders effect” and the study of huntergatherers assesses the utility of theories and methods
developed by the anthropological ‘founders’ of huntergatherer research. The role of past anthropological research
in relation to the continuing growth of archaeological
theory and method is addressed. Chapter 2 Human actors
and their role in the evolutionary play addresses the unique
characteristics of humans that influence responses to
environmental variables. Binford focuses on “planned or
volitional action and on the associated idea that goaldirected behavior, extreme behavioral plasticity, and the
capacity for culture itself are important human
characteristics that must be considered as constants in any
discussion of humans as actors in the ecological theater and
the evolutionary play” (p. 43). Chapter 3 The play of ideas
in the scientific theater discusses the important role that
science plays as a ‘learning strategy’ in relation to the
advancement of archaeological knowledge about human
behaviour in past societies. Two different kinds of analytical
tools, frames of reference and projections, are discussed. A
scientific investigation of the human past offers a
systematic and critical approach to the analysis of
archaeological data within the context of particular
environmental contexts. “The ability to tease out
Book Reviews
implications from repetitively patterned relationships
among variables (classes of data) provides the clues to the
way the world is organized and how it works in a dynamic
sense” (p. 48).
Part II Methods for Using Prior Knowledge: Building
Frames of Reference and Models consists of three chapters
that address the methods involved in the development of
frames of reference and models from ethnographic and
environmental data. “These three chapters outline the logic
and actual construction of intellectual models and frames of
reference and are central to the tactical exploration of the
procedures developed in subsequent chapters” (p. 4).
Chapter 4 Setting the stage for the evolutionary play: The
Earth’s climates, plants, and animals documents the
variability in environments in which hunter-gatherers are
known to have lived. Ecological variables that can be used
to record specific, relevant properties of global
environments are introduced and defined. These
environmental data provide the ecological basis for the
development of hypotheses regarding explanations for
ethnographic variability. Chapter 5 Designing frames of
reference and exploring projections integrates the
ecological variables documented in Chapter 4 and the
analytical tools (frames of reference and projections) as a
means to assess various aspects of hunter-gatherer
behaviour. For example, the distribution and population
density of hunter-gatherer communities across the globe is
related to net above-ground productivity of various plant
communities in these regions. In Chapter 6 Building a
baseline for analyzing niche variability among
ethnographically documented peoples Binford identifies
the range of habitats or ‘energetic domains’ within which
hunter-gatherers are known to have occupied. He then
employs a case study which approximates the subsistence
base, degree of mobility, and ethnic diversity that may have
characterised hunter-gatherer populations in particular
regions of Europe preceding agriculture.
Part III Recognizing Patterns and Generalizing about
What the World is Like consists of three chapters that
employ cross-cultural ethnographic comparisons of diverse
hunter-gatherer groups as a uniformitarian strategy to
reconstruct past hunter-gatherer behaviours. Chapter 7
Twenty-one generalizations in search of a theory employs
generalizations derived from pattern recognition studies to
build a model of factors that might relate to variability in
hunter-gatherer group size. In Chapter 8 A flat earth or a
“thick rotundity”?, the ethnographic hunter-gatherer data
set is employed to investigate how group size varies in a
wide range of circumstances. Variables addressed include
degree of mobility, social cooperation, food storage,
changes in the organisation of labour, and various means of
intensification in relation to food yields. Chapter 9 The
play’s the thing in the scientific theater addresses how
processes of intensification lead to increased sedentism and
changes in the organisation of hunter-gatherer cultural
systems.
Part IV Putting Ideas, Second-Order Derivative
Patterning, and Generalizations Together: Explorations in
Theory Building includes the final three chapters of the
book. This section of the book focuses on analysis and
integration in relation to the explanation of the observed
variability in the hunter-gatherer ethnographic data base.
Chapter 10 A disembodied observer looks at hunter-gatherer
responses to packing discusses the interaction of densityAustralian Archaeology, Number 60, 2005
dependent variables in relation to intensification processes.
Binford defines intensification as “the process that impels
hunter-gatherers to increase the amount of food they extract
from smaller and smaller segments of the landscape”
(p. 363). Intensification is linked to increases in population
size or ‘demographic packing’. It is argued that habitat
richness may condition the rate of demographic change and
that demographic packing is the factor forcing major
adaptive changes. Chapter 11 The evolution of system states
examines social complexity that arises from the integration
of previously independent systems. Ethnographic case
studies are employed to examine the causes of internal
social differentiation and ranking in hunter-gatherer
societies. Binford contrasts internally ranked and socially
stratified societies that were more dependent on aquatic
resources with hunter-gatherers primarily dependent upon
terrestrial plants who employed social differentiation based
on an individual’s progress through a series of age-graded
sodalities or ‘secret societies’. The former employed wealthbased distinctions, polygamous privileges for the elite, and
council-based decision making. In Chapter 12 The last act
crowns the play Binford considers the spatial and temporal
patterning of different system state conditions in order to
clarify the way in which archaeologists approach the
archaeological record. Density-dependent thresholds that
mark major interruptions in the character, complexity, and
tactical behaviour of hunter-gatherer groups are discussed in
an attempt to “build a theory about the self-organizing
processes of intensification among hunter-gatherers” (p.
461).
In the Epilogue, Binford discusses some of the broader
implications of major organizational changes that have
occurred when hunter-gatherer groups have reached and
exceeded the ‘packing threshold’. In addition, underlying
principles in relation to “the development of a method for
productively using ethnographic data in the service of
archaeological goals” are reviewed (p. 471).
Although major advances have been made in the
development of an anthropological archaeology, Binford
argues that the discipline remains in a nascient stage of
theory building. Constructing Frames of Reference makes a
significant contribution to ongoing theory building in
contemporary archaeology.
References
Binford, L. R. 1962 Archaeology as anthropology. American
Antiquity 28:217-225.
Binford, L. R. 1965 Archaeological systematics and the study of
culture process. American Antiquity 31:203-210.
Binford, L. R. 1968 Some comments on historical versus
processual archaeology. Southwestern Journal of
Anthropology 24:267-275.
Binford, L. R. 1972 An Archaeological Perspective. New York:
Seminar Press.
Binford, L. R. 1978 Nunamiut Ethnoarchaeology. New York:
Academic Press.
Binford, L. R. 1980 Willow smoke and dogs' tails: Huntergatherer settlement systems and archaeology. American
Antiquity 45:4-20.
Binford, L. R. 1981 Bones: Ancient Men and Modern Myths. New
York: Academic Press.
Binford, L. R. 1983 In Pursuit of the Past: Decoding the
Archaeological Record. London: Thames and Hudson.
Binford, L. R. 1989 Debating Archaeology. New York: Academic
Press.
Binford, S. R. and L. R. Binford (eds) 1968 New Perspectives in
Archaeology. Chicago: Aldine.
83
MINUTES OF THE 2004 ANNUAL GENERAL MEETING OF THE
AUSTRALIAN ARCHAEOLOGICAL ASSOCIATION Inc.
University of New England, Armidale
Tuesday, 14 December, 2004
1. Welcome
The President, Judith Field, welcomed members to the
AGM, and declared the meeting open at 5.35pm.
2. Apologies
Apologies were received from Richard Fullagar, Lesley
Head, Annie Ross, Isabel McBride, Pam Smith, and Ian Lilley.
3. Confirmation of the Minutes of the 2003 AAA AGM
The minutes of the 2003 AAA AGM held at the Station
Resort, Jindabyne, New South Wales, on 5 December 2003
were published in AA58:53-62. Motion: “that the 2003 AGM
minutes as published in AA58 be accepted.” Moved: Peter
White. Seconded: Pat Gaynor. Motion carried nem. con.
4. Business arising from the 2003 AGM
4.1 Proposed amendments to the AAA Code of Ethics
Sean Ulm representing Richard Fullagar, the Chair of the
Code of Ethics Review Subcommittee, and Nathan Woolford,
Indigenous representative of the Code of Ethics Review
Subcommittee spoke to this item. Sean explained the history
of the operation of the Subcommittee and provided a report of
progress during 2004. He stated that in 2002 Richard Fullagar
had raised the issue of amendments to the existing AAA Code
of Ethics which had been in operation for 13 years without
amendment, and was subsequently appointed Subcommittee
Chair. A co-opted Subcommittee of Sean Ulm, Ian Lilley,
Terry Blair, Nathan Woolford, Annie Ross, Mike Williams,
Peter Veth, Jo McDonald, Tim Murray, and Jane Anderson was
later established.
Proposed changes to the AAA Code of Ethics were
emailed to members in 2003 and covered the whole spectrum
of the existing document. These proposed changes were taken
to the 2003 Jindabyne conference where extensive reworking
to the clauses occurred. Submissions had been made to the
Subcommittee during 2004. This included the issue of the
constitution of the committee and its duties. The Code of
Ethics had been adopted subject to legal advice at the
Jindabyne conference.
Sean discussed that Jane Anderson had raised the issue of
the Code of Ethics codifying intellectual property was already
subject to individual agreements that archaeologists have
established with Indigenous communities and was
unnecessary in the Code of ethics. A particular problem was
raised in relation to clause 3.1 which gave primacy which was
counter to that position.
Comments on the Code of ethics by various members had
been obtained by Sean and these were presented. Clause 3.1
was then read in its present state “Members acknowledge the
primacy of Indigenous knowledge, intellectual property and
cultural rights in respect of Indigenous heritage and the
following articles reflect this principle”. The comments
received by Sean all centered on that this clause neither
recognised different domains of knowledge, and that in certain
84
situations expert archaeological advice would be invalidated by
the clause. The Subcommittee thought it necessary to change
clause 3.1.
Four different options were presented for voting at the 2004
AGM:
Option 1
• Members acknowledge the authority of Indigenous
knowledge, and recognise the Indigenous rights in their
cultural heritage. The following articles reflect this
principle.
Option 2
• Members acknowledge the authority of Indigenous
knowledge, and recognise the primacy of Indigenous
rights in their cultural heritage. The following articles
reflect this principle.
Option 3
• Delete clause 3.1 and renumber remaining clauses in
Section 3.
Option 4
• No change. Keep existing Code of Ethics Section 3.1.
• Members acknowledge the primacy of Indigenous
knowledge, intellectual property and cultural rights in
respect of Indigenous heritage and the following articles
reflect this principle.
Nathan Woolford, acting as Indigenous Representative on
the Code of Ethics Subcommittee spoke on these proposed
options. Nathan expressed disappointment that he was not able
to attend the 2003 AGM and stated that he did not support
clause 3.1 as it stood. He discussed his belief that
archaeological and Aboriginal knowledge were not the same
thing, and that it was wrong to circumscribe the discipline in
this respect. Rather than recognizing primacy, Nathan stated
that AAA should rather respect Aboriginal culture. He found it
patronizing that this clause existed, and pointed out that
archaeologists and Indigenous views are often different. He
asked the AGM to remove clause 3.1.
Comments were then taken from the floor. Steven Price
suggested choosing option 3 ‘in principle’ for the next 12
months, so that there was a working Code of Ethics.
Motion 1: Judith Field moved that option 4 be adopted;
that there be no change to the clause 3.1. Seconded: Jo
McDonald. Motion dismissed unanimously.
Jo McDonald asked whether the Code of ethics was
already an ‘open’ code.
Sean Ulm answered that it was, and that it could be
changed at any time, and that we could change the whole thing
at next years AGM if we wanted.
Motion 2: option 3; Delete clause 3.1 and renumber
remaining clauses in Section 3. Moved: Peter White. Seconded
Pip Rath. Those for 65, against 9. Motion carried.
4.2 Aboriginal Issues
Steven Free spoke on this issue. He acknowledged the
traditional owners of the area and provided a background on
himself, and then outlined three issues that he wanted to cover,
i) the formation of IAA (Indigenous Archaeologists
2004 AGM minutes
Association) this year with a membership of eight in their core
group, ii) the AAA Code of Ethics, iii) Indigenous heritage
issues. Steven suggested that IAA wanted to form a close
alignment with AAA, but had not been consulted by AAA at
any stage so far. He then discussed a statement written by Dave
Johnston which alleged possible breaches of the AAA Code of
Ethics by unnamed person(s), which he wished to table with
the AAA executive.
Steven stated that IAA was an association for all
archaeologists. There were different categories of membership,
and that all were welcome. He mentioned that with concern
with the AAA Code of Ethics some people had not wanted to
join AAA.
The statement by Dave Johnston was not tabled to the
AAA executive. Steven stated that the President of AAA had
undertaken to consult with him on this matter. Steven then read
a list of members of IAA.
Ken Markwell then spoke to the AGM. He said that he
was not speaking for all, just based on his experience at the
2004 conference. He experienced archaeology at ’the other
end’ teaching in the country. Ken stated that he had attended
numerous archaeology papers and that introductions consisted
of housekeeping, and introduction on what archaeology was,
and there was no acknowledgement of traditional owners. He
stated that in the course of the conference such
acknowledgement only happened five times. Ken stated that
we need to move beyond the question of who owned the past,
to who belongs to the past. He cited the need for associate
lecturers to teach Aboriginal people as a good idea, that
Aboriginal knowledge was another side, and that he had earnt
that knowledge.
Ken stated that there was an inherent need for teachers to
have Aboriginal people with qualifications but also earnt
knowledge. Australian Universities needed a process and had a
responsibility to teach with Aboriginal People in institutions,
and that we should not subject other people to the same things
that he has personally been through. Ken said that as a
Commonwealth employee they have an Indigenous recruitment
policy, and there was no excuse not to have a similar process
now.
Sue Hudson spoke stating that she was both a Rotarian
and an Aborigine. She stated that there was never a proper
welcome to country at the conference and no smoking
ceremony. She believed that there was only a token Aboriginal
representation at the conference and in the presentation of
papers, and this was a huge ongoing problem. No Aboriginal
views had in her opinion been incorporated into papers at the
AAA conference.
Robyn Bancroft spoke on behalf of Dave Johnston. She
said he had worked hard on the Code of Ethics and provided a
lot of support, and that AAA should have shown him greater
respect in relation to this, and that he should have been
included more.
Sean Ulm said that it was unfair to say that Indigenous
People had not been included in the Code of Ethics
Subcommittee, and that there were Aboriginal representatives
on the Subcommittee. Sean stated that Dave Johnston had an
opportunity to comment on the Code of Ethics and there had
been long meetings on this matter.
Robyn Bancroft replied that she appreciated that there had
been Indigenous input, but that it should have been pushed
harder.
Judith Field said that she would look at the composition of
the Code of Ethics Subcommittee and look for greater
Indigenous input.
Robyn Bancroft replied that Aboriginal groups had a good
idea of who worked with communities and who did not, and the
impact that it causes when archaeologists make mistakes.
Judith Field agreed and encouraged Aboriginal attendance
at AAA conferences. She noted that it is unfortunately very
expensive and took a great deal of organization to bring
Aboriginal communities to the conference, but that all
members of AAA would like Aboriginal representatives in
attendance.
Steve Free suggested that various organizations combine
funding for Aboriginal attendees.
Peter Veth noted that at AIATSIS conferences this was a
mandate, and that they aimed at maximizing Indigenous
participation. More lobbying was needed and the more we can
combine things the better. He also noted the ARC Indigenous
Research Support Scheme, and Mandy Thomas at AIATSIS as
a contact for post graduate and post doctoral funding for
Indigenous people.
Ken Markwell asked that AAA ask all Universities that
teach archaeology to report back on the progress of including
Indigenous aspects to the AAA AGM.
Jo McDonald stated that there was already a teaching and
Learning Subcommittee co opted by AAA.
Judith Field asked whether there was a program to develop
this Subcommittee and noted there is a need to present ways for
getting Indigenous People to AAA conferences. She stated that
conference attendance fees should be automatically waived and
that this issue would be worked through over the next 12
months.
Matthew Spriggs spoke regarding the AAA Teaching and
Learning Subcommittee. He said that AAA was an association
and not a University and that any checking up on Universities
sounded prescriptive. He stated that it was difficult to compel
any University to report on such matters.
Judith Field replied saying that she will pursue assistance
of getting Indigenous representatives to conferences and not
the issue of reporting on Universities.
Sean Ulm provided a counterpoint to these issues. He
asked whether the best place for raising issues of University
teaching might be in the Indigenous Archaeology Association,
and that AAA already had a role for Indigenous representation.
He believed there was no need for another Subcommittee and
the various issues raised should be fed into existing
Subcommittees.
Ken Markwell stated that he had not asked for a motion
and that if there are existing Subcommittees then he was
satisfied. He stated that he was originally asking about funding
in Universities, their role, and what they were actually doing in
respect of Indigenous concerns.
Sue Solomon stated that UNE had an existing indigenous
employment strategy.
4.3 Public Liability Insurance
AAA Treasurer, Joe Dortch spoke on this matter. He
stated that AAA had acquired Public Liability Insurance of $5
million and Association Liability Insurance of up to $1 million.
The costs of these had been $2135 and $802 respectively for
the year.
4.4 Kenniff Cave Conservation
Catherine Westcott and Luke Godwin spoke on this
matter. Catherine stated that at present Kenniff Cave had been
closed to the public under inherent threat of collapse following
85
2004 AGM minutes
a report by geologists. There had been correspondence with
the Queensland Minister for Environment regarding this
matter. Parks responses besides closing the site had been to
remove Kenniff Cave from all park brochures, however
there had been attempt to address the potential loss of a site
of immense archaeological significance. There was no
further intention to do anything to the site to ensure its
protection. Parks staff had replicated photographs of Walsh
in order to monitor changes but that was all.
Luke Godwin stated that this was a second rate
response based on misinformed management and then read
a prepared statement. In this statement the assertion that the
potential collapse due to natural processes was ridiculous,
and there should always be support for both the historical
and natural environment. He noted that other funding had
been available for environmental studies whilst Kenniff cave
had been neglected. The responses had been idiosyncratic
and the idea there was no money to look after this iconic site
was ridiculous. He stated that the consultative process had
been dubious and questioned whether independent advice
had indeed been obtained. Luke noted that the local
traditional owners approve of the closure but that their
decision should have been for cultural reasons and not
forced on them because of potential danger.
Luke suggested that the response of the Queensland
Government was consistent with their approach to cultural
heritage, that they possessed an unwillingness to commit to
resource something they claim ownership over. He
suggested that the local interpretive photos were inadequate
and ludicrous, as a cultural resource strategy, for a site that
is the oldest site in the Queensland highlands, and that the
Queensland Parks Service had missed much of the
significance of the site – the deposit. He suggested salvage
excavation possibilities, and photogrammetric survey, and
that a general cultural heritage management plan was
necessary for an open process.
Catherine Westcott agreed with the concerns of Luke
Godwin, but thought it unfair to target the interpretive
officer, suggesting that people were doing what they could.
Luke Godwin stated that the Queensland Government had
provided responses only at the lowest level.
Judith Field said that AAA had received no responses
to correspondence.
John Mulvaney stated that Kenniff Cave constituted the
first excavated site in Australia with a stratified sequence
dating to the Pleistocene and based on that it should demand
recognition.
Judith Field said that this issue required addressing at a
higher level than had so far been achieved.
John Mulvaney suggested that the Australian Heritage
Council should be approached.
Judy Birmingham suggested that a motion by put to
affirm the significance of and concern for such an iconic
site, so that something could be triggered.
Josephine Flood stated that AAA should put forward a
case for registration of the site with the Heritage Council so
as to short circuit the process, and the bring the issue to the
medias attention.
Judith Field proposed the motion that “the Australian
Archaeology Association is deeply alarmed at the
threatened state of Kenniff Cave in view of its iconic
significance as a place in Australian archaeology” Moved:
Judy Birmingham. Seconded: Sue Hudson. Motion carried
nem. con.
86
Steve Brown stated that AAA needed to follow up on
heritage listing the Kenniff Cave site. He proposed the
motion that “the Australian Archaeology Association submit
a nomination for Kenniff Cave to the Australian Heritage
Council for listing” Moved: Steve Brown. Seconded: Jo
McDonald. Motion carried nem. con.
Judith Field co opted a Subcommittee for matters
regarding this issue and appointed John Mulvaney,
Catherine Westcott and Luke Godwin to this Subcommittee.
5. Reports
5.1 President’s Report (Judith Field)
In my first year as president I have continued reassured
by the dynamic and inclusive group of people who are
committed to the discipline and have been always willing to
be involved in issues that arise and need professional input.
One of the most significant achievements 2004 has been
the completion of the digital archive of the journal. I think
that this marks an important move to make that journal more
accessible and as a result I expect an increase in citation rate.
The last 30 years have marked considerable development and
change within the association and the production of this
important compilation ensures that accessibility of the work
documented in those pages. Michael Haslam, Sean Ulm and
Luke Kirkwood require a special mention as it is a product
of their time and commitment.
The new AAA website has also recently been loaded
onto the web and this provides a very polished and easily
navigated public interface for the association. Luke
association is indebted to him.
Archaeological sites continued to be threatened either
by development proposals or the vagaries of nature, as has
been the case for Ngarrabullgan and Kenniff Cave. As per
usual either finding out details from the appropriate
authorities and or interacting with them has been challenging
to say the least. Luke Godwin has been extremely helpful in
his advice and help in trying to get a coherent response from
the relevant authorities.
In 2004 AAA was asked to contribute to the selection of
individuals for the expert panel of the ARC. This turned out
to be a complicated and drawn out process. While this year
we felt unable to nominate any one individual, not because
of the lack of good potential members, but the inability to
nominate someone from a university without their
DVC/PVC/Research approval. This is difficult as
universities nominate their own people and, from the advice
given to me will not sign off on people they are not prepared
to nominate themselves. There is a significant time
commitment from any individual which requires this release
by any institution. I will be writing to the DVC at Sydney to
get further confirmation and to the ARC before we go
through this process again. We must be able to act as we see
fit, which may well conflict with the institution. Forewarned
is forearmed and I will move to clarify the role of
organizations such as ours.
Earlier this year we asked to comment of the duty of
care guidelines by the Assistant director o the Cultural
Heritage Coordination Unit. We have responded but have no
indication whether our recommendations have been
incorporated into the document.
The committee has worked proactively to maintain and
improve the membership base and Joe Dortch and Amy
Stevens have been actively pursing this over the year.
2004 AGM minutes
5.2 Secretary’s Report (Michael Slack)
Correspondence to AAA in 2004 consisted of 79 emails
and 3 letters. By far the greatest amount of correspondence
concerned employment inquiries (usually from overseas).
Seven children contacted AAA about the possibilities of
completing work experience. Most of these occurred directly
after National Archaeology Week.
Other correspondence ranged from membership inquiries,
subscriptions to Australian Archaeology, scientific advice, and
five questions regarding ‘alternative’ archaeologies.
5.3 Treasurer's Report (Joe Dortch)
The AAA financial year ending 31 August 2004 saw
continuation of most of the activities initiated by the previous
executive, with the result that income and expenditure
remained high. Overall, the association’s finances are healthy.
An increase in income means that while AAA paid for both the
2003 and 2004 journal printings in the one financial year and
so made a loss, it remains well-positioned for next year's costs
and is placed to make a reasonable profit (Table 1).
2003
$
30,033
16,626
13,407
55,154
68,211
68,211
TOTAL INCOME
TOTAL EXPENSES
OPERATING PROFIT (LOSS)
Retained Profits
PROFIT AVAILABLE FOR
APPROPRIATION
RETAINED PROFITS
2004
$
33,636
44,356
(10,721)
68,211
57,490
57,490
Table 1 Profit and Loss Summary 31 August 2004.
As usual, most income is from subscriptions (78%),
which continue to be in good shape thanks to the sending of
reminders to members of the previous year and
encouragement of new student members. We also had better
than usual conference profits (11%), thanks to conference
sponsorship in 2003. These are net profits here, after subsidies
for student and Indigenous delegate travel and some workshop
costs are included. Bank interest (7%) derives largely from our
two investment accounts, detailed below. Sales of back issues
remain at a similar level (3%) to previous years but with the
new DVD of past back issues now on sale this source of
income may reduce in future. Copyright fees (1%) are at a
similar level to most previous years.
The highest expenditure was in the two journal printings
paid for in one financial year (81%). This expense should be
halved in future years. Other expenses have been in initiatives
such as National Archaeology Week (in this year we paid
design and production costs for posters, hence printing for
NAW posters, stationary, and postage (7%) are presented
separately from journal costs) and conference prizes (6%).
Auditor’s fees (3%) were high in 2004 but should be reduced
in future. Remaining expenditures represent internet fees (2%)
and the normal operating costs for our bank accounts (1%).
A summary of expected expenditure for 2005 follows. Having
now shifted to paying for the journal printing earlier in the
year, in future years we will revert to the normal situation of
paying for one journal printing per year (57%). Additional
printing, stationary, and postage costs will increase (10%).
There are new expenses in two areas, on public indemnity and
liability insurance (9%) and DVD archiving (9%), which we
paid for in the current financial year. Insurance will be an
annual and necessary cost for future years. The policies were
recommended by the same broker who recommended
insurance policies for AACAI. Conference prizes expenditures
will increase (8%). Sales from the DVD journal archive should
begin to repay the one-off costs of moving the entire journal to
PDF format, and may return a small profit in future years.
Future auditing costs (3%) may be reduced with the
introduction of accounting procedures recommended by the
2004 auditor. Moving to electronic banking (2%) has reduced
the costs of card transactions and it is hoped to fully automate
receipting, which will keep postage costs low (and workload).
Internet fees (2%) should remain similar to last year. With the
starting base of our current account balance plus an expected
income from 2005 subscriptions, future expenditure is amply
provided for.
The last statements for these accounts show a combined
balance of $39,080, more than the costs of two year's journal
printings, which is the intended purpose of this reserve (Tables
2, 3).
$
EQUITY
Reserves
68561 Retained profits
Represented by:
1 Sept 2003 CURRENT ASSETS
CBA Business Account
31,875
20,870 CBA Term Deposit
15,816 CBA Cash Management Trust
68,561 NET ASSETS
$
57,490
1 Sept 2004
18,638
22,345
16,507
57,490
Table 2 Balance Sheet 31 August 2004.
Statement Particulars
Date
3/12/2004 Term
deposit
Amount Term Maturity Interest
Paid
Invested Date
$22,384 180 days 18/12/2004 4.9%
on
maturity
1/10/2004 Commonwealth $16,696 Ongoing N/A
4.56% pa
Cash Management
Trust
Table 3 Investment Register.
Variations in income from 2003 to 2004 included higher
total subscriptions and net conference profits, which include
the costs of travel subsidies and workshops (Table 4).
2003
396
1936
1226
517
23,258
2700
30,033
INCOME
Copyright fees
Net conference profits
Interest received
Sales
Subscriptions
Workshop income
TOTAL INCOME
2004
199
4184
2264
1054
25,935
33,636
Figure 4 Profit and Loss - Income.
87
2004 AGM minutesLessons for the profession
Variations in expenditure from 2003 to 2004 included the
cost of two years’ journal production and reduced expenditure
on the annual conference, workshops, telephone, postage, and
bank fees (Table 5). Variations in internet and database fees are
due to some of these bills having been paid for two years in
advance. The adjustment of last year's audit is due to an
accounting error by the 2003 auditor.
Overall this has been another successful year for AAA, for
which thanks are due to the previous executive for their
excellent management of the association, and to the 2003
conference committee for a very successful conference.
2003
1320
1200
625
5900
500
145
687
26
0
3485
3365
453
16,626
13,407
EXPENDITURE
Auditor's remuneration - fees
Bank charges
Bankcard charges
Conference expenses
Database maintenance
General expenses
Internet fees
Licences, registrations, permits
Journal production costs
Printing, stationery and postage
Prizes/Medallions
Telephone
TOTAL EXPENDITURE
OPERATING PROFIT (LOSS )
55,154
RETAINED PROFITS 68,561
1 September 2003
Less: Adjustment for error in prior year 350
1 September 2003
1 September 2004
68,561
2004
1408
418
196
0
548
154
345
54
34,870
3813
2550
0
44,356
(10,721)
Table 5 Profit and Loss - Expenses.
5.4 Membership Secretary’s Report (Amy Stevens)
A the time of the AGM membership was 560, just short
of the previous year which was 565, however the AGM was
earlier this year. All categories except for ordinary members
had witnessed some small growth. Reminder forms again
proved by far the best way to encourage membership.
There were 110 first time new members of AAA in 2004
(Table 6), all of whom received a bonus back issue of AA, an
initiative continued from last year. Membership forms were
sent out to every archaeology department in Australia in an
attempt to encourage student members, and lecture ‘bashing’
was conducted at a number of Universities.
Membership by State was relatively steady, however
membership in the ACT and Victoria has decreased
significantly, and might need some attention (Table 7).
Year
2004
2003
2002
2001
2000
Members
560
568
497
367
363
New Members
110
126
83
67
38
Table 6 Number of members and new members for the
last five years.
88
State
NSW
QLD
ACT
VIC
WA
SA
NT
TAS
2004
154
104
66
66
47
31
20
9
2003
148
111
81
81
44
39
17
7
2002
127
124
55
51
41
37
15
11
Table 7 Membership numbers by State for the last three
years.
5.5 Webmaster’s Report (Luke Kirkwood)
The exponential success of the website in 2003 continued
unabated in 2004. By the end of the year, traffic to the website
had effectively doubled, with over 9 Gigabytes of data served.
The website had also received over 400,000 hits by the end of
the year at a rate of roughly 7,000 unique visits per month.
November was once again the most popular month of 2004
with 52,811 hits. This was most likely due to interest generated
by the conference.
2004 also saw the debut of the new and improved AAA
website. Following the successful migration and
implementation of the AAA membership database in 2003,
work began on redesigning the website to take advantage of
this new system. Four main objectives were identified and
directed the development of the new website. Firstly the
website must have a management system that does not rely on
competency in coding webpages, secondly it must be a system
that requires minimal training, thirdly it must have a minimum
operating lifetime of at least five years, and finally the website
shouldn’t be completely orange.
The new website is the result of 6 months of development
and testing and has met all of the objectives outlined. To
facilitate management of the website, a backend interface has
been designed which can be utilized through a web browser by
the administrator. Information can be updated across the entire
website through the use of simple web-based forms. The
simplicity of this approach allows the website to be managed
by new executives without the need to have an expert to update
basic content. This gives the Webmaster the ability to
concentrate on upgrading and adding new features that will
benefit AAA membership.
The technology involved in the new website is a
combination of PHP scripting language (http://www.php.net/)
and MySQL databases (http://www.mysql.com/), both of
which are open source software packages. Because these
software packages are well supported among web designers,
this will ease migration of the website to newer technology at
a later date. Lastly the website isn’t completely orange.
It was also with mixed feelings that I announced at the
2004 Annual General Meeting that 2005 will be my last year as
principle webmaster for the AAA website. When I was first
invited by Sean Ulm to join his executive team for 2002, I did
not expect to finish up with a website that attracts 50,000 hits
per month. My time as webmaster for AAA has been nothing
short of a dream job for me, marrying my two loves:
archaeology and computers. Being awarded a Life Membership
Award at the 2004 conference for my contributions to the AAA
website was the icing on an already delicious cake. I would like
to thank AAA and our association’s members for giving me the
opportunity to create a resource that will hopefully be an
2004 AGM minutes
indispensable tool for all Australian archaeologists and help to
promote the advancement of Archaeology.
5.6 Editors’ Report (Donald Pate)
To mark the 30th year of the publication of Australian
Archaeology, the AAA Electronic Archiving subcommittee
converted back issues Volumes 1-57 (1974-2003) to a digital
format. The resulting archive is now available on a single DVD
disc for AUD$50.00 including postage. Thanks to Michael
Haslam, Sean Ulm, and Luke Kirkwood for their efforts in
relation to this project. An order form for the DVD can be
found on page 74 of AA 59 (December 2004) or on the AAA
website at: www.australianarchaeologicalassociation.com.au/
australianarchaeology
AA journal subscriptions were maintained at record levels
of over 550 subscribers during 2004. The editors continued to
offer a complimentary copy of the Special 20th Year Volume
AA 39 (1994) with each new membership. Sales of back issues
of the journal continued to be strong in 2004 with most stocks
of older issues being depleted. An order form for remaining
back issues dating from 1996 can be found on the AAA
website.
The editors launched an AAA photo competition in 2004
to stimulate interest in archaeological photography in relation
to the submission of high quality images to the journal. Photos
in three categories were judged at the AAA conference in
Armidale. These categories included: Archaeological Site
Images (Winner: Matt Schlitz, Flinders University),
Archaeological Fieldwork or Lab Work in Progress (Winner:
Matt Schlitz, Flinders University), and Artefact Images
(Winner: Kerrie Grant, University of New England). Thanks to
all who submitted photos for the competition and to the judges
Judith Field and Mike Roach.
The December 2005 issue of AA will be a Special Volume
addressing Teaching and Learning in Australian Archaeology.
This volume will be edited by Sarah Colley (University of
Sydney), Sean Ulm (University of Queensland), and Donald
Pate (Flinders University).
The editors thank all members of the AA editorial team
and journal contributors and referees for their contributions
during the 2004 year. Finally, a special thanks to Ian Murray at
Graphic Print for his ongoing assistance with journal
production.
5.7 State Representative Reports
No State representative reports were received for 2004.
5.8 AA Electronic Archive Subcommittee Report
(Michael Haslam)
Michael Haslam explained how to use the new AA DVD
electronic archive and how the user could search it by keyword,
title etc. Total costs for this project were $3614.04 and a total
of 30 copies at $50 each had already been sold during the
conference. He thanked Ian Lilley for allowing them to use his
old copies of AA for the archive. Peter White said that this was
a great achievement and an enormous amount of work had
been undertaken on this project by the Subcommittee. Judith
Field suggested that due to time constraints of the AGM a full
demonstration of the features of the archive should be
convened for later during the AAA conference.
5.9 Media Liaison Officer Report (Johan Kamminga)
Early in 2004 a AAA media release was posted about
National Archaeology Week events around Australia. Preparing
this release involved considerable correspondence with
members of the NAW committee and the AAA webmaster,
Luke Kirkwood. Jo acknowledges with thanks Luke
Kirkwood's assistance and his considerable effort in revising
the media web page. While a number of research findings by
AAA members and related matters were canvassed in the
media, no other significant issues were referred to the media
liaison officer by the AAA executive or members during the
months leading up to the conference. Jo dealt with a number of
enquiries from the public, and liaised with media about the
theft and destruction of cultural heritage in Iraq. Media
releases concerning issues to be raised at the annual conference
are in the process of being posted on the web site and some
releases will be sent to specific media agencies as well as being
posted on the AAA web site.
Jo has also been updating the list of media contacts a
process that will be ongoing in 2005. Media interest in the
2004 annual conference mostly concerns the recent
archaeological discoveries in Flores and there have been a
number of enquiries in the last few weeks. A number of
journalists and other media representatives are attending the
conference and the organisers at UNE have provided dedicated
facilities for them. The AAA executive is currently considering
whether there is scope for the liaison officer to provide media
contacts for independent releases about research results by
AAA members.
5.10 Archaeology Teaching and Learning Subcommittee
Report (Sarah Colley)
Apologies were received by Richard Mackay, Claire
Smith, Alistair Paterson, and Martin Gibbs.
The T&L Sub-Committee acknowledged the continuing
support of AAA during 2004. The committee was re-named
the Joint Interim Standing Committee for Archaeology
Teaching and Learning (JISCATL) at the 2003 National
Archaeology Teaching and Learning Workshop to reflect its
aim to incorporate all areas of Australian archaeological
practice and the support of ASHA, AACAI and AIMA, as well
as AAA.
Committee business has been conducted by e-mail
through 2004, with one face to face meeting at the UNE
Conference.
The main focus of the committee’s activities has been the
‘Learning Archaeology’ Session held at the AAA Conference
at UNE. The December 2005 issue of the journal Australian
Archaeology (Volume 61) will be a special edition devoted to
papers from the session, plus additional invited contributions.
This will be edited by Sarah Colley and Sean Ulm in
collaboration with Donald Pate.
Current initiatives linked to the Committee’s aims (as
reflected in the RAT Charter developed at the National
Archaeology T&L Workshop) included the following:
1) Sean Ulm has started a project to profile the profession in
Australia
2) Sarah Colley, with input from Martin Gibbs, Sean Ulm,
Denis Gojak and others has started work towards defining
a list of core skills for archaeology graduates which could
be linked to a nationally recognized course in professional
practice.
3) Sarah Colley attended a session on Archaeological
Pedagogies organized by Thomas Dowson at the December
2003 TAG Conference, University of Lampeter UK and
presented a paper on issues in T&L/ professional training in
Australia and the work of the Committee. This was
89
2004 AGM minutesLessons for the profession
published in World Archaeology in 2004. Sarah Colley had
also publicized the work of the Committee and the
forthcoming issue of AA at the joint AIMA/ASHA
Conference in September 2004, and at the Sydney
Historical Archaeology Workshop in November 2004.
4) Wendy Beck, with Jane Balme, has conducted research
into benchmarking Honours programmes in Archaeology
in Australian universities.
Sarah Colley reported briefly on Committee activities for
2004. Sean Ulm presented a draft proposal for an AAA Award
for Excellence in Teaching & Learning. He had agreed to
submit a motion to the AAA AGM seeking in principle
support, with further details to be decided in 2005 (e.g. terms
of the award (university and/or public education); criteria;
value of the award).
Discussion of e-mail correspondence previously
circulated by Sean Ulm from David Roe expressing concern
about changes to Commonwealth Funding Clusters for
Archaeology which reduces funding compared to e.g.
Anthropology, Social Sciences and Science subjects. Funding
has been a problem for some time and these changes make no
effective difference to financial support allocated to e.g.
Archaeology practical work courses in some universities as this
depends on structure of degree programmes and internal
funding re-allocation in each institution. It was agreed to
submit a motion to AAA AGM seeking AAA support for a
campaign to lobby the Commonwealth Government on behalf
of Australian university teaching staff in Archaeology, in
conjunction with the wider profession, to secure a higher level
of funding for Archaeology units of study which, with practical
work and professional practice component, are more expensive
to deliver than other Humanities subjects. It was proposed to
investigate seeking support from senior archaeologists in the
Australian Academy of the Humanities, given the very small
number of Chairs of Australian Archaeology (i.e. limited
lobbying ability).
A Proposal to investigate a more workable structure and
membership of the JISCATL was presented. A model
suggested was a smaller Steering Committee composed of
active members with interest/expertise to liaise with
representatives of e.g. AAA, ACCAI, ASHA, AIMA and a
wider committee with nominated representatives from each
Australian university which offers an Archaeology programme
of study, plus other departments which offer Archaeology units
of study within other degree programmes. Useful input from
Thomas Dowson on UK model and experiences had been
provided and it may be sensible to include Continuing
Education Programmes (or equivalent) also. Sarah Colley will
be overseas for 2005 and wishes to step down as Chair of
JISCATL. Agreed that Sarah will remain as Chair until March
2005, then Wendy Beck and Jane Balme will take over.
Comments were invited from the floor. Peter White said
that a teaching award was very questionable because many
areas in archaeology teaching such as classical archaeology
were not covered by the association, and that there needed to be
some sort of assessment procedure on how this might be
accomplished. Andre Rosenfeld endorsed the general idea but
asked how an award might be assessed and who by. Sean Ulm
suggested that areas such as the motivation and commitment to
students and commitment to supporting teaching were crucial
aspects of the award, and that the recipient need not be teaching
in a university. Bryce Barker expressed concern of the
assessment process. Pip Rath said that if Sean Ulm’s
suggestions were adopted that students could decide. Sarah
90
Colley stated that the Subcommittee would look at these issues,
and that the award need not be a cash prize. Robin Torrence
state that she thought more thought was required and that the
motions be rejected until there was a firm plan in place. Val
Attenbrow expressed support for Robin Torrence’s comments.
A first motion was put to the floor; that “AAA provide in
principle support for a AAA Award for Excellence in
Archaeology Teaching”. Motion: Sarah Colley. Seconded:
none.
Second motion was presented: that “AAA provide in
principle support to approach the Australian Academy of
Humanities and other bodies to develop an effective strategy to
lobby the Commonwealth Government to increase university
funding allocated to Archaeology units of study under current
formula models”. David Roe commented that this issue was
complicated as the government does not provide as much for
archaeology but that seeking advice was a good idea. Motion:
Sarah Colley. Seconded: Peter White. Motion carried nem.
con.
A third motion was presented: that “AAA make available
funds up to $2500 for the Teaching and Learning
Subcommittee for 2005”. Motion: Sean Ulm. Seconded: Val
Attenbrow. Motion carried nem. con.
6. Other Business
6.1 Annual Conference 2005-2008
Judith Field presented the 2005 AAA Annual conference
progress. An email had been received by co-organiser Al
Paterson which was read. The AAA conference is to a joint
conference with AIMA and will be held at the Western
Australian Maritime museum in Fremantle, WA, from the 27th
of November to the 30th of November 2005. The theme of the
conference will be “the Archaeology of Trade and Exchange”.
The web site for the conference should be active by April 2005.
Proposed speakers at the conference are to include John
Adams. Judith Field expressed some concern over the cost of
accommodation but stated that organizers were looking into
booking a backpacker venue for attendees.
6.2 Archiving of Archaeology Honours Theses
Sean Ulm stated that attempts had been made to pursue
this issue. AIATSIS had been willing to accept Honours thesis
of Indigenous students. No further progress had been made,
and it was decided that the AAA executive would follow up on
this subject in 2005.
6.3 National Archaeology Week May 2005
Michael Westaway spoke on this topic. The dates for
National Archaeology Week had been set to commence from
the third Sunday in May each year. He stated that there would
be around 150 activities planned for this years NAW and that
he was investigating various sources for funding. Public
lectures, displays and open days were planned.
Michael asked AAA for a motion “to make up to $2500
available for National Archaeology Week for 2005”. Motion:
Sean Ulm. Seconded: Pip Rath. Motion carried nem. con.
6.4 Ngarrabullgan Cave Conservation Issues
Bruno David presented a background to the
Ngarrabullgan site and concerns over future development. He
explained that the whole of the Ngarrabullgan mountain was a
highly significant area in terms of archaeology, Aboriginal
spiritual importance and unique faunal communities. He
2004 AGM minutes
discussed archaeological investigations which have been
recently completed and that Aboriginal occupation at Nonda
rockshelter had recently been dated at between 60,000 – 50 000
years BP.
Luke Godwin discussed the plans for coal mining to occur
at Ngarrabullgan. A permit to explore had been obtained and
that an access agreement was in place for work to occur. He
described the process of exploration which will consist of 20 –
30 test holes as preliminary holes, but that he expected this
number will expand greatly. If coal mining is found to be
economically viable high wall and long wall mining will occur.
Luke stated that such methods will ultimately result in the
collapse of surface ground by a couple of meters along any
mined areas. He discussed various avenues open to AAA to
attempt to stop mining at Ngarrabullgan, including National
Heritage listing, and appealing to the Aboriginal and Torres
Strait Islander Protection Act. Meetings were planned for
February 2005 of the Resource Tribunal, and that he had
approached Bruno David to provide expert evidence at this
meeting. A motion was then presented.
Motion 1: that “The Australian Archaeological
Association recognises the immense cultural significance of
Ngarrabullgan to the Jungan people. The association notes that
the archaeological sites that have been investigated on
Ngarrabullgan comprise one element of the cultural values
with which the mountain is imbued. The association notes that
archaeological sites investigated there constitute some of the
oldest evidence yet found for the occupation of the Australian
mainland by Aborigines. The association fully supports the
Jungan people in their determination to protect their cultural
inheritance and with that in mind stands totally opposed to
mineral exploration and any consequent development of
Ngarrabullgan.” Moved: Judith Field. Seconded: Michael
Strong. Motion carried nem. con.
Luke Godwin asked that a motion be presented to assist
with travel for Bruno David and others to attend the planned
meetings. Kevin Tibbett asked if there was a definite amount
asked for. Luke Godwin stated that it needed to be a serious
amount if AAA wanted to pay for airfares. Bruno David said
that he was not sure of the overall costs at this stage. Judith
Field proposed that $2000 be made available. Eleanor Crosby
expressed concern that this might ‘open the floodgates’ for
other situations. Peter White suggested that it would not so
long as each appeal/situation was assessed individually. Judith
Field agreed stating that consultation with AAA needed to
occur in all situations. John Appleton stated that he was
worried about using AAA general funds, and that it might be
wise to set aside money in a discretionary fund rather than use
the general pool. Sean Ulm stated that a “grant in aid” might
be made by the AAA executive.
Judith Field as the President of AAA, committed to
release such funds to members of the association to the
maximum of $2,000 to assist members with such travel for the
purpose of tending evidence to any court hearings into issues
associated with cultural values and the development of
Ngarrabullgan.
6.5 Other matters
Robin Torrence suggested that it was appropriate that the
President of AAA on behalf of the association write a letter of
support to Bruce Veitch. Moved: Robin Torrence. Seconded:
Sheryl Sparkes. Motion carried nem. con.
7. Election of Officers of the Committee
Judith Field read out a list of the current officers of the
committee. She stated that all were willing to run for another
year. No other nominations were received. The nomination of
the AAA executive was accepted unanimously.
No nominations were received for the positions of media
liaison officer, state representatives, or public officer. Existing
officers who were all absent were re-elected unanimously. The
position of state representative for Western Australia was
subsequently filled by Fiona Hook.
8. Close of Meeting
The President of AAA, Judith Field, called the 2005 AAA
AGM to a close at 7.45pm.
91
92

Australian Archaeology (Number 60) June 2005

Transcription

Similar documents

Paula Brennan, Greater Dandenong

Hukay - UP Diliman Journals Online - University of the Philippines

Life, Death and Warriorism in Mycenaean Greece

Invitation to seminar and workshop

Artefact - University of Southampton

PDF - Monex Europe

Dredged Up from the Past

From: Smith, Roland [mailto:[email protected]] Sent: 09

Animals Toy Horses Toy horses are common enough to deserve

Market update – Australia March 2014