thesis abstracts 83 obituaries

number 66 June 2008

EditorialSean Ulm & Annie Ross iiARTICLESThe Upihoi Find: Wrecked Wooden Bevaia (Lagatoi) Hulls of Epemeavo Village, Gulf Province, Papua New GuineaBruno David, Nick Araho, Alois Kuaso, Ian Moffat & Nigel Tapper 1Tigershark Rockshelter (Baidamau Mudh): Seascape and Settlement Reconfi gurations on the Sacred Islet of Pulu, Western Zenadh Kes (Torres Strait)Ian J. McNiven, Joe Crouch, Marshall Weisler, Noel Kemp, Lucía Clayton Martínez, John Stanisic, Meredith Orr, Liam Brady, Scott Hocknull & Walter Boles 15Dating of Bush Turkey Rockshelter 3 in the Calvert Ranges establishes Early Holocene Occupation of the Little Sandy Desert, Western AustraliaPeter Veth, Jo McDonald & Beth White 33Glen Thirsty: The History and Archaeology of a Desert WellM.A. Smith & J. Ross 45

SHORT REPORTSTrialing Geophysical Techniques in the Identifi cation of Open Indigenous Sites in Australia: A Case Study from Inland Northwest QueenslandIan Moffat, Lynley A. Wallis, Alice Beale & Darren Kynuna 60

BOOK REVIEWSRenewing Women’s Business: A Documentary by Julie Drew & Wardaman Aboriginal CorporationReviewed by Sally Babidge 64Coastal Themes: An Archaeology of the Southern Curtis Coast, Queensland by Sean UlmReviewed by Bryce Barker 65The Social Archaeology of Australian Indigenous Societies edited by Bruno David, Bryce Barker & Ian J. McNivenReviewed by Richard Cosgrove 67Neolithic by Susan Foster McCarterReviewed by Phillip C. Edwards 68Archaeology of Ancient Australia by Peter HiscockReviewed by Brian Fagan 69Artifact Classifi cation: A Conceptual and Methodological Approach by Dwight W. ReadReviewed by David Frankel 71Whalers and Free Men: Life on Tasmania’s Colonial Whaling Stations by Susan LawenceReviewed by Martin Gibbs 72Inauthentic Archaeologies: Public Uses and Abuses of the Past by Troy LovataReviewed by Denis Gojak 72Lithics in the Land of the Lightning Brothers: The Archaeology of Wardaman Country, Northern Territory by Chris Clarkson Reviewed by Richard A. Gould 74Box Offi ce Archaeology: Refi ning Hollywood’s Portrayals of the Past edited by Julie M. SchablitskyReviewed by Peter Hiscock 75Salvage Excavation of Six Sites Along Caddies, Seconds Ponds, Smalls and Cattai Creeks in the Rouse Hill Development Area, NSW by Jo McDonald Cultural Heritage Management Pty LtdReviewed by Fiona Hook 76Archaeological Theory and the Politics of Cultural Heritage by Laurajane SmithReviewed by Thomas F. King 77The Archaeologist’s Fieldwork Companion by Barbara Ann KipferReviewed by Ian J. McNiven 78Landscapes, Rock-Art and the Dreaming: An Archaeology of Preunderstanding by Bruno DavidReviewed by Tim Murray 79A Companion to Social Archaeology edited by Lynn Meskell & Robert W. Preucel Reviewed by Alistair Paterson 81

In this issue

nu

mb

er 66 2008

THESIS ABSTRACTS 83

OBITUARIESLester Richard Hiatt (1931–2008) 86

BACKFILLMinutes of the 2007 Annual General Meeting of the Australian Archaeological Association Inc. 87

2007 AAA Conference Awards 94

Letters to the Editors 98

AustArch1: A Database of 14C and Luminescence Ages from Archaeological Sites in the Australian Arid ZoneA.N. Williams, M.A. Smith, C.S.M Turney and M.L Cupper 99The Australian Academy of the Humanities: 2007 Fellows 99

Lectures 100

Conferences 100

NOTES TO CONTRIBUTORS 101

ISSN 0312-2417

AUSTRALIAN ARCHAEOLOGICAL ASSOCIATION INC.Offi ce Bearers for 2008

Position Name Address

Executive

President Ian McNiven School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Secretary Tim Denham School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Treasurer Bruno David School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Membership Secretary Jeremy Ash School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Assistant Membership

Secretary

Liam Brady Centre for Australian Indigenous Studies, Monash University, Clayton,

VIC 3800

Public Offi cer Sally Brockwell Centre for Archaeological Research, Research School of Pacifi c & Asian

Studies, Australian National University, Canberra, ACT 0200

Webmaster Samantha Bolton Archaeology, M405, University of Western Australia, 35 Stirling Highway,

Crawley, WA 6009

Media Liaison Offi cer Peter Veth National Centre for Indigenous Studies, Australian National University,

Canberra, ACT 0200

Media Liaison Offi cer Wayne Brennan PO Box 217, Katoomba, NSW 2780

Australian Archaeology Editors

Editor Sean Ulm Aboriginal & Torres Strait Islander Studies Unit, University of Queensland,

Brisbane, QLD 4072

Editor Annie Ross School of Social Science, University of Queensland, Brisbane, QLD 4072

Editorial Assistant Linda Terry School of Social Science, University of Queensland, Brisbane, QLD 4072

Short Reports Editor Catherine Westcott Department of Main Roads, GPO Box 1412, Brisbane, QLD 4001

Short Reports Editor Lara Lamb School of Humanities & Communications, University of Southern

Queensland, Toowoomba, QLD 4350

Review Editor Chris Clarkson School of Social Science, University of Queensland, Brisbane, QLD 4072

Review Editor Jill Reid Department of Main Roads, GPO Box 1412, Brisbane, QLD 4001

Thesis Abstract Editor Stephen Nichols School of Social Science, University of Queensland, Brisbane QLD 4072

State Representatives

Australian Capital Territory Kyle Turner 22 Donaldson Street, Braddon, ACT 2620

New South Wales Jodie Benton OzArk Environmental & Heritage Management Pty Ltd, PO Box 2069,

Dubbo, NSW 2830

Northern Territory Daryl Guse School of Archaeology & Anthropology, Australian National University,

Canberra, ACT 0200

Queensland Lara Lamb School of Humanities & Communications, University of Southern


South Australia Lynley Wallis Department of Archaeology, Flinders University, GPO Box 2100, Adelaide,

SA 5001

Tasmania Denise Gaughwin Forest Practices Authority, PO Box 180, Kings Meadows, TAS 7249

Victoria Nicola Stern Department of Archaeology, La Trobe University, Bundoora, VIC 3086

Western Australia Stuart Rapley Archae-aus Pty Ltd, PO Box 177, South Fremantle, WA 6162

Australian Archaeology, the offi cial publication of the

Australian Archaeological Association Inc., is a refereed

journal published since 1974. It accepts original articles

in all fi elds of archaeology and other subjects relevant to

archaeological research and practice in Australia and nearby

areas. Contributions are accepted in six sections: Articles

(5000-8000 words), Short Reports (1000-3000), Obituaries

(500-2000), Thesis Abstracts (200-500), Book Reviews

(500-2000) and Backfi ll (which includes letters, conference

details, announcements and other material of interest to

members). Australian Archaeology is published twice a year,

in June and December.

Subscriptions are available to individuals through

membership of the Australian Archaeological Association

Inc. or to organisations through institutional subscription.

Subscription application/renewal forms are available at

http://www.australianarchaeologicalassociation.com.au.

Graphic Design: Lovehate Design

Printing: Screen Offset Printing

Cover: Excavations at Glen Thirsty 1, August 2004, looking

southeast (L to R: Mike Smith and Sharon Overend)

(Photograph: M.A. Smith).

All correspondence and submissions should be addressed to:

Australian Archaeology

PO Box 6088

St Lucia QLD 4067

AUSTRALIA

Email: [email protected]

URL: http://www.australianarchaeologicalassociation.com.au

The views expressed in this journal are not necessarily those of

the Australian Archaeological Association Inc. or the Editors.

© Australian Archaeological Association Inc., 2008

ISSN 0312-2417

Editors

Sean Ulm University of Queensland

Annie Ross University of Queensland

Editorial Advisory Board

Val Attenbrow Australian Museum

Margaret Conkey University of California, Berkeley

Bruno David Monash University

Andrew Fairbairn University of Queensland

Richard Fullagar Scarp Archaeology

Martin Gibbs University of Sydney

Luke Godwin Central Queensland Cultural Heritage

Management Pty Ltd

Simon Holdaway University of Auckland

Susan Lawrence La Trobe University

Jo McDonald JMcD CHM Pty Ltd

& Australian National University

Tim Murray La Trobe University

Jim O’Connell University of Utah

Fiona Petchey University of Waikato

Paul Rainbird University of Wales

Lynette Russell Monash University

Robin Torrence Australian Museum

Peter Veth Australian National University

Editorial Assistant

Linda Terry University of Queensland

Short Report Editors

Lara Lamb University of Southern Queensland

Catherine Westcott Department of Main Roads (Qld)

Book Review Editors

Chris Clarkson University of Queensland

Jill Reid Department of Main Roads (Qld)

Thesis Abstract EditorStephen Nichols University of Queensland

iNumber 66, June 2008

EditorialSean Ulm & Annie Ross ii

ARTICLESThe Upihoi Find: Wrecked Wooden Bevaia (Lagatoi) Hulls of Epemeavo Village, Gulf Province, Papua New GuineaBruno David, Nick Araho, Alois Kuaso, Ian Moffat & Nigel Tapper 1

Tigershark Rockshelter (Baidamau Mudh): Seascape and Settlement Reconfi gurations on the Sacred Islet of Pulu, Western Zenadh Kes (Torres Strait)Ian J. McNiven, Joe Crouch, Marshall Weisler, Noel Kemp, Lucía Clayton Martínez, John Stanisic, Meredith Orr, Liam Brady, Scott Hocknull & Walter Boles 15

Dating of Bush Turkey Rockshelter 3 in the Calvert Ranges establishes Early Holocene Occupation of the Little Sandy Desert, Western AustraliaPeter Veth, Jo McDonald & Beth White 33

Glen Thirsty: The History and Archaeology of a Desert WellM.A. Smith & J. Ross 45


BOOK REVIEWSRenewing Women’s Business: A Documentary by Julie Drew & Wardaman Aboriginal CorporationReviewed by Sally Babidge 64

Coastal Themes: An Archaeology of the Southern Curtis Coast, Queensland by Sean UlmReviewed by Bryce Barker 65

The Social Archaeology of Australian Indigenous Societies edited by Bruno David, Bryce Barker & Ian J. McNivenReviewed by Richard Cosgrove 67

Neolithic by Susan Foster McCarterReviewed by Phillip C. Edwards 68

Archaeology of Ancient Australia by Peter HiscockReviewed by Brian Fagan 69

Artifact Classifi cation: A Conceptual and Methodological Approach by Dwight W. ReadReviewed by David Frankel 71

Whalers and Free Men: Life on Tasmania’s Colonial Whaling Stations by Susan LawenceReviewed by Martin Gibbs 72

Inauthentic Archaeologies: Public Uses and Abuses of the Past by Troy LovataReviewed by Denis Gojak 72

Lithics in the Land of the Lightning Brothers: The Archaeology of Wardaman Country, Northern Territory by Chris Clarkson Reviewed by Richard A. Gould 74

Box Offi ce Archaeology: Refi ning Hollywood’s Portrayals of the Past edited by Julie M. SchablitskyReviewed by Peter Hiscock 75

Salvage Excavation of Six Sites Along Caddies, Seconds Ponds, Smalls and Cattai Creeks in the Rouse Hill Development Area, NSW by Jo McDonald Cultural Heritage Management Pty LtdReviewed by Fiona Hook 76

Archaeological Theory and the Politics of Cultural Heritage by Laurajane SmithReviewed by Thomas F. King 77

The Archaeologist’s Fieldwork Companion by Barbara Ann KipferReviewed by Ian J. McNiven 78

Landscapes, Rock-Art and the Dreaming: An Archaeology of Preunderstanding by Bruno DavidReviewed by Tim Murray 79

A Companion to Social Archaeology edited by Lynn Meskell & Robert W. Preucel Reviewed by Alistair Paterson 81

THESIS ABSTRACTS 83

OBITUARIESLester Richard Hiatt (1931–2008) 86BACKFILLMinutes of the 2007 Annual General Meeting of the Australian Archaeological Association Inc. 87



AustArch1: A Database of 14C and Luminescence Ages from Archaeological Sites in the Australian Arid ZoneA.N. Williams, M.A. Smith, C.S.M Turney and M.L Cupper 99

The Australian Academy of the Humanities: 2007 Fellows 99

Lectures 100

Conferences 100


TABLE OF CONTENTS

ii Number 66, June 2008

Welcome to the fi rst issue of AA for 2008. In this issue we have

four articles: two papers on coastal archaeology in northern

Australasia and two on the arid zone. Bruno David et al. document

the exciting discovery of a trading canoe from coastal Papua New

Guinea and place this fi nd in both ancient (archaeological) and

modern (anthropological) contexts. Continuing the theme of

recent change in the Torres Strait region, Ian McNiven et al. report

on excavations of a rockshelter in Western Torres Strait that

demonstrate marked variation in use during the late Holocene,

supporting wider arguments for restructuring of land-use in the

region in the recent past. Moving to the arid zone, Peter Veth et

al. analyse results from a rockshelter on the Canning Stock Route

that fi lls the early Holocene temporal gap in much arid zone

archaeology. Mike Smith and June Ross also examine an arid

zone rockshelter and demonstrate once again the importance of

the late Holocene in Australian arid zone archaeology. There is

only one short report and we are keen to see an increase in these

short comments in the journal.

Even before opening the cover of AA66 readers will notice

the expanded size of this volume. This issue is even bigger,

however, than it might fi rst appear – not only is it over 100 pages,

but a further six pages of detailed supplementary information

relating to the articles is available online on the AAA website at

www.australianarchaeologicalassociation.com.au

We have taken the initiative of placing supplementary material

online for several reasons. For some time we have been grappling

with the problem of publishing large datasets and large numbers

of tables and fi gures in AA. We recognise the value of publishing

such data and realise that few venues are willing to publish

such detailed information. Conventionally, such information is

published in monograph form in series such as Terra Australis

and British Archaeological Reports. In an ideal world all of this

material should be published in hard copy. However, this is just

not practical anymore for AA. It simply costs too much money

to make the journal much bigger than it is already.

Another concern with publishing only a few data-rich papers

within the page limits and budget available for each issue of the

journal is that this would impact on the diversity of content and,

therefore, the broad appeal of the journal.

These issues came to a head in preparing copy for the current

issue with several data-rich papers included. We saw no viable

alternative other than to place some of this material as online-only

supplementary information to be published on the AAA website

at the same time as the print issue is circulated. Supplementary

information relating to articles may be accepted for online

publication where such materials are directly relevant to the

arguments presented in a paper. Such additional information

may include, for example, detailed data tables where there is only

space for a summary table in the printed version. Supplementary

material will normally be restricted to tables and fi gures, but may

also include video or audio fi les. All supplementary information

will be peer-reviewed, at the same time as the normal refereeing

process for the paper.

Critically, we feel that a paper in its printed form has to retain

its own integrity, meaning that anything published as online

supplementary information is exactly that, supplementary data.

The hard copy version of papers published in AA must stand

alone and the conclusions able to be comprehensible on the basis

of the data printed with the paper itself.

Another new initiative launched with AA66 is a Referee

Mentoring Programme which aims to increase AA reviewer

expertise by allowing referees to view the other assessments

of papers they review. If reviewers choose to participate in the

programme, their reports are deidentifi ed and forwarded to the

other reviewers of the manuscript who elect to take part in the

programme, on a strictly confi dential basis. We feel that this

programme will be of particular benefi t to junior referees and

others wishing to build their effectiveness as a reviewer.

In another milestone for AA, the Australian Archaeological

Association Inc., as publisher of AA, has been awarded a ‘Green’

rating by the OAK (Open Access to Knowledge) List Project,

part of the broader OAK Law Project based at the Queensland

University of Technology, Brisbane. This rating is in recognition

of the journal’s policies of allowing authors to retain copyright in

their work and allowing them to self-archive both pre-print and

post-print versions of work published in AA. See more details at

www.oaklaw.qut.edu.au

In recognition of the increasingly strong fl ow of manuscripts

to AA, particularly in the areas of historical archaeology and

cultural heritage management, we are pleased to welcome Susan

Lawrence and Luke Godwin to the Editorial Advisory Board.

We extend our congratulations to Sue O’Connor (Australian

National University) who was elected as a Fellow of the

Australian Academy of the Humanities and Ian McNiven

(Monash University), Sue O’Connor (again!), Rainer Grun

(Australian National University), Mike Smith (National Museum

of Australia) and Gocha Tsetskhladze (University of Melbourne)

who have been elected to The Society of Antiquaries of London.

Congratulations also to Meg Conkey, Editorial Advisory Board

member, who was recently elected as Society for American

Archaeology President for 2009.

We note with sadness the recent passing of Les Hiatt and

extend our condolences to his family and friends. Aboriginal

and Torres Strait Islander readers are warned that a photograph

of Les and the late Frank Gurrmanamana accompanies Les’

obituary as well as a photograph of the late Syd Coulthard in

Mike Smith and June Ross’ article in this issue. The photograph

of Frank Gurrmanamana is reproduced with the permission of

his daughter Betty Ngurrabangurraba.

In bringing AA66 to print we thank contributors, referees,

Editorial Committee, Editorial Advisory Board and John

Reid (Lovehate Design). Clair Harris, Michelle Langley and

Jo Bowman provided support in the Editorial Offi ces. The

Aboriginal and Torres Strait Islander Studies Unit and the School

of Social Science at the University of Queensland provided funds

to employ Linda Terry as an Editorial Assistant.

Sean Ulm and Annie Ross

EDITORIAL

1Number 66, June 2008

THE UPIHOI FIND: Wrecked Wooden Bevaia (Lagatoi) Hulls of Epemeavo Village, Gulf Province, Papua New GuineaBruno David1, Nick Araho2, Alois Kuaso2, Ian Moffat3,4 and Nigel Tapper5

1 Programme for Australian Indigenous Archaeology, School of Geography and Environmental Science, Monash University, Clayton, VIC 3800, Australia

2 Papua New Guinea National Museum and Art Gallery, PO Box 5560, Boroko, NCD, Papua New Guinea

3 Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia

4 Department of Archaeology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia

5 Atmospheric Science Programme, School of Geography and Environmental Science, Monash University, Clayton, VIC 3800, Australia

Lagatoi dug up on Gulf beach

Two lagatoi hulls, believed to be from the popular Hiri

Moale trade between the coastal villages of Central and Gulf

provinces, have been dug up along the coast of Kerema, Gulf

Province. The canoe hulls, believed to be the oldest and used

during the Hiri trade, were discovered by villagers of Keakea

and Epemeamo. The villagers dug out the canoes from the

sandy beach, buried more than two metres under ground. The

hulls, forming a lagatoi, were located about 20 metres from

the sea and still in good shape. Also discovered in the canoes

were claypot pieces believed to be from the popular Hiri trade.

National Broadcasting Commission Kerema manager Timothy

Akia said the villagers discovered the lagatoi canoe hulls in

July and managed to dig them up early this month. Mr Akia

said the lagatoi was very old and elders said it was one of those

used during the Hiri Moale trade many years ago.

They said the lagatoi was ‘one of those two that didn’t make

it back after bad weather’ from its Hiri Moale trade trip back

from Central Province. ‘Locals believe that while the lagatoi

found was one that didn’t make it back, the other one arrived

safely but this one didn’t return home after bad weather

caught up with it out at sea,’ Mr Akia said. The canoe hulls,

with all the features of a lagatoi, measured 30 metres in length

(Anon. 2007).

AbstractOn 20 August 2007, Epemeavo and Kea Kea villagers from

the eastern end of the Gulf Province of Papua New Guinea

reported fi nding two lagatoi hulls deeply buried in beach

sands at Upihoi, near Epemeavo village, parts of a trading

vessel associated with the renowned Motu hiri trade of

former times. This paper presents results of an emergency

investigation of these fi nds by staff of the Papua New Guinea

National Museum and Art Gallery and Monash University,

describing the fi nd, its environmental, cultural and social

settings and contexts of discovery, radiocarbon dating,

historical assessments, and signifi cance.

IntroductionOn 20 August 2007, six days after Timothy Akia of Radio Gulf

announced the news on the local radio station, the Post-Courier

– one of Papua New Guinea’s leading newspapers – reported

the fi nding of two wooden lagatoi hulls by Epemeavo and Kea

Kea villagers, between the Vailala River and Kerema in the Gulf

Province of Papua New Guinea (PNG). These were fi ndings of

no small concern to the PNG National Museum and Art Gallery

that administers the National Cultural Property (Preservation)

Act 1965 (No.26 of 1965), as amended by the Act of 1967 (No.65

of 1967) (later Chapter No.156 of the laws of the Independent

State of Papua New Guinea). This Act aims to protect cultural

property in PNG. No formerly active, historical lagatoi ships are

known to exist. The lagatoi and the hiri trade with which they

are associated are iconic traditions of southern PNG culture and

continue to be commemorated today in the Hiri Moale cultural

festivals associated with the annual 16 September Independence

Day national celebrations.

The hiri is an ethnographically-reported trade system involving

Austronesian-speaking (Motu) ceramic pot manufacturers and

traders journeying annually to coastal villages of the Gulf of

Papua in fl eets of lagatoi sailing ships. Formal trade transactions

were made between long-standing and often hereditary trade

partners. These villages then served as redistribution centres

for inland villages of nearby river systems (e.g. Chester 1878:9;

Dutton 1982; Herbert 1917; Oram 1977, 1982). The hiri trade

journeys are well-documented in late nineteenth and early

twentieth century literature (e.g. Barton 1910; Chalmers 1895;

Chester 1878; see Oram 1982 for a review). Trade voyagers set off

in fl eets of double- or multihulled lagatoi from Central Province

pottery-manufacturing villages when the southeast trade winds

blew, typically in October or November, and returned with the

monsoons around January in poorly manoeuvrable multihulled

vessels (see Chalmers 1895:74-92 for a fi rst-hand account of a

late nineteenth century hiri expedition) (Figure 1). These trading

expeditions brought ceramic pots and shell artefacts to Gulf

Province villages in exchange for sago and hulls. So large were

these expeditions that Fort (1886:15) wrote in his government

report in 1886 that annually ‘20,000 pots were taken, for which

they would bring back in exchange about 150 tons of sago’; other

estimates indicate up to 30,000 pots and 500 tons of sago (e.g. see

Allen 1976, 1977 for reviews). Sailing fl eets of 20 lagatoi were not

unusual (e.g. Barton 1910) (Figure 2).

Archaeological research since the late 1960s indicates that

the ethnographically-recognisable hiri system and its associated

ceramic traditions probably began around 500 years ago. Older

ceramic traditions across the Gulf and Central Provinces also

suggest that the historical hiri descended from a further 1500 years

or more of formalised long-distance maritime trade relations

across the region (e.g. Allen 1976, 1977; Bulmer 1978, 1982;

Rhoads 1982; see David in press for a review and signifi cantly

expanded radiocarbon dating results). At the other end of the

chronological spectrum, hiri expeditions were severely disrupted

during World War II when Motu villages were evacuated and

also as a result of increasing involvement in the wage economy

since the mid-1900s (Ryan 1970; see also Johnston 1974 cited

2 Number 66, June 2008

The Upihoi Find: Wrecked Wooden Bevaia (Lagatoi) Hulls of Epemeavo Village, Gulf Province, Papua New Guinea

in May and Tuckson 2000:59). Formal hiri trade expeditions

continued sporadically into the 1960s, although they largely

ceased in the late 1950s following the sinking of a lagatoi off

the coast of Boera village in 1957 (a then-predominant Motu

hiri pottery manufacturing centre), when several lives were lost.

However, long-held trading partnerships between villages have

in many cases been maintained, despite the demise of formal hiri

expeditions (Vincent Eka, Kerema resident and descendent of

historical hiri trade partner, pers. comm., 2007).

With these historical and cultural details in mind, the Post-

Courier report of the Upihoi lagatoi fi nd created considerable local,

national and international attention, including interest from local

villagers for whom the hiri voyages stirred memories of their own

recent pasts and signifi ed distant and recent social connections

and cultural practices; politicians who well understood the iconic

value and signifi cance of the hiri in local identity; and academics

who had long imagined the chance fi nding of sunken lagatoi and

the potential wealth of historical information that such a fi nd

could generate. However, given the apparent disturbance of the

site resulting from the extraction of the hulls from their buried

contexts, and the likely fragility of the wooden objects, the

reported fi nds also implied considerable urgency of assessment

and, most probably, management and conservation. Therefore,

the day following its newspaper reporting, Nick Araho, in his

capacity as Chief Curator of Prehistory at the PNG National

Museum and Art Gallery, asked Bruno David to participate in

urgent investigations of the newspaper claims to determine the

fi nds’ and the site’s cultural and scientifi c signifi cance, to protect

the site from further disturbance, to stabilise the fi nds from

potential damage and, if necessary, to plan a more extensive

follow-up fi eld season at the site and to initiate appropriate

conservation work on the fi nds. This paper reports on this initial

visit, and provides three radiocarbon determinations on the two

hulls and one of the masts.

The People of Epemeavo and Kea KeaThe Upihoi fi nd was made on the sand beach in the upper

intertidal zone immediately west of Epemeavo village, 20km

southeast of Orokolo village, 7km east of the Vailala River and

1.2km west of Kea Kea village (Figure 3). The place name ‘Upihoi’

is a compound word given by Epemeavo villagers when asked

about the name of the site where the lagatoi hulls were found. It

refers to Upi, the name of the general location, and hoi meaning

‘road’ in the Haula language – ‘road to Upi’ – in reference to

the beach as the road along which local villagers travel (see

also Williams 1940:16 for a description of the nearby Orokolo

Bay beach as a ‘highway’ in the 1930s). The Epemeavo and Kea

Kea villagers are close kin and identify as Keauru, speakers of

the Haula language. The Keauru are part of a much broader

cultural group historically-known as the Elema by seafaring

Motuans (see also Holmes 1903:125). The Epemeavo villagers

are all members of the Miaro clan, and show allegiance to fi ve

clan leaders and elders: Tom Haeae, Hoahe Merere, James Korea,

Kora Erope and Albert Moie. These fi ve clan leaders represent the

oldest male generation of the village, the eldest being born in the

1930s. According to John Erope of Epemeavo, in the 1890s the

Figure 1 ‘East End lakatoi at Purari Delta’, moored off Kaimare village around Christmas 1915 (Photograph: Ernest Sterne Usher, South Australian Museum, AA835 C93; see also Pike and Craig 1999:234, 248).


Bruno David, Nick Araho, Alois Kuaso, Ian Moffat and Nigel Tapper

ancestors of Epemeavo travelled south from inland locations to

establish the original Epemeavo village along the coast. Kea Kea

River which now fronts Epemeavo village in those days fl owed

further to the east than its present position. Epemeavo villagers

generally trace their recent history back to Kaveharo, an inland

village beyond a sago swamp near the mission village of Belepa,

5km inland and immediately north of Epemeavo (in 1928 John

Korea, John Erope’s grandfather, became a missionary at Belepa).

From Kaveharo, the ancestors moved to Helau near the present-

day small village of Poiva, and from Helau to Epemeavo. Sam

Nao points out that the old people used to say that the present

coastline, including today’s coastal villages eastward to ‘The

Bluff ’ (17km east of Kea Kea village), were in the past located

under the sea, the old coastline being located immediately south

of Belepa some 5km inland (see Rhoads 1994:53 for similar

claims of a sand ridge representing an ancient shoreline at Popo

2.5km inland, 22.5km to the northwest of Epemeavo). Based

on fi eldwork undertaken between 1923 and 1937, Williams

(1940:28) reports that:

the coast of Orokolo Bay has evidently been making ground in

recent times, and the population, who are so attached to beach

life, have moved forward with it. Indeed the sites of the present

villages were mostly, within living memory, under the sea.

Today, Kea Kea villagers commonly expose these buried ancient

shorelines during gardening and swamp drainage channelling

a few tens of centimetres below the ground surface. However,

the present Epemeavo village is the fourth village of that name,

the previous three being 100–200m south of the present village,

under the sea as a result of now-north-migrating shorelines and

a shifting Kea Kea River mouth.

Similarly, the Kea Kea villagers also trace their ancestry

to inland villages. Sam Nao of Kea Kea states that local oral

traditions indicate that the ancestors of today’s village came

from the inland village of Popo (for discussion of the ancestral

Popo site, see Williams 1940:28 for regional oral traditions and

Rhoads 1994 for archaeological investigations). According to

Sam Nao, in ancient times some Popo villagers went inland

travelling east to establish the ancestral village of Ahia

immediately to the west of the Vailala River; others went south

to establish Orokolo and nearby villages along the coast. Ahia

villagers then followed the Vailala River to the coast, where

some people then followed the coastline westward, and others

eastward to establish what is now Kea Kea village. Present-

day Kea Kea villagers explain that the ancestors divided from

Popo to follow different migration routes – some going inland

while others southwards directly to the coast. Today, numerous

languages are spoken across the region, including Haula by the

people of Kea Kea and Orokolo by those of Orokolo. According

to Kea Kea villagers, this is a consequence of the fi ssioned

migration history from Popo. Bastard (1922:70-79 cited in

Haddon 1937:210), writes that the ‘Keuru [Keauru] group of

villages, 7 miles east of Vailala ... has made its appearance on

the coast only within the past few years’. Holmes (1903:132)

similarly notes that:

The Haura [Keauru, who speak Haula language] tribe has only

made its appearance at the coast, with the intention of making a

claim on the coast line for future settlement, within the past few

years. The representatives of this tribe, now on the coast and in

its immediate vicinity, are all located between the east bank of

the Vailala river, and the spur of the Albert Range, known as the

‘Cupola,’ the east boundary of Kerema Bay. These communities

are known in every-day parlance, by the names of the respective

localities in which they live, as Helau, a village near Vailala, Keakea,

another village a few miles east of Vailala.

He further notes that ‘By occasional visits to the tribes living on

the east bank of this [Vailala] river, and a casual acquaintance

with their movements, [Holmes] has observed for many years

a marked tendency to migrate coastward’; and ‘From time

immemorial, there has been a tendency among these tribes

to migrate southwardly, wittingly or otherwise, toward the

coast’ (Holmes 1903:132, 133). Holmes makes no mention of

Epemeavo village in his 1903 paper, despite listing the coastal

village names for this region (but he does mention Helau, the

village from which the people of Epemeavo came according

to present-day oral traditions), implying its likely more recent

origins (see Holmes 1903 for details of origin stories).

In recounting ancestral migrations from inland locations, Kea

Kea villagers also noted that oral traditions identifi ed hilltops as

the locations of ancestral villages for reasons of defence in the

case of attack by hostile groups.

The Upihoi FindThe Upihoi fi nd was exposed after the collapse of large coconut

trees revealed the tops of the hulls in the sandy uppermost

reaches of the intertidal zone (Figures 4-5). Here the subcoastal

zone is of very fl at gradient and thickly vegetated by an expansive

coconut grove of great height. The shoreline edge of the grove is

marked by a 1.2m-high erosion face of thickly-matted coconut

palm roots marking the limits of high tide. South of this gully,

and below the base of the matted coconut palm roots, sediments

consist of the uniform dark mineral sands of the intertidal zone.

The hulls were found just below the base of this matted layer

of coconut roots, signalling that the lagatoi wreck predates the

growth of the coconut grove.

It is not clear exactly when in July 2007 the canoe hulls were

fi rst discovered by John Erope and Moisen Iavi of Epemeavo

village. However, in early August, shortly after discovery, a

group of 18 local villagers dug up and emptied the hulls of

Figure 2 ‘Fleet of lakatois starting for the west’ (Chalmers 1895:75).



their surrounding beach sand, and at high tide fl oated them

to the nearby village. John Erope said that one of the hulls was

buried in up to 2m of sediment. Initially only the front c.1m of

the most complete hull (see below) was visible, the stern along

with much of the hull resting in sand immediately beneath the

coconut roots. Resting sideways in a horizontal position, this

hull was dug out by the Epemeavo villagers after a coconut tree

fell down, creating a space for digging between the matted roots

around it. Digging by the 18 villagers began around 8am and

did not stop until about 5pm when the tide came in. The hull

was dug with two shovels and an old container, with seawater

being constantly thrown around the hull to make the digging

process easier. The sand was extracted by hand from inside

the hull. The coming of high tide facilitated removal of the

hull from its surrounding sands, and enabled it to be fl oated

to Epemeavo village where it is currently displayed on low

wooden rollers.

According to John Erope, the uppermost 50cm or so of

sediment above the complete hull consisted of a ‘black’ mineral

sand (the humic surface sediments of the coconut grove), below

which was found a ‘brown’ mineral sand (probably, but not

certainly, corresponding to the level of the matted coconut roots).

Below this and surrounding the canoe was a ‘white’ mineral sand,

very slightly lighter in colour than the ‘black’ and ‘brown’ sands

above. The villagers stopped digging when they reached the base

of the hull; the ‘white’ sand surrounding and fi lling the hull

continued below this level.

The second, broken hull was found completely exposed

in the middle of Kea Kea Creek near the mouth of the creek,

some 50m north-northeast (upstream) of the complete

hull. The hull was broken when a coconut tree fell on it.

Both the broken and the complete hulls were discovered at

approximately the same time. The former was easily retrieved

from the creek; Epemeavo villagers did not have to dig deeply

to move the hull as it was largely free of sand. A rope was tied

around the hull and it was pulled out of the creek and taken

to the village.

In total two large wooden hulls made of ihoea wood

(in Haula language; equivalent to iposea wood in Toaripi

language, and ipa-a in Yare language), one complete and one

fragmented mast made of mangrove wood, a wooden bung

also of mangrove wood, and a small undecorated ceramic

sherd were retrieved by the Epemeavo villagers. The masts,

wooden bung and ceramic sherd were found well inside the

more complete of the two hulls.

The Two HullsThe two recovered hulls are here referred to as ‘the complete

hull’ and ‘the broken hull’ (Figures 6-7). Both hulls are similarly

shaped, consisting of parallel-sided dugout logs of slightly wider

Figure 3 Map of the central southern coast of Papua New Guinea.

Figure 4 Sketch map of complete hull fi nd site, Upihoi (not to scale). Figure 5 The complete hull fi nd site at Upihoi. The hull was found in the clear area between coconut tree roots (Photograph: Bruno David).



beam towards the bow than the stern. Vincent Eka of Kerema

(pers. comm., 28 August 2007) has pointed out that in this region

the root end of a tree always forms the bow (or, in local parlance,

‘head’, as opposed to the ‘tail’ or stern) of a canoe or hull as it is the

thickest end which possesses the greater strength to push through

the waves. The bow of both hulls is fronted by a narrow projecting

beak-head at the top of the hulls, with the leading stem of the

bow sloping forward over a short distance. The stern of each hull

slopes backwards and does not possess decoration. Small, square

holes immediately below the gunwales (the rim of the hull) were

cut out to accommodate a series of transverse beams by which to

link the hulls of the vessel (Figure 8). These are found at regular,

parallel intervals on each side of each hull (as the gunwales of the

broken hull are extensively damaged, these are not always visible

on this hull). The two hulls are of comparable size. We describe

each separately below. Table 1 presents measurements of each

and compares them with published ethnographic canoe and hull

types from southern PNG.

The larger of the two hulls (the complete hull) has a maximum

length of 13.17m along the centre-line, and a beam of 80cm and

outer height of 80cm at midships (Figures 6-7). The sheer-line is

generally fl at rather than curved, with the fl at beak also horizontal

and only slightly raised. The cylindrical hull’s width between the

slightly tapering gunwales is 38cm at midships, and the interior

hull width also at midships is 70cm. The beak-head of the hull

projects from the bow by 45cm. A shallow, 5cm-deep platform

is inset in the bow. The walls of the hull are of fairly uniform

thickness, measuring c.5cm. The ends of the hull at the bow and

stern measure 52cm and 35cm maximum thickness respectively.

Each gunwale rim possesses 10 square holes (the sides of which

measures 5–7cm each; Figure 8). The holes are typically located

some 6cm below the gunwales. A single, 9cm-wide and 7cm-high

raised transverse stiffening member is found on the fl oor of the

hull about three-quarters along its length towards the stern. Such

features are typically used as re-enforcement in thin-walled

dugouts to strengthen the hull and to stop fl exing and bowing

of the canoe, as well as to stop the weight of the superstructure

from pulling apart the walls of the hull. The hull’s outer wall

surface is smooth and does not feature cut-marks; in general,

the interior wall surfaces are also devoid of clear cut-marks, but

the slanting interior surface of the bow and the square gunwale

holes contain clear marks made by metal blades (Figures 8-9). In

general the hull is complete, with only a small localised section

of the stern, and a small length of the gunwale astern of midships

being broken.

The broken hull is similarly shaped and of similar size to the

complete hull, albeit marginally smaller (Figures 6-7). Only the

front two-thirds of this hull is present, at a maximum length of

9.51m, and 70cm beam and 75cm height at midships. Like the

complete hull, the broken hull’s profi le tapers slightly from bow to

stern. The individual features of the broken hull are like those of

the complete hull, except that here the raised transverse stiffening

member is found closer to the bow (here it is 6cm-wide and 5cm-

high). Most of the gunwale has disintegrated, and it is therefore

not possible to determine exactly how many square holes to

accommodate the transverse beams were originally present; two

can still be seen, similarly spaced to those of the complete hull.

Like the complete hull also, the cylindrical broken hull has 5cm-

thick walls. The width of the opening between the gunwales is

39cm, while the maximum interior width of the hull is 60cm at

what would originally have approximated midships (based on the

complete hull’s length). Again, like the complete hull, the outer

surface of the broken hull is smooth, with metal cut-marks most

clearly visible on the sloping interior surface of the bow.

The Two MastsEach of the two masts consists of a single, linear but not

straight mangrove (poro wood in Haula) tree trunk with rough,

outer surfaces intact. One of the masts is complete, the other

fragmented. The complete mast is 4.68m-long and 9cm-wide at

its base. Numerous knots are evident along its length, although

these are fl ush with the surrounding surface of the wood.

The broken mast consists of four conjoining fragments, the

longest of which is 88cm-long and 9cm-wide. Only a proportion

of this mast was recovered during the extraction of the Upihoi

fi nd from its original setting by Epemeavo villagers.

The Mangrove BungA small piece of tapering mangrove wood identifi ed by John

Erope as a bung to plug leaks was recovered from inside the

Hull Max. Length

Outer Width at Midships

Interior Width at Midships

Width between

Gunwales

Circumference at Midships

Length of Beak

Distance of Holes Below Gunwales

(m) (cm) (cm) (cm) (m) (cm) (cm)Upihoi complete hull 13.17 80 70 38 2.80 45 c.6

Upihoi broken hull >9.51 70 60 39 2.35 67 c.7

Gar/gem, Torres Strait (Haddon 1937:193-98)*

9-19

Pe, Mawata & Fly River estuary (Haddon 1937:200-201, 203; Landtman 1927)*

14.68 96.2 81.28

Kauma of the Kaipi (Haddon 1937:211-212)*

12.2

Haruka of Orokolo (Haddon 1937:213)*

15.06 71.1 55.9 7-8

1886 Motuan lagatoi (after Barton 1910:96-120)*

14.53 80 4.75** 69

Table 1 Measurements of the Upihoi hulls compared to ethnographic hull dimensions, southern Papua New Guinea. * = Original imperial measurements converted to metric measurements. ** = Maximum dimension, rather than at midships.



Figure 7 Diagrams of the two Upihoi hulls.

Figure 9 Adze marks on the slanting interior of the complete Upihoi hull’s bow (Photograph: Nick Araho).

Figure 10 (A) The Upihoi bung. (B) The Upihoi ceramic sherd (Photograph: Steve Morton).

Figure 6 The two Upihoi hulls after they were taken to Epemeavo village, 26 August 2007 (complete hull in foreground; broken hull bow visible in background) (Photograph: Bruno David).

Figure 8 Two cut square holes from the gunwale rims on the complete Upihoi hull (Photograph: Bruno David).



complete hull during the extraction of the Upihoi fi nd (Figure

10A). It measures 6.3cm x 2.3cm x 1.9cm.

The Ceramic SherdAlso recovered from inside the complete hull during extraction

was a single ceramic body sherd (Figure 10B). It measures 3.6cm

x 2.8cm by a uniform 4.9–5.9mm thickness, and weighs 6.82g.

The sherd is not decorated and is extensively fi re-stained on its

exterior surface. Systematic surveys by BD and NA together with

Epemeavo villagers for c.50m around the Upihoi fi nd location on

26 August 2007 failed to fi nd any further cultural objects along

the beach.

Discussion of the Upihoi FindThe two hulls from Upihoi are of very similar size and

construction, as is typical of the hulls of multihulled vessels such

as lagatoi. However, there are other kinds of watercraft from

southern PNG that could potentially fi t the Upihoi fi nd. Here

we discuss a range of possibilities to determine whether or not

they indeed came from a lagatoi.

Among the large canoes and hulls of southern PNG, nine

potential candidates can be considered (we do not explore here

watercraft east of the Motuan villages of the Central Province,

nor single-hulled canoes without outriggers or platforms as

these can be dismissed due to their lack of holes along the

gunwale rims):

1. Gar or gem hulls of western and eastern Torres Strait

respectively.

2. Pe of Mawata.

3. Pe of the Fly River estuary.

4. Peri of the Bamu River estuary.

5. Large kauma of the Kaipi.

6. Haruka of Orokolo.

7. Bevaia of Orokolo Bay.

8. Toaripi olote.

9. Motu lagatoi of Central Province.

The details of each of these has been reviewed in Haddon

(1937:193-231) and Williams (1976:48-72), and will not therefore

be described beyond their salient features here. Summary details

are presented in Table 1 and discussed below.

We can discount Torres Strait canoes, the pe of Mawata and

of the Fly River estuary, and the peri of the Bamu River estuary

as accounting for the Upihoi fi nd because of differences in the

details of hull construction. As Haddon (1937:193) notes, rather

than possessing a fl at sheer-line with slightly raised fl at beak as

do the hulls of the Upihoi fi nd, here the hulls ‘gradually slope

up to a blunt point’ (Torres Strait gar/gem). Typically among the

gar/gem of Torres Strait, and the peri of the Bamu River, the stern

is cut square, whereas those of the Upihoi fi nd incline. The pe

of Mawata and of the Fly River estuary have ‘both ends raked

and produced to a blunt squared point’, and in the peri of the

Bamu River estuary the ‘bow has usually a more decided rake

[slope] than the stern and is more or less open in the front; the

stern is cut off squarely so that it looks like the cutting edge

of a gouge’ (Haddon 1937:205). In contrast, both ends of the

Upihoi hulls rake to a similar degree, each possessing a shallow

sunken platform towards the bow, fronted by a pronounced fl at

beak. The large kauma of the Kaipi are elaborately carved at the

bow and stern with various designs often of fi gurative forms,

unlike the undecorated Upihoi hulls (Haddon 1937:211-212;

see also Landtman 1927:210-211). In Torres Strait, Mawata, the

Fly River estuary, the Bamu River estuary, and among the Kaipi,

the dugout is accommodated with one or two outriggers and

superimposed or central platform, but these never extend along

the entire length of the hull, and the gunwales do not possess

square holes from bow to stern.

The above large canoes, mostly from the west of Orokolo Bay,

each possess design features that are signifi cantly different from

the Upihoi hulls, thus eliminating them as likely candidates. This

leaves in question the large ships of Orokolo Bay eastward to the

Motu lagatoi of the Central Province.

The peoples of Orokolo Bay-Vailala in the eastern part of

the Gulf Province have a number of large sailing ships with

hulls of comparable size to the Upihoi fi nds. One of these is the

haruka or haruka-iroki (Williams 1976:50), the ‘great double

canoes’ of Haddon (1937:213), which have all the characteristics

found on the Upihoi hulls except that no ethnographic details

are recorded of their bows (including beaks) and sterns. These

traditional watercraft are said to have been originally paddled

and sometimes used with temporary sails, giving way to the

bevaia sailing ships of the same region in the early twentieth

century (Haddon 1937:213; Williams 1976:50). Both the haruka-

iroki and bevaia were used ‘for long voyages on the deep sea and

for bringing in large quantities of food at feast times’ (Holmes

1903 cited by Haddon 1937:213).

According to Williams (1976:50-51) who in the early 1930s

witnessed the commencement of their use (said to have started in

the 1920s at the mouth of the Vailala River, and in 1930 elsewhere

in the region), the bevaia sailing ships of the Aivei River-Kerema

region are ‘imitation lakatoi’. Williams (1976:50-51) writes:

I do not know what the word signifi es [clan leaders at Epemeavo

village in 2007 translated the word to the generic ‘sailing ship’],

but it is the traditional name for the vessels which the Gulf

natives used to sight every year when the Motu, or as they were

called, the Derai-Haera [lagatoi], arrived about September. It is

often smaller: I saw none with more than three dugouts, whereas

the Motuan lakatoi often sail with four or fi ve; and in many cases

as yet it is not so well put together. The names of the parts differ,

the Gulf natives apparently retaining many of those that they

formerly used in the construction of their own haruka-iroki; but

the lakatoi pattern is followed very closely, even to the ornamental

fi nish of the cane lashings.

In his more detailed ethnography of Orokolo, Williams (1940:15-

16) writes:

The Orokolo natives, however, are not as a rule far-penetrating

bushmen, but people of the coast. Nor are they really at home in the

sea. They have their small outrigger canoes for shark-fi shing and in

these the lone paddler will venture some miles from the shore. But

of larger craft they possess very few. In the old days they would man

long double canoes [haruka-iroki] with paddlers and travel along

the coast to Motumotu or even beyond Cape Possession to Yule

Island, their object being to trade arrows and bamboo bows for the

shell ornaments which had found their way from the Melanesians



farther east. But they did not know the use of sails except in the

form of a mat temporarily rigged on a couple of poles. The long

expedition in which they now fairly often engage is a thing of

recent times. First they adopted oars, copied from the whaleboats

of traders and missionaries, and fi tted their double canoes with

fi xed rowlocks. Then – and this is acknowledged to be quite a recent

innovation – they began to copy the Motuan lakatoi which had for

generations before been bringing to their shores the cargoes of pots

to be exchanged for sago [hiri]. It is strange that this experiment

should have been so long delayed, but now there is a fairly well-

established series of expeditions running in the opposite direction

to those of the Motuans. The Gulf mariners sail with cargoes of

sago towards the end of the north-west season, and nowadays may

venture far beyond Port Moresby. Their object is to trade, not for

more pots, but for shell ornaments; and they return – if they have

got through – with a great fl ourish and blowing of shell-trumpets,

before the south-easterly trade winds have got up too strongly.

They copy, or attempt to copy, the Motuan lakatoi in the minutest

detail; but their inexperience is shown by the great proportion that

come to grief. Time and again the bevaia, as it is called, overloaded

and grossly over-manned with would-be travellers to the Papuan

metropolis, sinks almost before the journey has begun, and this

despite a great deal of magical preparation. So far the people of

Orokolo Bay are not so much mariners as enthusiastic learners. It

remains to be seen whether their enthusiasm will survive. During

my last visit [1937] it seemed rather to have dropped off.

Williams (1976:52) further notes that the ‘frequent failures’ of

their bevaia:

have been due in part to ill construction. Whereas the Motu have

had generations of practice, the Gulf natives are new to the game.

Not infrequently one may see the Motu visitors helping in or even

supervising the construction of a bevaia; but the independent

product may lack the fi nish in lashing and caulking which in such

a craft is essential to seaworthiness. A second cause of ill-success

is sheer overloading (see Figure 11; compare with Figure 1).

Williams (1976:53) lists the incidence of two bevaia wreckages

from Vailala between the end of 1931 and beginning of 1932, and

six from Orokolo for that same period of time (three ‘sank at

their moorings’; two ‘were driven ashore and broken’; one was

‘abandoned at sea on the very day of her sailing’).

The frequency of wreckage notwithstanding, the 1930s saw a

proliferation of attempts to build and sail bevaia in the Orokolo

Bay-Vailala region:

The large village of Vailala, situated at the mouth of the Vailala

River [7km to the west of Epemeavo], claims to have been making

bevaia for the last ten years or more, but I doubt whether they

have ever previously reached the pitch of enthusiasm which

has caused them in this last season (i.e., the end of 1931 and

the beginning of 1932) to send out twelve bevaia and one large

Figure 11 ‘Hareho’s bevaia waterlogged. Vailala, December 1934’. (Photograph: F.E. Williams, courtesy of the National Archives of Australia:A6003, 344.2).



haruka-iroki. Beyond Vailala, the fi rst experiments were made

only in the previous season, 1930-31, when three were equipped.

But during last season, while I was in the neighbourhood, three

were fi tted out at Orokolo and four at the neighbouring village

of Arihava (Williams 1976:51).

The famous Motuan lagatoi upon which the bevaia were based

have been well-described by late nineteenth and early-to-mid

twentieth century commentators. The classic and most detailed

account is by Barton (1910:96-120) which is paraphrased by

Haddon (1937:227) in the following way:

A lakatoi is composed of three or more asi, which are made of a

soft-wood tree (ilimo) of great size that grows close by rivers in

their low alluvial reaches in the Papuan Gulf district. The Gulf

natives fell the trees and fl oat them to the lakatoi that have arrived

on a trading expedition (hiri). The trunks are hauled on to the

bank of the river, where the visitors hollow them out and shape

them. Fire is not employed in this operation. An asi is a clumsy

dugout with rounded or squared ends prolonged above into a

projecting fl at beak ...

The asi are secured together by numerous cross beams which

are tied by lashings that pass through square holes cut in their

gunwales. Over these beams a large platform or deck (ilaha) is

constructed; this must be made very strong as it has to stand

the strain of the great waves of the Gulf. The platform extends

beyond the asi especially fore and aft. In 1884 the largest lakatoi

which arrived at Port Moresby from the Gulf consisted of 14 asi

and measured 59 by 51 feet [18.0m x 15.5m]; two smaller ones

measured 54 by 37 feet [16.5m x 11.3m].

Lagatoi typically have two masts ‘made of the stem of a

sapling of a species of mangrove’ (Pratt 1906:72; see also

Chalmers 1895:76), although single-masted and various

confi gurations of multimasted vessels are known (e.g. Haddon

1937:227-231).

From these details, and given the dimensions presented in

Table 1, it is not possible to distinguish the Upihoi hulls from

the haruka-iroki, bevaia and lagatoi, as insuffi cient structural

details have been published of the former to allow for this, and

the bevaia were modelled precisely on the lagatoi, at times with

direct Motuan advice and involvement during the construction

process. Similarly, Haddon (1937:216, following Chalmers

1898:327), Holmes (1903) and Abel (1902:72-79), note that the

large, multihulled olote sailing ships of the Toaripi (‘Motumotu’)

towards the eastern end of the Gulf Province are also ‘the same

as the Motu lakatoi’. On metrical and design characteristics

alone, the Upihoi fi nd could thus relate to any of these types.

However, we suggest below that other details allow a narrowing

of possibilities.

The Age of the FindsSix independent lines of evidence can be used to date the Upihoi

fi nd. The results of each support the other: radiocarbon dating;

cut-marks; geomorphic evidence of an old shoreline; stratigraphic

location a short distance below coconut roots; historical records;

and two hulls but lack of pottery.

Radiocarbon DatingWood samples from the outer parts of each of the two hulls plus

from a piece of the broken mangrove mast were submitted for

radiocarbon dating (Table 2). In most circumstances, the ‘old

wood problem’ (e.g. Schiffer 1986) would warrant great caution

in interpretation of radiocarbon results. In our case, however,

the ‘old wood problem’ is not signifi cant, for in each of the three

radiocarbon samples the only part dated was the outer layer of

the trunk. In the case of the complete hull (Wk-22225), a c.1mm-

thick layer of the outer wood was peeled from the base of the

hull (originally representing the outer layer of the tree). In the

case of the broken hull (Wk-22224), a 1cm-thick piece of wood

was taken from the outer surface of the fragmented stern end

of the hull. A 14.8cm-long section of the broken mast was also

submitted (Wk-22223); only the outer layer was extracted for

dating at the Waikato Radiocarbon Dating Laboratory. In each

case, the dating of the outer layer of wood only has meant a

potential old wood component of a few years maximum. The

radiocarbon results thus approximate the timing of death of the

trees themselves.

The three radiocarbon determinations each revealed

comparable results, in agreement with the assumed

contemporaneity of the fi nds; the near-identical 13C values for the

two hulls, both made of the same taxon of ihoea wood, further

testify to the reliability of results (Table 2). In each of the three

determinations, the highest probability 2 sigma result indicates

an age of somewhere between AD 1800 and 1960 – unfortunately

a period of time that exhibits considerable fl attening in the

radiocarbon calibration curve and that thus limits more precise

calibration (combining the three ages in this instance would not

resolve this problem of atmospheric radiocarbon uncertainty).

Object Haula Language Name of

Wood

Lab. No. δ13C‰ (±0.2)

% Modern (±0.4)

14C Age (years BP)

Calibrated Age AD (1σ) (probability)

Calibrated Age AD (2σ)

(probability)

Complete hull ihoea Wk-22225 -23.7 98.6 110±36 1700-1730 (8.7%) 1800-1930 (59.5%)

1680-1730 (16.1%)1800-1960 (79.3%)

Broken hull ihoea Wk-22224 -24.0 97.9 171±35 1670-1740 (24.1%)1790-1820 (6.3%)1830-1900 (24.4%)1920-1960 (13.4%)

1670-1960 (95.4%)

Broken mast poro Wk-22223 -27.1 98.5 125±35 1680-1740 (19.8%)1800-1890 (38.8%)1900-1930 (9.6%)

1670-1780 (36.7%)1790-1950 (58.7%)

Table 2 Conventional radiocarbon determinations, objects of the Upihoi fi nd. Radiocarbon ages were calibrated using OxCal 3.10 (Bronk Ramsay 1995, 2001) and the Southern Hemisphere calibration dataset (McCormac et al. 2004).



However, the absence of a bomb pulse signal indicates that these

samples all pre-date the period of atomic testing in the 1950s.

Cut-MarksNeither of the hulls exhibits any obvious cut-marks on its exterior

surface, but the slanting interior surfaces of both bows, and the

square holes below the gunwales have clear marks made by metal

adze and axe blades (Figures 8-9). To the immediate west of

Orokolo Bay, at Ukiravi village in the Purari River delta, Williams

(in Young and Clark 2001:68) captioned a photograph of a canoe

hull being shaped with a metal adze in 1922 in the following way:

‘The rough work of hollowing is done with a trade axe ... When

the canoe had been roughly hollowed, the fi nishing touches

are put to the interior with an adze’. Metal became available in

southern PNG well before European settlement in 1873, through

trade with passing ships and internal exchange networks (e.g.

Swadling 1996; see also Hughes 1977). However, metal axes and

adzes did not replace stone tools along the Gulf Province coastline

until the late 1800s into the early 1900s, when a proliferation of

new Western administrative centres, government patrols and

missionary stations saw the onset of sustained availability (e.g.

Bevan 1890; Chalmers 1895; see also Salisbury 1962; Young and

Clark 2001:68, 121, 192, cf. Young and Clark 2001:235, 240, 242,

245, 248, 263, 273). The Upihoi hulls, manufactured by metal axe

and adze blades, were thus most probably fashioned sometime

after the late 1800s.

Geomorphic Evidence of Old ShorelineThe coastline along which the Upihoi fi nd was made is today, as

in the recent past, geomorphologically active, as evidenced by

inland relic shorelines (some of which have been radiocarbon

dated) and oral traditions. In particular, past villages containing

rich ceramic sherd assemblages are evident along relic shorelines

located well inland (e.g. at Popo, Keveoki). At Popo, a radiocarbon

determination of 410±80 BP (ANU-2181) indicates the presence

of the shoreline 2.5km inland in the very recent past (Rhoads

1994:53-55). Oral traditions of the coastline shifting ‘within living

memory’, some discussed by Williams (1940:28) for the 1920s-

1930s and others presented by Kea Kea and Epemeavo villagers

in 2007 (see above), testify to the rapid southward migration of

the coast during the late nineteenth into the twentieth century.

In such actively aggrading coastal settings, the settling of a lagatoi

wreck on or very near the present shoreline indicates a relatively

recent event measurable in decades rather than centuries.

Stratigraphic Location a Short Distance Below Coconut RootsThe complete Upihoi hull was exposed following the collapse

of large coconut palms on the presently surviving edge of an

expansive coconut grove a short distance to the west of Epemeavo

village. The roots of the collapsed palm trees – then-level with

those of the presently standing trees nearby – lay just above the

complete hull. The implication of shallow sediment build-up

between the base of the coconut palm roots and the top of the

hull is that the latter came to rest in its buried position shortly

before establishment of the coconut grove. Lewis (1996:96)

notes that coconut plantations began to be established a few

kilometres away after 1908 by ‘Donald McDonald at Vaiviri

[c.10km northwest of Epemeavo]; Robert Mawson, father of the

Antarctic explorer, at Orokolo [c.20km northwest of Epemeavo]

... Mrs Ashton at Huiva on the Ie Inlet [c.6km east of Epemeavo]’,

and Maira on the Vailala River ‘above Vaiviri’, 6.5km north of

Epemeavo. In 1937, the Vaiviri plantation was the most extensive

at 350 acres. However, no coconut plantations are known near

Epemeavo, the palms adjacent to Upihoi being ‘natural’ and

locally planted growths.

We can estimate the likely age of the trees by reference to

their great height and plant ecology. Tall palms typically grow

for 60–80 years, reaching heights of 20–30m, although some

are known to grow longer and taller. With these general points

in mind, Epemeavo’s coconut trees near Upihoi appear to have

commenced growing in the 1940s or slightly earlier. Therefore,

we suggest that the Upihoi fi nd predates the 1940s given its

location below the coconut palm roots.

Historical RecordsHistorical records do not reveal direct evidence of the antiquity

of this particular fi nd. However, two points are pertinent. First,

it is clear from late nineteenth and early twentieth century

historical documents, ethnographic writings and present-day

oral traditions that all seafaring in this region was hazardous due

to the rough seas and considerable swell. Stories of hazardous

seas are regularly made in the historical literature – especially

when discussing bevaia and lagatoi expeditions – while those

of shipwrecks abound, particularly with respect to bevaia (e.g.

Chalmers 1895; Williams 1940, 1976). The bevaia of the 1930s

saw particularly high rates of wreckage as a result of poor

construction and insuffi cient handling skill (Williams 1976).

Based on these accounts, a large number of bevaia are known to

have foundered along the Orokolo to Kerema coast in the 1930s,

and therefore represent the most likely candidates for the Upihoi

fi nd, although in itself this does not exclude Motuan lagatoi

from consideration.

Two Hulls but Lack of PotteryThe Upihoi fi nd contains two hulls but a paucity of ceramics.

If it represents a west-bound, storm-wrecked lagatoi, we would

expect large quantities of ceramic sherds to have been found

by the Epemeavo villagers as they dug out the hulls, and also

evidence of broken ceramics in the sand below the coconut

palm roots. With the single exception of the small sherd found

in the complete hull, no other evidence of pottery was found

at the site. The implication is that the ship was probably not

laden with pots but rather with degradable materials such as is

the case with east-bound trading vessels carrying sago. However,

such latter ships tend to contain multiple hulls rather than two,

for on their arrival in recipient villages the lagatoi are refi tted

with extra hulls for the return journey (as suitable trees do not

grow in the Motuan homelands). We suggest that either more,

yet-undiscovered hulls are buried nearby, or, more probably, the

Upihoi fi nd represents an east-bound vessel with two or slightly

more hulls laden with sago for trade. In the latter case, a locally-

manufactured haruka-iroki or bevaia is implicated. However, the

presence of two masts in the Upihoi fi nd would suggest a bevaia

is more likely than a haruka-iroki. As bevaia began to be made in

the 1920s to early 1930s, such an interpretation would suggest a

vessel dating to sometime after the mid-1920s, most likely 1930s

during their peak production, use and wreckage. The presence



of metal cut- marks also supports the interpretation as a bevaia

rather than the earlier haruka-iroki.

Meteorological ConsiderationsThe Gulf region of Papua New Guinea lies within the region

of infl uence of the Asian-Australian monsoon characterised by

the seasonal alternation of trade wind easterlies with monsoon

westerlies. However the northwest monsoon fl ow across eastern

Papua New Guinea is rather weaker than it is further west over

the Banda and Arafura Seas (Tapper 2002). Despite this, the

coastline of the eastern Gulf is one of atmospheric convergence

under northwest fl ow conditions (McAlpine et al. 1983),

convergence that is exacerbated by a land breeze/katabatic fl ow

regime associated with the high country to the north and east of

Kerema. The eastern Gulf region is therefore one of four areas

of atmospheric convergence and storminess around Papua New

Guinea identifi ed to be associated with monsoon northwest fl ow

(McAlpine et al. 1983:48). Interaction of airfl ows in the eastern

Gulf is known to produce quite severe squalls (short-lived bursts

of high winds accompanied by thunder, lightning and intense

precipitation). The ‘Guba’ is a well-known monsoon season

squall that occurs in the Gulf area, especially around Kerema

(McAlpine et al. 1983). These squalls mainly occur in the early

hours of the day and are associated with northwest winds that

have been recorded up to 170km/h, with strong winds lasting

for up to 30 minutes. The squalls are frequently associated with

very heavy rainfall. It is likely that one of these squall events was

linked to the wreck of the canoe found at Epemeavo.

Discussion of AntiquitySix independent lines of evidence conclusively indicate that the

Upihoi fi nd dates to sometime between the nineteenth and mid-

twentieth century. The presence of metal cut marks, and the

highest probability radiocarbon calibrations narrow down the

fi nd’s most likely age to some time between the late 1800s (when

metal tools became readily available) and 1950. The location of

the hulls a short distance below the coconut roots suggests a pre-

1940s antiquity. The two most likely vessel types are bevaia and

lagatoi, of identical hull construction; the Toaripi olote can be

discounted as these east-bound vessels are found some 60km to

the east of Upihoi, and are not known to have ventured this far

westward. The haruka-iroki are possible candidates based on size,

except for the presence of two long masts, in this region most

typical of lagatoi and bevaia, and that the distinctive fl at beaks

of the lagatoi (and presumably lagatoi-derived bevaia) have not

been recorded from this region on other kinds of watercraft.

We are thus left with lagatoi and bevaia as the two most likely

candidates for the Upihoi fi nd.

The presence of two masts, and probably also the distinctive

beaks on the Upihoi hulls, thus point to a late-nineteenth to early-

twentieth century lagatoi or bevaia as the likely ship represented

by the Upihoi fi nd. The great frequency of shipwrecked bevaia

along the coastline between Orokolo and Kerema during the

1930s, precisely in the region of Upihoi, strongly suggests that

the Upihoi fi nd is one of the wrecked bevaia documented by

Williams (1976) for the 1930s. We note in this regard that present-

day Epemeavo and Kea Kea villagers, including the clan leaders

(who are also the elders, the eldest born in the 1930s), were during

our visit entirely unaware of the 1930s bevaia ‘experiments’ and

wreckages discussed by Williams (1976). Nor do they recognise

the Upihoi hull designs, indicating likely foreign infl uence or

manufacture (as applies to historical bevaia and lagatoi alike,

both of Motuan design; in contrast, the two-hulled haruka-iroki

are of local design).

Future WorkThe likelihood of further cultural materials associated with the

Upihoi fi nd remaining in situ is uncertain. However, given the

signifi cance of the site, its rarity, and the stratigraphically intact

nature of the fi nd spot’s immediate surroundings, geophysical

survey will be undertaken in November 2007 to investigate

the area surrounding the Upihoi fi nd location and a recently

discovered nearby archaeological village site to locate subsurface

items of material culture and provide further locations for

direct investigation.

DiscussionThe Upihoi fi nd represents the only known remains of once-

active historical bevaia or lagatoi (of identical design). Based on

morphological and metrical characteristics and historical details,

the hulls and associated paraphernalia are almost certainly from

a Gulf Province bevaia that aimed to copy the classic lagatoi, in

either case of Motuan design. Six independent lines of evidence

have been used to determine the fi nd’s antiquity, each in close

agreement with the other and together indicating that the ship

was most likely built, used and wrecked in the 1930s.

The Upihoi fi nd is an unprecedented archaeological discovery

of an important aspect of PNG’s cultural history. However, the

major signifi cance of this fi nd would be lost were it simply

couched in terms of its ‘archaeological’ or ‘scientifi c’ criteria.

Rather, to understand the value of the Upihoi hulls, masts, bung

and ceramic sherd requires a broader consideration of the

social contexts of trade across the Central and Gulf Provinces

and an understanding of Indigenous transcultural networks,

in relation to PNG’s colonial history. Such canoe hulls are

known from ethnography but have never previously been

found archaeologically. As such the Upihoi fi nd represents an

opportunity to further explore social and cultural contexts of

long-distance maritime (hiri) trade in southern PNG during

the early colonial period, including a consideration of social

relationships between trading partners, acculturation processes,

and traditional navigation.

Based on contemporary interviews and personal

observations, Williams (1976:52) suggested that the peoples

of Orokolo Bay-Kerema began building bevaia and sailing

them to the Motuan homelands – and in doing so mimicking

their own hiri expeditions in reverse – to access the new wage

market economies of the colonial centres, in particular Port

Moresby. Williams (1976:52) speculates that ‘There are two

fairly obvious reasons’ as to why the Elema embarked on such

a new entrepreneurial enterprise:

One is that of the tax. This may drive him afi eld to seek money,

and it should be noted that the purpose of the hahi expedition

[the eastward trading voyages originally undertaken on haruka-

iroki but subsequently those of the bevaia] is not so much to

collect pots (the lakatoi from the east bring enough of them) as to

seek shell ornaments, trade goods and money. The other reason



is found in the after effects of the Vailala Madness, that startling

religious movement which, temporarily in some parts, and (it

may be) permanently in others, involved the destruction of the

ceremonies. This reason applies especially to Vailala itself, which

remains strangely obdurate against a revival of the old customs,

though the surrounding villages have been gradually taking

them up again. A number of informants at Vailala spoke with

enthusiasm of the present state of affairs. Formerly they were

always busy in preparation for some ceremony or festival; now

they have ample time for enterprises of a more profi table nature

– at least commercially. As one man literally put it, it was their ‘one

thought’ nowadays to make money. Others did not lay so much

stress on the money as on the actual bevaia. Previously they were

peraia, ‘glad’ with hevehe, kovave, and so on (i.e. the really fi ne

mask ceremonies of the Gulf); now they were peraia with their

bevaia. And it is a truly surprising thing to see this thriving village

continually at work in shaping the enormous dugouts, in lashing

them together, in building the superstructures, and in rigging

the masts, all with meticulous attention to detail, or engaged in

the making of sago, collecting betel, or the manufacture of dyed

sago-leaf skirts for cargo. Nowadays, indeed, Vailala is more a

port of native trade than any Motu village. Its inhabitants have

discovered, unassisted, a substitute for the interest of the old

ceremonies: they are quite mad on bevaia.

The 1930s followed a period of destruction of traditional

ceremonial paraphernalia and a cessation of the ceremonies

themselves during that keenly followed socio-religious and

economic movement known as the ‘Vailala Madness’, when in

1919 the peoples of the broader Vailala region began to expect

the return of their deceased relatives and more distant ancestors

‘in a large steamer, which was to be loaded with cases of gifts

– tobacco, calico, knives, axes, food-stuffs, and the like’ (Williams

1976:341). ‘In some early versions of the belief the steamer was

to have on board a consignment of rifl es, which would be used

in driving the present white inhabitants of Papua, out of the

country ... it seems that vague ideas of Papua for the Papuans

were current at the time’ (Williams 1976:342). A core aspect of

the Vailala Madness was a conviction ‘that the old customs are

no good’ and ‘the condemnation of the old customs’ by local

community members, resulting in a destruction of previously

important ceremonies and ritual objects (see Williams 1976:331-

395 for detailed descriptions of the Vailala Madness). ‘The main

teaching’, writes Williams (1976:386), ‘was that the old customs

and ceremonies must be done away with ... There can be no

doubt that misunderstood Mission teaching had something to

do with this aspect of the Madness’.

The bevaia movement of Orokolo Bay-Kerema came in the

aftermath of the Vailala Madness, when soon after the destruction

of the ceremonial objects, and in the failure of the living to catch

the spirits of the dead which, they said, fl ashed past them, came

‘a defi nite move to reinstate some of the ceremonies’ (Williams

1976:385). Yet many of the masks and other paraphernalia

had been destroyed. The new bevaia ventures focused local

communities onto a promising social endeavour that both

required concerted communal efforts (the construction of the

bevaia and accumulation of associated trade goods) and brought

anticipated returns (Motuan objects of value, Western goods,

money) to the Gulf communities. In this sense these new socio-

commercial ventures can be understood only in relation to their

contemporary social contexts as historical momentum. The new

bevaia movement was a response to colonial period forces that

at once disrupted customary ceremonial and socio-economic

practices while offering the prospect of new socio-economic

opportunities. They are, in this sense, an example of transcultural

dynamics of the early-to-mid-colonial period in PNG, and an

assertion of active Indigenous voice and agency in the face of

missionary activity, colonial economics and customary socio-

cosmological practices. The Upihoi fi nd represents this Indigenous

response to colonial period disruption as a manifestation of

Indigenous agency, and is an important symbol of cultural

survival with change through a period of colonial rule into the

post-independence period. It is an instance of what Nakata

(2007) has called the ‘cultural interface’, necessitating locating

Indigenous innovation and practice within local aspirations and

local processes of social engagement. The Upihoi fi nd does more

than intimate cultural change and survival across a temporal

frontier that was the colonial period. Rather, it highlights also

the entrepreneurial and agenda-setting initiatives of local Gulf

communities together with their Motuan lagatoi craftsmen,

advisors in bevaia construction and trade partners – the latter

arguably PNG’s most marginalised group at the proximal end of

the colonial power base, Port Moresby – and for this reason its

symbolic value cannot be underestimated.

Objects call on us to think and behave in certain ways.

They materially connect the observer as an embodied being

to an engaged world; they allow biographies to be had: cars

give birth to drivers, cricket bats to cricket players, cameras to

photographers. In this sense the Upihoi fi nd is more than the

material remains of an ancient ship: through navigation and

the lagatoi (Motuan)-inspired bevaia experiments of the 1920s–

1930s, this Gulf Province trade vessel intimates to the world a

connected Elema and Motuan past, and in doing so nuances

and (re)affi rms the region’s history as it relates to its present-day

Indigenous communities. For, after all, it is through history that

we come to know ourselves and our relationships with others

– our achievements, our mistakes as much as our rights. The

Upihoi fi nd presents Gulf Province as well as Motuan history not

as fi xed in a past glory forever changed by colonial circumstances,

nor as geographically atomised, but as rich in local endeavours

and actively shifting and connected in relation to the changing

world. The Upihoi fi nd couples the present with the past not

simply through the static object, but more importantly through

its connections with past bevaia enterprises made possible in

networks of transcultural relations.

On a broader geographical and historical note, the Upihoi fi nd

and the bevaia experiments of the 1920s–1930s also allow us to

better understand processes of social interaction and cultural and

technological innovation and adoption between linguistic groups

across Island Melanesia. The last 20 years in particular have seen

complex debates between researchers attempting to understand

the mechanics of the spread of long-distance ‘maritime’

(especially Lapita ceramic) industries across the Western Pacifi c

(e.g. Bedford and Clark 2001; Bellwood 1991; Kirch 2000; Spriggs

2003; Terrell et al. 2001). The bevaia experiments of the Orokolo

Bay-Kerema region illustrate well the culturally dynamic nature

of recent transcultural practices between Austronesian and non-

Austronesian speaking long-distance maritime trade partners in



this region, with implications for longer-term dynamics. While

Motuan (Austronesian-speaking) ceramic manufacturers by all

accounts appear to have initiated through their seafaring prowess

long-distance maritime trade with non-Austronesian sago

producers of the Gulf Province, in time the latter came to adopt

the Motuan seafaring technology that so signifi cantly infl uenced

their lives through the annual hiri trade and the associated

establishment of ongoing long-distance trade partnerships. In

this instance it is neither pottery-making nor the manufacture

of shell valuables that came to be adopted by the Gulf Province

recipients of the maritime Motuan hiri traders – these fi nished

products remained important trade imports – but the sailing

vessels themselves, imitations of which they began to construct

(as an Indigenous enterprise but with Motuan advice) and sail in

a ‘reverse hiri’, demonstrating the selective nature of transcultural

impacts and adoptive practices across the Austronesian/non-

Austronesian divide. There are other examples in southern PNG

of non-Austronesian shipbuilders, most of whom were involved

in maritime trade and who perhaps learnt their shipbuilding

skills from Austronesian speaking peoples (e.g. Kiwai, Mailu,

Murik Lakes). Historical considerations of these promise to shed

important insights into processes of information transmission

across communities, regions and language groups.

As a fi nal note, the Upihoi fi nd is part of a collection of

national icons, and in this sense resonates with Mandui’s (2006)

notion that archaeology – and by extension material culture –

can play an important part in nation-building. Often in PNG (as

elsewhere) items of material culture attain national signifi cance

and status as a result of their scarcity, ingenuity of manufacture

or creativity. However, in PNG, widespread agreement about

national signifi cance can be a diffi cult process given the presence

of 800 linguistic groupings each of which differentially prioritises

what is important to their heritage. Historically, items regarded

as National Cultural Property have been defi ned, listed and are

protected under national law by the National Cultural Property

Preservation Act 1967, a piece of legislation administered by

the PNG National Museum and Art Gallery. The Upihoi fi nd

is a symbol not of a common history nor of a ‘shared history’,

but rather of social connectivities and social responses within

and across local communities that testify to PNG’s position in

local and global, and pre-colonial to post-colonial contexts. The

materiality of such connections and of responses to colonialism

are found in objects such as the Upihoi fi nd, and in this lies their

power to symbolise and communicate key dimensions of PNG’s

history. With these concerns in mind, the PNG National Museum

and Art Gallery is currently in negotiation with the people of

Epemeavo for the acquisition, preservation and display of the

Upihoi fi nd.

AcknowledgementsWe thank the people of Epemeavo and Kea Kea villages for

reporting the fi nd and for their warm welcome. At Epemeavo,

we thank in particular: John Erope, Moisen Iavi, Ouku Kora,

Koivi Opu, Ouhu Jim, Michael Martin, Haikoro Kora, Ivan

John, George Iavi, Hari Iavi, Haihe Hoa, Auri Hoahe, Mukari

Kouri, Auri Sam, Hearo Koivi, Jack Poi, Ken Moie, Kora Erope;

the Miaro clan elders Tom Haeae, Hoahe Merere, James Korea,

Kora Erope and Albert Moie; Eau Saropa, and Ivan Oree and

Bethy Oree for their hospitality and ailave and bilum gifts. At

Kea Kea, thanks to Sam Nao and the people of Kea Kea. Many

thanks also to Francis Bafmatuk, Acting Deputy Director of

the PNG National Museum and Art Gallery, for support for

this project; Timothy Akia of Radio Gulf for useful discussions

and friendship, and the Deputy Provincial Administrator ‘Mac’

Orisuru Avai at Kerema; Tim Denham of Monash University for

forwarding the Post-Courier announcement of the Upihoi fi nd

to BD; the School of Geography and Environmental Science,

Professor Rae Francis and the Faculty of Arts, and Professor

Edwina Cornish (DVC of Research) at Monash University, and

the Australian Research Council for grant and QEII Fellowship

DP0877782 to BD for funding this project; Kara Rasmanis for

drafting the Figures; Steve Morton for photography (Figure 10);

Brad Duncan and Mike Green for advice; the staff of the PNG

National Museum and Art Gallery for support; Alan Hogg at the

Waikato Radiocarbon Dating Laboratory for urgent radiocarbon

dating and discussions; Judith Powell, Andrew Viduka and Vicki

Richards for conservation advice; Barry Craig and the South

Australian Museum for permission to reproduce Figure 1 and

the National Archives of Australia for Figure 11; and Barry Craig,

Michael Quinnell and Paul Rainbird for insightful comments of

an earlier draft.

ReferencesAbel, C.W. 1902 Savage Life in New Guinea. London: London Missionary Society.

Allen, J.F. 1976 Fishing for wallabies: Trade as a mechanism for social interaction,

integration and elaboration on the central Papuan coast. In J. Friedman and

M.J. Rowlands (eds), The Evolution of Social Systems, pp.419-455. London:

Duckworth.

Allen, J.F. 1977 Sea traffi c, trade and expanding horizons. In J.F. Allen, J. Golson

and R. Jones (eds), Sunda and Sahul: Prehistoric Studies in Southeast Asia,

Melanesia and Australia, pp.387-417. London: Academic Press.

Anon. 2007 Lagatoi dug up on Gulf beach. Post-Courier 20 August 2007. Retrieved

18 March 2008 from http://www.postcourier.com.pg/20070820/news09.htm.

Barton, F.R. 1910 Trading voyages to the Gulf of Papua. In C.E. Seligman, The

Melanesians of British New Guinea, pp.96-120. Cambridge: Cambridge

University Press.

Bedford, S. and G. Clark 2001 The rise and rise of the incised and applied relief

tradition: A review and reassessment. In G.R. Clark, A.J. Anderson and T.

Vunidilo (eds), The Archaeology of Lapita Dispersal in Oceania, pp.69-82.

Terra Australis 17. Canberra: Pandanus Books.

Bellwood, P. 1991 The Austronesian dispersal and the origin of languages. Scientifi c

American 265:70.

Bevan, T. 1890 Toil, Travel and Discovery in British New Guinea. London: Kegan

Paul, Trench, Trübner and Co.

Bronk Ramsey, C. 1995 Radiocarbon calibration and analysis of stratigraphy: The

OxCal program. Radiocarbon 37(2):425-430.

Bronk Ramsey, C. 2001 Development of the radiocarbon calibration program

OxCal. Radiocarbon 43(2A):355-363.

Bulmer, S. 1978 Prehistoric Culture Change in the Port Moresby Region. Unpublished

PhD thesis, University of Papua New Guinea, Port Moresby.

Bulmer, S. 1982 West of Bootless Inlet: Archaeological evidence for prehistoric

trade in the Port Moresby area and the origins of the Hiri. In T.E. Dutton (ed.),

The Hiri in History: Further Aspects of Long Distance Motu Trade in Central

Papua, pp.117-130. Pacifi c Research Monograph 8. Canberra: Research School

of Pacifi c Studies, Australian National University.

Chalmers, J. 1895 Pioneer Life and Work in New Guinea 1877-1894. London:

Religious Tract Society.

Chalmers, J. 1898 Toaripi. Journal of the Royal Anthropological Institute 27:326-334.



Chester, H.M. 1878 Narrative of Expedition to New Guinea. Brisbane: Government

Printer.

David, B. in press. Rethinking cultural chronologies and past landscape

engagement in the Kopi region, Gulf Province, Papua New Guinea: 100 new

AMS radiocarbon dates. The Holocene.

Dutton, T. (ed.). 1982 The Hiri in History: Further Aspects of Long-Distance Motu

Trade in Central Papua. Pacifi c Research Monograph 8. Canberra: Research

School of Pacifi c Studies, Australian National University.

Fort, G.S. 1886 Report. British New Guinea: Report on British New Guinea,

from Data and Notes by the Late Sir Peter Scratchley, Her Majesty’s Special

Commissioner, pp.1-27. Melbourne: Government Printer.

Haddon, A.C. 1937 Canoes of Oceania: Volume 2 – The Canoes of Melanesia,

Queensland, and New Guinea. Honolulu: Bernice P. Bishop Museum Special

Publication 28.

Herbert, C.L. 1917 Delta Division. In Papua Annual Report for the Year 1914-15,

pp.86-89. Melbourne: Government Printer.

Holmes, J.H. 1903 Notes on the Elema tribe of the Papuan Gulf. Journal of the Royal

Anthropological Institute 33:125-134.

Hughes, I. 1977 New Guinea Stone Age Trade. Terra Australis 3. Canberra: Research

School of Pacifi c Studies, Australian National University.

Kirch, P. 2000 On the Road of the Winds: An Archaeological History of the Pacifi c

Islands Before European Contact. Berkeley: University of California Press.

Landtman, G. 1927 The Kiwai Papuans of British New Guinea. London: Macmillan

and Co.

Lewis, D.C. 1996 The Plantation Dream: Developing British New Guinea and Papua,

1884-1942. Canberra: The Journal of Pacifi c History.

Mandui, H. 2006 What is the future of our past? Papua New Guineans and cultural

heritage. In I. Lilley (ed.), Archaeology of Oceania: Australia and the Pacifi c

Islands, pp.379-82. Oxford: Blackwell.

May, P. and M. Tuckson 2000 The Traditional Pottery of Papua New Guinea.

Honolulu: University of Hawai‘i Press.

McAlpine, J.R., G. Keig and R. Falls 1983 Climate of Papua New Guinea. Canberra:

CSIRO and Australian National University Press.

McCormac, F.G., A.G. Hogg, P.G. Blackwell, C.E. Buck, T.F.G. Higham and P.J.

Reimer 2004 SHCal04 Southern Hemisphere calibration 11.0 cal kyr cal BP.

Radiocarbon 46:1087-1092.

Nakata, M. 2007 Disciplining the Savages: Savaging the Disciplines. Canberra:

Aboriginal Studies Press.

Oram, N. 1977 Environment, migration and site selection in the Port Moresby

coastal area. In J. Winslow (ed.), The Melanesian Environment, pp.74-99.

Canberra: Australian National University.

Oram, N.D. 1982 Pots for sago: The hiri trading network. In T.E. Dutton (ed.), The

Hiri in History: Further Aspects of Long Distance Motu Trade in Central Papua,

pp.1-33. Pacifi c Research Monograph 8. Canberra: Research School of Pacifi c

Studies, Australian National University.

Pike, G.R. and B. Craig 1999 The Usher photographic collection from the south-

west Pacifi c. Records of the South Australian Museum 31(2):215-253.

Pratt, A.E. 1906 Two Years among New Guinea Cannibals. London: Seeley and Co.

Rhoads, J.W. 1982 Prehistoric Papuan exchange systems: The hiri and its antecedents.

In T.E. Dutton (ed.), The Hiri in History: Further Aspects of Long Distance Motu

Trade in Central Papua, pp.131-151. Pacifi c Research Monograph 8. Canberra:

Research School of Pacifi c Studies, Australian National University.

Rhoads, J.W. 1994 The Popo site. In D. Frankel and J.W. Rhoads (eds), Archaeology

of a Coastal Exchange System: Sites and Ceramics of the Papuan Gulf, pp.51-69.

Research Papers in Archaeology and Natural History 25. Canberra: Research

School of Pacifi c and Asian Studies, Australian National University.

Ryan, D. 1970 Rural and urban villages: A bi-local social system in Papua.

Unpublished PhD thesis, University of Hawai‘i, Honolulu.

Salisbury, R.F. 1962 From Stone to Steel: Economic Consequences of a Technological

Change in New Guinea. Parkville: Melbourne University Press.

Schiffer, M.B. 1986 Radiocarbon dating and the ‘old wood’ problem: The case of the

Hohokam chronology. Journal of Archaeological Science 13:13-30.

Spriggs, M.J.T. 2003 Post-Lapita evolutions in Island Melanesia. In C. Sand (ed.),

Pacifi c Archaeology: Assessments and Prospects, pp.205-212. Les Cahiers de

l’Archéologie en Nouvelle-Calédonie 15. Nouméa: Départment Archéologie,

Service des Musées et du Patrimoine de Nouvelle-Calédonie.

Swadling, P. 1996 Plumes from Paradise. Boroko and Coorparoo: Papua New Guinea

National Museum and Robert Brown and Associates Pty Ltd.

Tapper, N.J. 2002 Climate, climatic variability and atmospheric circulation patterns

in the maritime continent region. In P. Kershaw, B. David, N. Tapper, D. Penny

and J. Brown (eds), Bridging Wallace’s Line: The Environmental and Cultural

History and Dynamics of the SE Asian-Australian Region, pp.5-28. Advances

in Geoecology 34. Reiskirchen: Catena Verlag.

Terrell, J.E., K.M. Kelly and P. Rainbird 2001 Foregone conclusions: In search of

‘Papuans’ and ‘Austronesians’. Current Anthropology 41(1):97-124.

Williams, F.E. 1940 Drama of Orokolo. Oxford: Oxford University Press.

Williams, F.E. 1976 ‘The Vailala Madness’ and Other Essays. St Lucia: University

of Queensland Press.

Young, M.W. and J. Clark 2001 An Anthropologist in Papua: The Photography of F.E.

Williams, 1922-39. Adelaide: Crawford House Publishing.


TIGERSHARK ROCKSHELTER (BAIDAMAU MUDH): Seascape and Settlement Reconfi gurations on the Sacred Islet of Pulu, Western Zenadh Kes (Torres Strait)Ian J. McNiven1, Joe Crouch2, Marshall Weisler3, Noel Kemp4, Lucía Clayton Martínez5,

John Stanisic6, Meredith Orr7, Liam Brady2, Scott Hocknull6 and Walter Boles8

1 Programme for Australian Indigenous Archaeology, School of Geography and Environmental Science, Monash University, Clayton, VIC 3800, Australia

² Centre for Australian Indigenous Studies, Monash University, Clayton, VIC 3800, Australia

3 School of Social Science, University of Queensland, Brisbane, QLD 4072, Australia

4 50 Derwent Avenue, Lindisfarne, TAS 7015, Australia5 Universidad Autónoma de Madrid, 28049 Madrid, Spain6 Queensland Museum, PO Box 3300, South Brisbane, QLD 4101,

Australia7 School of Geography and Environmental Science, Monash University,

Clayton, VIC 3800, Australia8 Australian Museum, 6 College Street, Sydney, NSW 2010, Australia

AbstractTigershark Rockshelter, a small midden site on the sacred islet

of Pulu in central western Zenadh Kes (Torres Strait), was

visited intermittently by small groups of marine specialists

between 500 and 1300 years ago. The diverse faunal

assemblage demonstrates procurement of turtle, dugong,

shellfi sh, fi sh, shark and ray from mangrove, reef and open

water environments. Apart from a characteristic fl aked

quartz technology, the site contains shell body adornments.

Establishment of Tigershark Rockshelter reveals increasing

preference for shoreline settlements possibly for enhanced

intervisibility, intimacy and liminality between newly-

conceptualised territorial land- and seascapes. Intensifi ed

occupation 500–700 years ago matches concomitant

demographic expansions across the region. As local

settlement patterns focused on large open village sites 500

years ago, Tigershark Rockshelter became obsolete and was

abandoned. These settlement reconfi gurations were part

of broader social transformations that eventually saw the

status of Pulu change from a residential to a ceremonial and

sacred place.

IntroductionIn 1996, one of us (IM) was invited by cultural anthropologists

John Cordell and Judith Fitzpatrick to participate in the ‘Torres

Strait Culture Site Documentation Project’ (Fitzpatrick et al.

1998; McNiven et al. 2004a). A key aim of our work was to

better understand Torres Strait Islander cultural heritage issues

and needs and towards this end the Mabuyag (aka Mabuiag)

community of central western Zenadh Kes introduced us

to the small (550m-long) islet of Pulu located off the west

coast of Mabuyag (Figure 1). Haddon (1904:3, 1935:58,

408) documented the special status of Pulu as a ‘sacred islet’

revered across western Zenadh Kes as the place where the

sacred items of the warrior hero Kuiam were housed in the

sacred totemic skull cave (Awgadhalkula) and where the

great kod (ceremonial place) of the Goemulgal (people of

Mabuyag) is located at the northern end of the islet. To assist

future management and protection of Pulu, the Mabuyag

community, along with McNiven and Fitzpatrick, conducted

site surveys during 1998 to support nomination of the islet

as an Indigenous Protected Area (IPA) under the Australian

Government’s IPA Program. To enhance understanding of the

history, cultural heritage and associated management needs of

Pulu, the Mabuyag community, through the Goemulgau Kod

cultural organisation, proposed archaeological research and

conservation of a small rockshelter midden site (subsequently

named Tigershark Rockshelter) located at the southern end of

Pulu. Unlike Awgadhalkula and the kod, this enigmatic site did

not feature in local oral histories. Three key aims of the work

were to: (1) allow a cross-cultural exchange which would assist

the Mabuyag community to become better acquainted with

the pros and cons of archaeological excavation, and outside

researchers to be better acquainted with appropriate cultural

protocols for researching Torres Strait Islander cultural places;

(2) enhance existing historical and cultural knowledge of Pulu;

and (3) help protect the site from further pig disturbance

by placing galvanised wire mesh across the surface of the

midden (McNiven et al. 2004a:83). The joint project between

the Mabuyag community, McNiven and Fitzpatrick was the

fi rst community-based, partnership archaeological project in

Figure 1 Zenadh Kes (Torres Strait).


Tigershark Rockshelter (Baidamau Mudh): Seascape and Settlement Reconfi gurations on Pulu

N

01m

Base of boulder

Base of boulder

C

A

B

bedrock

Datum

bed rock

bed rock

SL

OP

E

S L O P E

4

grassy flat

SL

OP

E

grassy flat

grassy flat

grassy flat

grassy flat

termitemound

A¹

2

3

1

D

A

excavation pit

granite

midden(main surface exposure)

boulder overhang(outer edge)

surface collected item

Figure 3 Plan and cross-section of Tigershark Rockshelter.

Figure 2 Excavation of Square A by Iona Mooka with John Bani reading Volume 5 of the Haddon Reports, Tigershark Rockshelter, looking south (Photograph: Ian J. McNiven).


Ian J. McNiven et al.

western Zenadh Kes. Focusing on our second project aim, this

paper details excavation results and an analysis of changing

Torres Strait Islander seascape engagements over the past

1500 years.

Tigershark RockshelterTigershark Rockshelter is a picturesque site set amongst scattered

trees (including vine forest) and grass. It has commanding views

over territorial seas of the Goemulgal and is well-positioned for

easy access to intertidal reefs and a range of other marine resource

zones (Figure 2). The rockshelter is formed by two large granite

boulders resting on granite bedrock (Figure 3). The southern

edge of the overhang forming the rockshelter is 22m inland, and

7m above the high water mark. Cultural deposit, in the form of

black sandy sediments peppered with marine shells and small

bone fragments, is located underneath much of the area sheltered

by the double boulder overhang. It was the presence of large tiger

shark teeth on the surface and within the excavated deposits that

gave rise to the name of the site which translates in Mabuyag

language (Western-Central Language) as Baidamau Mudh

(Baidamau = the possessive form of Baidam = shark, mudh =

house). Lush vine growth covers much of the western half of the

midden during the wet season (December-April). The western

margins of the deposit are sectioned by water erosion (rainfall

run-off) to a depth of c.20cm down to bedrock. A 2m-wide zone

of bedrock is exposed within the dripline across the western and

southern margins of the site. Camping within the rockshelter is

restricted to a 10m x 5m area under the large southern boulder

as the fl oor area under the northern boulder mostly has a low

<20cm-high roof. However, considerable midden deposit occurs

beneath the northern boulder overhang. The midden covers an

area of approximately 9m (N-S) by 6m (E-W).

ExcavationsFour 50cm x 50cm excavation squares (Squares A, B, C and D)

were positioned across the site to sample spatial variation in the

midden (Figure 3). Squares A and C were positioned towards

the middle of the occupation zone where midden deposit was

predicted to be deepest (based on topography of surrounding

bedrock to the north and south) with minimal treadage impact

(overhead roofl ine elevated <1m above midden surface).

Squares B and D were located closer to the dripline. Square

D was positioned to recover the only fragments of bottle

glass seen on the surface. Squares A, B and C were excavated

to bedrock. As Square D was excavated primarily to recover

surface glass artefacts, it was stopped arbitrarily after a single

Excavation Unit (XU). The number of XUs and maximum

depth of midden exposed in each square is as follows: Square

A (11 XUs, 32.5cm); Square B (10 XUs, 29.0cm); Square C (10

XUs, 31.5cm); and Square D (1 XU, 2.5cm). XUs were thin to

increase chronological and stratigraphic control, averaging

2.7cm-thick (range: 1.2–5.1cm). The weight of each XU was

recorded and depth elevations were taken at the corners and

centre of each square at the beginning and end of each XU.

All excavated deposit was wet-sieved through 3mm and 1mm

mesh using seawater adjacent to the site, followed by thorough

rinsing in freshwater and drying back on Mabuyag. Analyses in

this paper are based on the 3mm sieve materials from Squares

A, B and C (and only bottle glass and shark teeth from Square

D). The 1mm materials are unanalysed. Samples of <3mm dry

sieved sediment were taken for each XU. A total of 302.6kg of

deposit was excavated. Squares A, B and C were backfi lled with

coralline sands from the adjacent beach. This site was excavated

between 27 March and 6 April 1999 at the end of the wet season

(Figure 3).

Radiocarbon Dating and ChronologyChronology of the midden was determined with seven

radiocarbon dates on wood charcoal – four from Square A

and three from Square C (Table 1). While the species of dated

wood charcoal is unknown, it probably derives from small dead

branches of local trees that were easy to harvest by hand (stone

axes being uncommon in the region). Old wood, in the form

of dead trees, is a rare occurrence in Torres Strait given the

tropical climate, termites, and frequent bush fi res. Furthermore,

small branch wood is better suited to small camp fi res in a

small rockshelter. As such, the ‘inbuilt age’ and ‘old wood’

problem of radiocarbon dates is unlikely to be more than a few

decades. Dating was undertaken by the University of Waikato

Radiocarbon Dating Laboratory and the ANTARES facility at

ANSTO (Fink et al. 2004). Radiocarbon dates were calibrated

into calendar years using OxCal 4.0 (Bronk Ramsey 1995, 2001)

and the southern hemisphere calibration dataset (SHCal04)

(McCormac et al. 2004). For convenience and heuristic purposes,

a single, central best-point estimate was calculated for the

irregular probability distribution of each date using the median

calibrated age (following Telford et al. 2004). These single dates

are also expressed as ‘years ago’ (cal BP – i.e. before AD 1950)

and have been rounded to the nearest 50 years to emphasise

their approximate (circa) status. All dates reveal increasing age

with depth suggesting the deposit has high stratigraphic integrity

with minimal vertical movement of materials. The basal XU of

Square A (XU11) and Square C (XU10), both resting on bedrock,

produced similar dates of 1390±50 BP (1250 years ago) and

1476±31 BP (1300 years ago) respectively. As such, camping at

the site commenced approximately 1300 years ago. The surface

(XU1) of Squares A and C dated to 490±45 BP (500 years ago) and

595±40 BP (550 years ago) respectively, suggesting that regular

camping at the site ceased by 500 years ago. However, bottle glass

fragments indicate minor activity in the late nineteenth/early

twentieth century (see below). Dates for the dramatic increase in

shell density in Square A (XU7) and Square C (XU7) are 830±45

BP (700 years ago) and 655±40 BP (600 years ago) respectively.

Thus, the shell-rich upper levels of the site, taking in XUs 1–7 of

Squares A and C and the top 17–23cm of the deposit, accumulated

between 500 and 700 years ago, whereas the lower levels of the

site, taking in XUs 8–10/11 of Squares A and C and the bottom

10–12cm of deposit, accumulated between 700 and 1300 years

ago. It is likely that the lower (shell-poor) zone accumulated at

a reasonably constant rate given that a date of 1102±29 BP (950

years ago) from midway through this zone (XU9, Square A) is

also approximately halfway between the chronological span of

this zone. This constancy of accumulation is consistent with the

relative homogeneity of sediments (see below).

Stratigraphy and Sedimentary EnvironmentSediment in all four squares was a homogenous black (Munsell:

10YR 2/1), loose (unconsolidated), soft, granitic sand. Cultural



materials (e.g. marine shells, bones, stone artefacts and charcoal)

and small fragments of granite occur through the deposit. Lack

of sterile layers indicates that sediments accumulated only

during periods of occupation. No stratigraphic changes were

observed (Figure 4). The upper c.20cm of sediment is moist and

shell-rich with pH values ranging from 9.5–10.0 (surface) to 8.5

with depth. In contrast, the lower c.10cm of sediment resting

on bedrock is wet and bone-rich with pH values mostly 7.0 to

6.5. The wet nature of lower sediments is most likely a result of

rainwater seeping into the rockshelter and being trapped by a

slight depression in the underlying granite bedrock (Figure 3).

Subsequent visits to the site in the late dry season (September-

November) revealed considerable drying of the midden deposit

and dieback of the vine cover.

Sediment Particle Size AnalysisLaser diffraction particle size analyses were undertaken on the

<3mm sediment samples from Square A by one of us (MO).

Two samples of less than 1cm³ each were selected in the form of

multiple small aggregates from each of the XU1 to XU11 samples.

The materials were readily mechanically dispersible and the

samples were vigorously shaken in distilled water and placed in a

vortex mixer for at least one minute. These were allowed to settle

for a few minutes before being poured through a 1mm sieve

into a Beckman Coulter LS100Q particle size analyser. Results of

the two samples for each excavation unit were compared and if

there was any discrepancy in the particle size distribution curves

then a third sample would be analysed. Averaged values for the

percentages by volume of clay and silt (<20µm), fi ne sand (20–

200µm), medium sand (200–600µm) and coarse sand less than

1mm (600–1000µm) were graphed by depth (Figure 5).

Total variability in the proportions of clay and silt, fi ne sand,

medium sand and coarse sand less than 1mm in size with depth

through the profi le is low. Three samples (XU1, XU4 and XU9)

have minor depletions of clay and silt size material of 3–8% by

volume relative to adjacent samples, and excluding these samples

there is a minor increase in clay content down the profi le, of

4% by volume from XU2 to XU11. Samples XU7, XU8 and

XU9 have slightly higher amounts (3–5% by volume) of coarse

sand less than 1mm in size relative to the surrounding materials.

Overall, the variations are small and Square A can be considered

as having a uniform profi le in terms of particle size distributions.

Though small, the greatest variations lie in the relative depletion

of clays and silts at the surface 0–2cm (XU1) and at 6.2–8.5cm

depth (XU4), which suggests some degree of surface exposure

after deposition with selective removal of fi ne materials by either

post-depositional surface disturbance or eluviation. The removal

of clays and silts from the surface (XU1) is consistent with

exposure to the elements for c.500 years. The higher silt and clay

content (37%) at 8.5–13.9cm depth (XU5 and XU6) suggests less

post-depositional disturbance (an hypothesis consistent with

increased sedimentation rates – see below) and/or fi nes illuviation

related to soil formation. Relative homogeneity of sediments

from XU7 to XU11 and the lack of associated obvious erosional

features further support the inference drawn from radiocarbon

dating that the lower levels of the site exhibit relatively consistent

depositional rates without major occupational/depositional

hiatuses. Indeed, sediments in Square A reveal minimal change

moving from the shell-poor lower levels (XUs 8–11) into shell-

rich levels (XUs 1–7).

Sedimentation RatesRadiocarbon dates reveal that the rate of sediment accumulation

changed dramatically through time. For example, the basal

lower levels of Square A (XUs 10–11) represent 6.1cm of deposit

accumulated over 300 years (>950 to 1250 years ago). This

translates to a sedimentation rate of 2.0cm/100 years. Similarly,

the lower levels of Square A (XUs 8–9) represent 6.1cm of

deposit accumulated over 250 years (>700 to 950 years ago)

and a sedimentation rate of 2.4cm/100 years. In contrast, the

upper levels of Square A (XUs 1–7) represent 16.6cm of deposit

accumulated over 200 years (500 to 700 years ago) and a much

higher sedimentation rate of 8.3cm/100 years. In other words,

the sedimentation rate for the upper levels of the site is 3.5–4

times higher than for lower levels.

Land SnailsA total of 13 land snail taxa were recovered from Squares A, B

and C (Table 2). The diversity of snail species in upper levels of

the site (XUs 1–6) ranges from 8 to 11 whereas in lower levels

Square XU Type/Sample Weight

(g)

Depth (cm)

Lab. No. δ13

C‰

14C Age

(years BP)

Calibrated Age BP 1σ

(probability)


(probability)

Calibrated Age BP Median

Years Ago

(cal BP) circa

A 1 C8.6

0-2.0 Wk-8496 -26.5±0.2

490±45 534-486 (64.0%) 481-474 (4.2%)

550-446 (89.9%) 361-334 (5.5%)

503 500

A 7 C 7.0

13.9-16.6 Wk-8497 -25.8±0.2

830±45 732-675 (68.2%) 788-658 (95.4%) 709 700

A 9 C 10.5

19.4-22.7 Wk-19548 -26.2±0.2

1102±29 969-929 (68.2%) 1053-1030 (9.5%)1005-921 (85.9%)

954 950

A 11 C 13.0

25.7-28.8 Wk-7304 -26.1±0.2

1390±50 1302-1236 (53.1%)1205-1183 (15.1%)

1348-1170 (94.2%)1159-1144 (1.2%)

1259 1250

C 1 AMS0.3

0-2.8 OZH831 -28.2 595±40 624-610 (14.6%)560-525 (53.6%)

637-592 (29.5%)570-507 (65.9%)

551 550

C 7 AMS 0.4

17.7-22.8 OZH832 -25.7 655±40 640-591 (50.7%)571-553 (17.5%)

656-545 (95.4%) 604 600

C 10 AMS 3.5

29.3-32.6 NZA-26585 -27.1 ±0.2

1476±31 1342-1298 (68.2%) 1378-1287 (95.4%) 1324 1300

Table 1 Radiocarbon dates for Tigershark Rockshelter. C = Conventional.



(XUs 7–11) the range is 0 to 6 (Table 2). The drop-off in species

diversity (and abundance) in lower levels most likely refl ects poor

preservation conditions (see below). Preservation issues aside,

the range of species is typical of the land snail fauna found in

the western and central islands of Zenadh Kes (Stanisic in press).

Most of these species have been recorded from the nearby islands

of Badu and Mua and are indicative of a vine thicket habitat. Vine

thicket occurs in both foredune and rock outcrop situations on

many of the Zenadh Kes islands. Most of the species recovered

from Tigershark Rockshelter are obligate vine thicket species

(Hadra funiculata, Austrochloritis buxtoni, Pravonitor krefftii,

Torresiropa spaldingi, Pleuropoma gouldiana) and very few

(Gastrocopta pediculus, Pupoides pacifi cus, Eremopeas tuckeri) are

also inhabitants of savannah woodland. The number of species

recorded in XUs 1–6 is within expectations for a mature Zenadh

Kes vine thicket land snail community (Stanisic in press). Vine

thicket is a rainforest community that relies on both moisture

and the absence of fi re for survival. As anthropogenic landscape

burning was most likely a feature of Pulu in the past, as it was

on many Zenadh Kes islands (e.g. McNiven 2008; Rowe 2006),

the temporal persistence of vine thicket land snails indicates that

between at least 500 and 700 years ago Tigershark Rockshelter

and its immediate surrounds were deliberately excluded from

fi re impact as part of a broader fi re management strategy for

the islet. Controlled landscape burning on Pulu 500–700 years

ago continued through to the nineteenth century as Haddon

(1904:370) records that one of the key roles for resident caretakers

of Pulu was ‘bush fi re’ control.

Marine Invertebrates

Shellfi shOf the 3678.8g of marine shell recovered, most (99.0% or 3640.6g)

was economic shell – defi ned as shells (including fragments)

with a known or extrapolated whole length of ≥15mm. Most

shell occurs within the upper sections of the deposit with the

bulk (80%) recovered from Squares A and C (Figure 6). The

fundamental difference in shell occurrence though the deposit is

revealed by the combined discard rate of economic marine shells

in Squares A and C: 9.2g/100 years in lower levels (XUs 8–10/11)

dated >700–1300 years ago and 1451.1g/100 years in upper levels

(XU1–7) dated 500–700 years ago.

A total MNI of 933 economic shells was represented by 181

bivalves, 752 gastropods, and 15 species with foraging focused

on the mudwhelk Terebralia sulcata from mangrove habitats

(64.1% of total MNI) (Table 3). Of secondary importance

(32.0% of total MNI) was Nerita spp., Paphies striata, Pinctada

spp. and Tellina spp. from mangrove, coral reef and sandy

habitats. No signifi cant vertical patterning was evident in the

relative occurrence of these fi ve species. The remaining 10

species were obtained infrequently and intermittently from

mangrove, coral reef and sandy habitats, with a combined

MNI of 36 (3.9% of total MNI). All shellfi sh could have been

obtained from marine environments within 1km of the site.

While shellfi sh species at Tigershark Rockshelter fall within the

range of western Zenadh Kes shellfi shing practices, the focus

on T. sulcata is atypical as is the absence of mudu (Anadara

sp.) and Chama sp. (e.g. David and Weisler 2006; Ghaleb

1990; McNiven 2006). Variations in the range and focus of

shellfi shing at different sites is attributed mostly to differences

in local availability and dietary preference.

Shell TaphonomyDoes the negligible representation of shell in the lower levels

of the deposit refl ect minimal shell discard or poor shell

preservation? According to Stein (1992) groundwater saturation

of middens can produce complex chemical changes and decrease

shell preservation conditions. Yet it is unlikely rainwater seepage

chemically removed shell from the lower levels of Tigershark

Rockshelter for the following reasons. First, the depth at which

the shell midden appears is on average 10.5cm below datum

in Square A (i.e. XU7), 0.5cm below datum in Square C (i.e.

XU7), and 2cm above datum in Square B (i.e. XU3). If rainwater

pooling across the lower sections of the deposit was responsible

for removing shell then the interface level between preserved and

unpreserved shell should be similar for all squares, not differ by

10–12.5cm. Second, conditions conducive to shell preservation

in lower levels are indicated by pH values ranging from 9.0 to 6.5

and the presence of bones, including tiny fi sh bones (see below).

Third, no obvious patterning of increasing shell weathering and

fragmentation with depth occurs through the deposit. However,

in contrast to upper levels, lower levels of deposit contain no

foraminifera discs and few snail shells. Foraminifera discs were

probably brought inadvertently from the adjacent beach to the

site attached to people’s wet feet after alighting from canoes (as

they did after alighting from the dinghy during excavation of the

site). As such, foraminifera should occur throughout the deposit,

not just the upper levels. Similarly, land snails would have been

a feature of the site throughout its history. Overall, taphonomic

analysis suggests that while preservation may have been an issue

for more fragile marine shells in the lower levels of the site, the

paucity of economic shell in these lower levels is largely cultural

and a refl ection of low shellfi sh consumption between 700 and

1300 years ago.

CrustaceaNone of the 24.9g of crustacean exoskeleton fragments

identifi ed (e.g. mouth parts) is crayfi sh. Most fragments are from

crabs (mostly claw fragments from small and medium-sized

individuals) readily available from around granite rocks amongst

mangroves adjacent to the site and from nearby reef fl ats at low

Figure 4 Stratigraphy of Squares A, B and C, Tigershark Rockshelter.



tide (cf. Johannes and MacFarlane 1991:213, 216, 218). Crab

exoskeletons form a small part of faunal assemblages from other

excavated sites in western Zenadh Kes dating to the past 2600

years (e.g. Crouch et al. 2007; David and Weisler 2006; Ghaleb

1990:274; Harris et al. 1985:17).

Marine VertebratesNearly 40,000 bone fragments weighing 2766.9g were

recovered. The small size of bone fragments is revealed by the

mean weight (0.07g) and mean maximum length (7.3mm) of

the 12,663 bone fragments in Square A. Relative consistency

exists in the density of bone through the upper seven XUs of

each square dating between 500 and 700 years ago (Figures

6-7). Major differences in bone density in the lower levels of

squares probably reflects spatial differences in bone discard

across the site >700–1300 years ago. Despite spatial changes

in discard, the discard rate of bone in Square A increased

through time from 92.2g/100 years in lower levels (XUs 8–11)

dated >700–1300 years ago to 153.5g/100 years in upper levels

(XU1–7) dated 500–700 years ago.

Most (70.7%) of the bone assemblage was designated ‘large

vertebrate’ which is likely to represent mixed turtle/dugong

bone given the range of known taxa and morphology of bones.

The remaining assemblage comprised mostly ‘small vertebrate’

(13.0%), dugong (12.3%) and turtle (3.8%) with minor traces

of bird (0.2%), snake/lizard (0.1%) and rodent (<0.1%). Most of

the ‘small vertebrate’ is probably fi sh, a view confi rmed by follow-

up recategorisation of 81% of ‘small vertebrate’ bone in Square

A as ‘fi sh’ by MW. Overall, each level of the site is dominated by

‘large vertebrate’ followed by ‘small vertebrate’ (i.e. mostly fi sh) in

upper levels dating 500–700 years ago and dugong in lower levels

dating >700 to 1300 years ago.

TurtleThe 104.8g of turtle (waru) bone (mostly osteoderm fragments)

is considered an extreme under-estimate of the representation

of turtle in the bone assemblage. Most of the ‘large vertebrate’

category of bone is probably turtle as little ‘large vertebrate’ bone

appeared to be dugong. As such, turtle is considered to have

been the focus of marine subsistence activities at Tigershark

Rockshelter. Such dominance follows ethnographic recordings

of the past 150 years where the green sea turtle (Chelonia mydas),

weighing 100–150kg, is the key marine protein contributor to

Torres Strait Islander diets (Johannes and MacFarlane 1991;

McNiven and Hitchcock 2004; Smith and Bliege Bird 2000).

Turtle is also a key food item recovered from other archaeological

sites in western Zenadh Kes, including Mask Cave (also on Pulu)

1500–4000 years ago (McNiven et al. 2006), Berberass on Badu

2500–4000 years ago (Crouch et al. 2007), and from the last

1000 years at Goemu on Mabuyag (Ghaleb 1990, 1998), Dauan

Figure 5 Sediment distribution, Square A, Tigershark Rockshelter.

Au

stro

chlo

riti

s b

uxt

on

i

Trac

hio

psi

s st

ran

gu

lata

Ere

mo

pea

s tu

cker

i

Hel

icar

ion

id s

p.

Torr

esit

rach

ia

torr

esia

na

Pra

von

ito

r k

reff

ti

Pu

po

ides

pac

ifi c

us

Ple

uro

po

ma

go

uld

ian

a

Had

ra f

un

icu

lata

Dis

coch

aro

pa

aper

ta

Tru

nca

tell

a g

uer

inii

Torr

esir

op

a sp

ald

ing

i

Gas

tro

cop

ta

ped

icu

lus

Nu

mb

er o

f ta

xa

XU

A B C A B C A B C A B C A B C A C A B C A C A C A A B C A C B C A B C Tota

l

1 • • • • • • • • • • • • • • • • • • • 7 7 5 8

2 • • • • • • • • • • • • • • • • • • • 7 6 6 9

3 • • • • • • • • • • • • • • • • • • • 8 5 6 9

4 • • • • • • • • • • • • • • • • • 6 1 10 11

5 • • • • • • • • • • • • • • • • • • • 9 1 9 10

6 • • • • • • • • • • • • • • • 6 0 9 10

7 • • • • • • • • • 3 0 6 6

8 • • • • • • 5 0 1 5

9 • 0 0 1 1

10 0 0 0 0

Table 2 Land snails, Squares A, B and C, Tigershark Rockshelter.



4 (McNiven 2006), and Kurturniaiwak on Badu (David and

Weisler 2006).

DugongDugong (Dugong dugon) (dhangal) bone (mostly rib fragments)

(340.2g) was recovered from all levels of the site, albeit with

uneven distribution between squares (Figure 7). Dugong

hunting and consumption was an important activity and

continues to be so for the Mabuyag community (Kwan 2005).

That people of western Zenadh Kes have been hunting dugong

for at least the past 4000 years is revealed by excavations at

occupation sites (e.g. Berberass – Crouch et al. 2007; Mask Cave

– McNiven et al. 2006; Goemu – Ghaleb 1990, 1998) and by

ritual dugong bone mounds dating to the past 400–500 years

(David and Mura Badulgal Committee 2006; McNiven and

Bedingfi eld 2008; McNiven and Feldman 2003).

FishAll fi sh and ‘small vertebrate’ bones were given to MW for

identifi cation to nearest taxon. Most of the ‘small vertebrate’ bones

are highly fragmented fi sh bones that, for the most part, could

not be identifi ed further. For the purposes of this paper, analysis

was limited to the 144g of fi sh bone from Square A. In general, all

bone was well-preserved with <5% by weight burnt. Similar to

other assemblages from Zenadh Kes (e.g. David and Weisler 2006;

McNiven et al. 2006), the bone comes from exceptionally small

individuals. Consequently, the entire collection was examined

under 10x magnifi cation with a binocular scope. The diameter

of all measurable vertebrae and widths of lower pharyngeals of

parrotfi sh (Scaridae) and wrasses (Labridae) were taken with

digital callipers to two decimal places. Measurements were taken

to provide an approximate indication of individual fi sh size.

Identifi cation to nearest taxon was made with a reference

collection consisting of approximately 400 specimens representing

50 families (Weisler 2001:Appendix 3, plus subsequent additions).

Quantifi cation was by number of identifi ed specimens (NISP)

using the standard fi ve-paired cranial bones (dentary, articular,

quadrate, premaxillary and maxilla) as well as unique elements

including, in this analysis, dermal spines of porcupinefi shes

(Diodontidae), dermal denticles of stingrays, vertebrae of sharks/

rays (Elasmobranchii), the fourth epihyal of parrotfi sh, and

pharyngeals of wrasses and parrotfi sh.

Table 4 lists the NISP of fi sh (including sharks/rays) by XU

for Square A. Shark/ray vertebrae and dermal denticles are found

throughout all XUs; however, even small sharks have more than

150 vertebrae and individual rays can have thousands of dermal

denticles making the recovery of these more likely. Additionally,

each porcupinefi sh can have more than 200 dense dermal spines

and are often over-represented in assemblages. Considering

these quantifi cation biases, the rank-order abundance of all taxa

based on NISP is: 1–sharks/rays (Elasmobranchii); 2–wrasses

(Labridae); 3–emperors (Lethrinidae); 4–groupers (Serranidae);

5–parrotfi sh (Scaridae); 6–puffers (Tetraodontidae); and

7–trevallies (Carangidae), porcupinefi shes (Diodontidae) and

snappers (Lutjanidae). If only the fi ve-paired cranial bones are

considered (which will reduce sampling bias), the rank-order

abundance of taxa based on NISP is: 1–emperors; 2–wrasses;

3–groupers; 4–puffers; 5–parrotfi sh; and 6–trevallies and

snappers. Both bones of trevally (dentary and articular) are from

small individuals similar to the genus Caranx. Two dermal spine

fragments of porcupinefi sh are probably from the same small

individual. This is a new archaeological recording for Zenadh

Kes for a fi sh that continues to be caught today (Johannes and

MacFarlane 1991:217).

Some 163 non-shark/ray vertebrae averaged 3.72±1.82mm,

ranging from 1.59 to 12.09mm, with a median of 3.17mm.

Comparisons with vertebrae from the modern reference

collection suggest that the reconstructed length of the average

fi sh from the Tigershark Rockshelter assemblage is well below

10cm in length. Wrasses tend to be over-represented as they have

Taxon MNI % Tidal Zone SubstrateGastropods

Terebralia sulcata 598 64.1 intertidal mangrove mud

Nerita spp. 125 13.4 upper intertidal rocks + mangrove trees

Monodonta labio 8 0.9 upper intertidal rocks + mangrove trees

Turbo spp. 4 0.4 intertidal + subtidal rocky/coral reefs

Cerithiidae 3 0.3 intertidal mud/sand

Cantharus fumosus 2 0.2 intertidal coral reefs

Oliva spp. 2 0.2 intertidal sand/coral reefs

Cypraea spp. 2 0.2 intertidal rocky/coral reefs

Muricidae 1 0.1 intertidal + subtidal sand/coral reefs

Patelloida spp. 1 0.1 intertidal rocky

Bivalves

Paphies striata 76 8.1 intertidal + subtidal sand

Pinctada spp. 59 6.3 intertidal + subtidal mud/sand/coral reefs

Tellina spp. 39 4.2 intertidal sand/muddy sand

Polymesoda erosa 5 0.5 intertidal mangrove mud

Asaphis violascens 2 0.2 intertidal + subtidal sand

Chitons

Polyplacophera 6 0.6 intertidal + subtidal rocks

Total MNI 933 100

Table 3 Economic shellfi sh MNI, Squares A, B and C combined, Tigershark Rockshelter.



upper and lower pharyngeal grinding plates that are covered

with dense enamel that preserves well, even when fragmentary.

The width of nine lower pharyngeals averaged 14.27±6.11mm,

ranging between 9.76 and 28.68mm with a median of 13.01mm.

Reconstructed total lengths of fi sh are generally less than 10cm

when compared to the bones of modern reference specimens.

Some of these elements are consistent with the wrasse (cf.

Bodianus sp.). Emperors represented some of the largest

individuals in the bony fi sh assemblage with half approaching

20cm in length when compared to data from the modern

reference collection. Two snapper bones (both right articular

and dentary) are probably from the same small individual.

Parrotfi sh are extremely common in Pacifi c sites, yet are

uncommon in this assemblage. One lower pharyngeal grinding

plate measured only 4.96mm in width and is from a fi sh <10cm

in total length (after Fleming 1986). Only one of the six bones

identifi ed is from a medium-sized fi sh. Groupers are also some

of the larger fi sh in the assemblage with three medium-sized

individuals and the rest quite small. Puffers are only known

from three other sites in Zenadh Kes (Crouch et al. 2007; David

and Weisler 2006; McNiven et al. 2006), yet the taxon has dense

enamel portions on the beak-like dentary and premaxillary that

preserve well. The specimens here represent extremely small

individuals (again, probably <10cm in length) and may have

been captured for their toxic entrails and skin, used elsewhere

in the Pacifi c as a poison.

The small size of the fi sh represented in this assemblage,

and many sites throughout Zenadh Kes, refl ects a casual,

opportunistic capture strategy by people walking along the reef

fl at at low tide gleaning fi sh stranded in tide pools and shallow

channels (see McNiven et al. 2006:61-62 for further discussion of

reef gleaning by men, women and children). In this context, fi sh

could be easily grabbed by hand or speared. It is also possible

gleaning was assisted by chemically stunning fi sh. On Mabuyag,

‘a fi sh poison known as itamar was used in coastal pools among

the rocks (until its use was made illegal)’ (Barham and Harris

1987:30). Many Thursday Island residents prefer smaller fi sh

as they taste ‘sweeter’ (Vic McGrath, pers. comm.) and with

an abundance of easily obtained large ‘meat packages’ such as

marine turtle and dugong, fi nfi sh may never have been a much-

desired protein source, but were consumed to add diversity to

the diet. Certainly the entire assemblage in Square A would not

total more than ~1kg of fi sh and perhaps ~5kg of small sharks

and rays. Even taking this low overall weight of the fi nfi sh and

sharks/rays represented in Square A, there are two interesting

trends in the weight of this bone. The early period of occupation

(represented by XUs 8–11) represents only 9.1g/100 years of fi sh

bone accumulation, whereas the later period (XUs 1–7) mirrors

those trends documented for the marine shellfi sh with a much

more rapid accumulation; that is, 44.6g/100 years, nearly a fi ve-

fold increase. This marked increase in fi sh consumption may have

placed hunting pressure on the limited abundance of inshore

species as evidenced by a decline in average fi sh size as represented

by the vertebrae diameter: early period = 4.24±2.21mm (n=42)

versus late period = 3.54±1.63mm (n=121). While this appears to

be a signifi cant decline, the overall amount of fi nfi sh represented

in these two periods is exceedingly small in contrast to other

protein sources.

The dominance of wrasses (Labridae) and emperors

(Lethrinidae) in the Tigershark Rockshelter fi sh assemblage is

variably matched at other sites across Zenadh Kes. For example,

at Berberass, wrasses dominated 2500–2600 years ago while

emperors dominated 2600–4000 year ago (Crouch et al. 2007).

Wrasses and parrotfi sh are key taxa in fi sh assemblages on Mer

and Dauar (eastern islands) for the past 2600 years (Carter 2004)

and at Goemu on Mabuyag and Kurturniaiwak on Badu for the

past 800 years (David and Weisler 2006; Ghaleb 1998).

SharkFurther insights into shark taxa at Tigershark Rockshelter is

provided by 60 shark teeth recovered from Squares A, B, C and

D. Teeth were identifi ed by one of us (NK) using comprehensive

comparative shark teeth collections at CSIRO Marine Research,

Hobart. Nine shark taxa representing four shark families

were identifi ed (Table 5). Most (n=27, 45%) teeth came from

wobbegongs, Orectolobus spp. and tiger sharks, Caleocerdo

cuvier (n=14, 23%), with the remaining 32% represented

by whalers, Carcharhinus spp. (n=7), blacktip reef sharks,

Carcharhinus melanopterus (n=2), Australian blacktip shark,

Carcharhinus cf. tilstoni (n=2), silky sharks, Carcharhinus cf.

Figure 6 Vertical changes in shell and bone density, Squares A, B and C, Tigershark Rockshelter.



falciformis (n=2), lemon sharks, Negaprion acutidens (n=2),

hammerheads, Sphyrna spp. (n=2), and grey nurses, Carcharias

taurus (n=1). As wobbegong and tiger shark teeth are also

distributed through levels down to XU9, Square A, it is clear

that shark procurement at Tigershark Rockshelter focused on

the capture of these two species between 500 and 950 years ago.

No shark teeth were recovered from basal levels of the site dated

950 to 1300 years ago.

Around 50 species of shark frequent the waters of Zenadh

Kes (Allen 1997; Grant 1995). As such, the nine shark taxa at

Tigershark Rockshelter, combined with a focus on two species,

reveals highly selective and specialised shark fi shing strategies by

Torres Strait Islanders 500 to 950 years ago. However, it needs to be

kept in mind that in Zenadh Kes the whaler genus Carcharhinus

is represented by at least 17 species and the wobbegong genus

Orectolobus is represented by two species. Thus, the range of

shark species represented by the shark teeth assemblage is likely

to be greater than that documented. Despite these identifi cation

issues, the restricted range of shark species is better illustrated

by the fact that only three of the 16 shark families known for the

Strait are found in the site.

It is diffi cult to determine shark size from tooth size given

the wide range of tooth sizes within a single individual shark’s

jaw. Despite this limitation, the large tiger shark tooth from XU7,

Square A measures 25mm across and comes from an individual

2.5–3m in length and weighing over 100kg. Yet size estimates

taken for shark/ray vertebrae indicate most other sharks

represented in the assemblage are small individuals. The width of

51 vertebrae averaged 4.37±1.32mm, range 2.60–8.75mm with a

median of 4.28mm. One modern reference specimen of whaler

or requiem shark (Carcharinidae Carcharinus sp.) weighed 1750g,

735mm-long and has 157 vertebrae with an average width of

6.05±2.01mm, ranging from 1.58 to 9.56mm. With the majority

of the identifi ed shark teeth assigned to this family, it is likely that

most of the vertebrae represent sharks probably weighing less

than 2kg and under 1m long.

All nine shark taxa are known to frequent the intertidal zone

and inshore reef environments (Allen 1997; Last and Stevens

1994). In Zenadh Kes, sharks are often attracted to shallow (<1m-

deep) water near the shoreline when blood runs down the beach

during butchering of dugongs and turtles (McNiven pers. obs.).

While smaller sharks at the site could have been caught with a

hook and line or speared, the larger sharks were most probably

speared or harpooned from either a canoe or the shoreline

during beach butchering.

Tigershark Rockshelter provides the fi rst detailed and

comprehensive archaeological evidence for shark consumption

by Torres Strait Islanders. Previous archaeological evidence for

shark consumption includes ‘Elasmobranchii (sharks and rays)’

from nearby Mask Cave (also on Pulu) in levels dated 1500–2600

years ago (McNiven et al. 2006:61); a ‘tooth from a small shark

(Carcharhinus sp.)’ from Sokoli on Dauar in the eastern Strait in

levels dated 900–1400 years ago (Carter 2004:220, 259, 489); a lower

jaw tooth from a 1.2–1.5m-long whaler (Carcharhinus sp.) from

Goemu on Mabuyag dating to within the past 600 years (Ghaleb

1990:292, 1998); ‘Elasmobranchii (sharks and rays)’ and a probable

shark tooth from Kurturniaiwak on Badu dated c.650–750 years

ago (David and Weisler 2006); and a shark tooth fragment from

Dauan 4 in a level dating 450–600 years ago (McNiven 2006).

Torres Strait Islanders continue to eat a range of shark

species – banded wobbegong (Orectolobus ornatus), tiger

shark (Galeocerdo cuvier), blacktip reef shark (Carcharhinus

melanopterus), blind shark (Brachaelurus colcloughi), epaulette

shark (Hemiscyllium ocellatum), spot-tail shark (Carcharhinus

sorrah), blacktip shark (Carcharhinus limbatus) and blackspot

shark (Carcharhinus sealei) (Harris et al. 1995:12; Johannes and

MacFarlane 1991:214, 221-222). Of these eight species, three

were also procured by people camping at Tigershark Rockshelter

at least 500 years ago – wobbegongs, tiger sharks and blacktip

reef sharks. Thus, the number of shark species caught and eaten

by Torres Strait Islanders has remained fairly constant over the

past 1000 years.

RaysShark/ray vertebrae (centra) and ray dermal denticles (18.4g)

occurred in all levels of the site and reveal that ray procurement

and consumption took place throughout the site’s history.

While none of the vertebrae has been identifi ed to species level,

the dermal denticles are mostly from the spiky armoured skin

of the aptly-named porcupine ray (Urogymnus africanus, often

Figure 7 Vertical changes in vertebrate taxa, Squares A, B and C, Tigershark Rockshelter.



identifi ed incorrectly as U. asperrimus – Randall et al. 1997:30).

The skin exhibits hundreds and often thousands of bony

dermal denticles. Across western Zenadh Kes the porcupine

ray – referred to as tapin(u)/tupamur/tupmul/tupmur – has

totemic signifi cance (Haddon 1904:154-55; Ray 1907:123)

and ‘when fat’ is a ‘much esteemed’ food in western Zenadh

Kes (Johannes and MacFarlane 1991:218). Porcupine rays can

provide considerable food as they grow up to 3m in length

with a head-body (disc) diameter of over 1m (Last and Stevens

1994:414; Randall et al. 1997:30). In central western Zenadh

Kes, tupmul are typically speared in shallow water adjacent

to the beach where the species seasonally congregates during

the wet season, especially December and January (David and

Weisler 2006:31). It is one of 13 species of rays ethnographically

documented as food items for Torres Strait Islanders (McNiven

and Hitchcock 2004:145-146). Previous archaeological evidence

for rays in Zenadh Kes include ‘Elasmobranchii (sharks and

rays)’ and ‘Myliobatidiformes (rays)’ from Mask Cave in levels

dated 1500–2600 years ago (McNiven et al. 2006:61), ‘dorsal

denticles’ from ‘rays’ from midden sites 04 and 02 on Mua

(Harris et al. 1985:16), ‘Elasmobranchii (sharks and rays)’ and

Urogymnus asperrimus dermal denticles from Kurturniaiwak

(David and Weisler 2006), and ‘dasyatids, or rays’ from Goemu

(Ghaleb 1998).

Terrestrial Vertebrates

BirdSeventy-one bird bones were identified from Squares A

(XUs 1–6, 8–10), B (XU3) and C (XUs 3, 5, 6–7) by IM and

sent to WB for more detailed description. Following the

general trend in bone deposition at the site, most bird bones

(n=61, 86%) were recovered from XUs 1–7 in levels dating

to between 500 and 700 years ago. The majority of bones

(80%) are miscellaneous limb fragments. Of the remaining

identifiable bone elements, most are vertebrae (n=5), lower

leg bones (tibiotarsus and tarsometatarsus) (n=5) and wing

bones (ulna and phalanx) (n=2). As such, 92% of bird bones

are limb elements. For the six bones where aging was possible,

five are from juveniles (as indicated by porous bone surface

and lack of complete ossification). The only bird bones where

more detailed taxonomic information was possible were a

distal tibiotarsus (XU3, Square B) from a small to medium-

size bird, a distal tarsometatarsus and distal tibiotarsus

(XU3, Square A) belonging to a medium-sized bird, and a

tarsometatarsus (lower leg) bone (XU2, Square A) from a

small heron (Ardeidae), closest to striated heron (Butorides

striatus). The striated heron, known locally as gaur (Boigu)

and gauti (Badu), is ‘eaten very occasionally’ by contemporary

Torres Strait Islanders (Johannes and MacFarlane 1991:213).

It is a ‘resident’ of western Zenadh Kes and frequents exposed

reefs and mudflats during the day when feeding (Draffan et

al. 1983:212). Striated herons are one of at least 20 birds eaten

across the region (McNiven and Hitchcock 2004:107, 144)

and the Tigershark Rockshelter remains suggest it has been on

the menu for at least 500–600 years. Previous archaeological

insights into Torres Strait Islander bird use are limited to a

limb fragment from Goemu on Mabuyag (Ghaleb 1990:306)

and 52 bird bones excavated from three sites in the Murray

Islands in the eastern Strait (Carter 2004:489, 493, 495). While

none of the Murray Islands bones has been identified beyond

‘bird’, the remains suggest use of larger birds by eastern Torres

Strait Islanders over the past 2600 years.

Lizard/SnakeA small amount (1.9g) of skink/lizard/snake bone (represented

by 21 vertebrae) was found in levels down to XU9, Square A

dating to between 500 and 950 years ago. Finer taxonomic

identifi cations were undertaken by SH. Skink (Scincidae) (n=10)

occurs in Square A (XUs 1, 2, 4, 6, 8 and 9) and Square B (XU2).

These small reptiles possibly died at the site naturally. Monitor

lizards (Varanus sp.) (n=10) occur in Square A (XU6), Square

Taxon Common Name

Excavation Unit (XU)1 2 3 4 5 6 7 8 9 10 11 Total

Elasmobranchii (vertebra)

Sharks/rays 5 2 9 4 7 3 12 1 8 7 0 58

Selachii (teeth) Shark 1 0 2 0 1 0 2 0 4 0 0 10

Elasmobranchii (dermal denticles)

Rays 0 0 1 0 0 0 0 1 0 0 1 3

Dasyatidae U. africanus (dermal denticles)

Porcupine Ray

7 10 12 13 12 5 4 4 4 0 1 72

Serranidae Groupers 2 1 2 0 0 1 2 1 0 0 0 9

Carangidae Trevallies 1 0 0 0 0 0 1 0 0 0 0 2

Lutjanidae Snappers 0 0 0 0 0 0 1 1 0 0 0 2

Lethrinidae Emperors 0 0 1 1 3 1 2 1 1 4 0 14

Labridae Wrasses 2 4 3 0 3 3 6 3 14 10 7 55

Scaridae Parrotfi sh 1 0 1 0 2 0 1 0 1 0 0 6

Tetraodontidae Puffers 0 2 1 0 0 0 1 0 2 0 0 6

Diodontidae Porcupinefi sh 0 0 0 0 0 1 1 0 0 0 0 2

Total 19 19 32 18 28 14 33 12 34 21 9 239

Table 4 Fish (including sharks/rays) NISP (number of identifi ed specimens), Square A, Tigershark Rockshelter.



B (XU3) and Square C (XUs 4, 5 and 7) and range in size from

small (juvenile) to adults c.1.2m in length. The single snake

(XU1, Square B) is either a tree (colubrid) or venomous (elapid)

snake (i.e. not a python) probably <1m in length. The monitor

lizards and snake probably entered the site as dietary items. In

this connection, ethnographic evidence for terrestrial reptile

consumption across Zenadh Kes concerns goannas (e.g. Bliege

Bird et al. 1995:5; Haddon 1912:138) and snakes (e.g. Haddon

1912:138; 1935:175, 303). Prior to the excavation of Tigershark

Rockshelter, archaeological evidence for terrestrial reptile

consumption by western Torres Strait Islanders was limited to

a ‘few’ bones of ‘reptile’ from Goemu (Ghaleb 1998), 0.05g of

snake and lizard bone from Berberass (Crouch et al. 2007), and

a mangrove monitor (Varanus indicus) vertebrae from Long

Beach on Mua (Rowland 1985:129). More extensive collections

including goannas, small lizards and small snakes dating to the

past 2600 years were recovered from the Murray Islands by Carter

(2004:Appendices F-I).

RodentsFive rodent bones (0.5g) were recovered – vertebra (n=1),

maxillary fragment with molar (n=1) and incisors (n=3).

As the bones were restricted to the top four XUs they date

to between 500 and 700 years ago. All bones are ascribed

to Melomys cf. capensis (identifications by Steve Van Dyck,

Senior Curator of Vertebrates, Queensland Museum). Cape

York melomys (Melomys capensis) has never been recorded for

Zenadh Kes islands (McNiven and Hitchcock 2004:Appendix

1); however, its presence in the rockshelter deposits is not

unexpected given that it occurs ‘on Cape York Peninsula and

some offshore islands’ (Leung 1995:634). This small, semi-

arboreal rodent grows to about 14cm (head-body length) and

its ‘preferred habitat is rainforest with a dense understorey

of saplings and vines’ (Leung 1995:634-635), elements of

which occur in the vicinity of Tigershark Rockshelter. While

Melomys capensis, like any species of rodent in Zenadh Kes,

is a potential food source, it is more likely that the rare

occurrence of the species at Tigershark Rockshelter reflects

occasional natural deaths. That rodents were rarely, if ever,

eaten across western Zenadh Kes is revealed by their absence

in ethnographic recordings of subsistence. Archaeological

evidence is equally scant: a ‘few’ bones of ‘rodent’ and rodent

scats at Goemu (Ghaleb 1990, 1998), 0.01g of rodent bone

from Berberass (Crouch et al. 2007), and 0.5g of rodent

scats (but no bones) from Kurturniaiwak (David and Weisler

2006). Carter (2004:490, 493, 496, 498) recovered 66 ‘rodent’

bones dating back to 2600 years ago from three sites in the

Murray Islands. Most were identified as Rattus rattus and

Carter (2004:326) implies that their presence was natural and

unrelated to subsistence.

Material Culture

Dugong Bone ArtefactsTwo conjoinable pieces of dugong rib with evidence of

deliberate shaping were recovered from XU10, Square A dated

from 950 to 1250 years ago (Figure 8). The implement has a

combined weight of 9.2g with a maximum length of 59mm.

The fracture surface has a convex conchoidal topography

consistent with longitudinal splitting of a rib by application

of percussion force, possibly while resting on an anvil. This

fracture surface has then been chopped into on the margins

with a sharp tool (probably a stone artefact) towards one end

to create two distinct notches. Tiny parallel striations running

perpendicular to the long axis of the implement are associated

with one of the notches. The pair of notches suggests that this

artefact was hafted and used as a tool. As the tapering (intact)

end of the implement exhibits no obvious signs of use, it is

likely that the ‘use end’ of the implement has snapped off and

is no longer present. What the intact implement was used

for remains a mystery as it is a unique artefact for the region.

Ethnographic references to hafted bone implements from

Zenadh Kes are rare. In southwest Zenadh Kes, MacGillivray

(1852, II:18, 24; cf. Haddon 1890:351) recorded spears tipped

with the ‘leg-bone of a kangaroo six inches long, sharpened

at each end’ and a dugong harpoon point made ‘of bone, four

inches long barbed all round’. Despite these observations,

all nineteenth century ethnographic Zenadh Kes harpoon

points collected by Haddon are carved from wood (Haddon

1912:169-170; Moore 1984:43, Plate 3). The only other

known archaeological example of a Zenadh Kes ‘bone point’

(unknown form and function) comes from midden Site 02

or 04 on Mua (age unknown) and is similarly made from

‘dugong rib’ (Harris et al. 1985:18).

Wg T W R B S L G H I TotalXU A B C D A B C D B C C B B C C C C B Teeth Taxa1 0 1 1 1 1 1 0 1 3 0 0 0 1 0 0 0 1 0 11 5

2 0 2 2 0 0 1 0 0 1 1 0 0 0 0 0 1 0 0 8 4

3 0 2 1 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 5 2

4 0 2 0 0 0 0 0 0 1 0 1 1 0 0 0 0 1 0 6 5

5 1 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 1

6 0 0 5 0 0 0 4 0 0 1 0 1 0 0 1 0 0 0 12 5

7 1 0 1 0 1 0 0 0 0 0 0 0 0 0 1 0 0 1 5 4

8 0 1 1 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 4 3

9 1 0 1 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 5 2

10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Table 5 Number of shark teeth, Squares, A, B, C and D, Tigershark Rockshelter. Wg=Wobbegong; T=Tiger Shark; W=Whaler; R=Reef; B=Blacktip; S=Silky; L=Lemon; G=Grey Nurse; H=Hammerhead; I=Indeterminate.



Shell ArtefactsSeven shell artefacts were recovered:

A. A U-shaped piece of pearl shell with edges smoothed by

abrasion (Figure 9a). Both faces expose fresh nacre indicating

that the original outer weathered surface of the shell has

been removed, probably through abrasion/grinding. This

105mm-long item is likely to be a broken pearl shell chest

pendant – a distinctive male status body adornment recorded

ethnographically for the region (Edmundson and Boylan

1999:25; Florek 2005:80, 94; Haddon 1912:43, 204, 1935:293,

399, Plate IX). However, the Tigershark Rockshelter pendant

differs slightly in form to ethnographically-known crescentic

pearl shell pendants (mai). It was collected from the surface

(Surface Collected Item #2) and is dated ≤500 years ago.

B. A 60mm-long piece of pearl shell with two margins

shaped by fl aking and a slightly curved margin with linear

furrows revealing that it was cut from both sides using

a sharp tool in a sawing action (Figure 9b). As with the U-

shaped pendant, both faces expose fresh nacre indicating

that the original outer weathered surface of the shell has

been removed, probably through abrasion/grinding. This

item may be an incompletely-made adornment, an off-

cut from the manufacture of a large pearl shell pendant,

or perhaps a form of cutting/scraping tool as recorded

ethnographically for the region (Haddon 1912:65). It was

recovered from XU3, Square A and dates to 500–700 years

ago. Both pearl shell items are made from large pearl shells

– either black lipped pearl shell (Pinctada margaritifera) which

grows to 17cm or gold lipped pearl shell (Pinctada maxima)

which grows to 28cm. Pearl shell artefacts, either tools or

adornments, are rare archaeological fi nds for Zenadh Kes

with previous examples restricted to Goemu on Mabuyag

(Ghaleb 1990:196-208; see also Barham et al. 2004:30) and

Tudu (McNiven et al. 2004b).

C. A 71mm-long, subrectangular-shaped section of large marine

gastropod, probably a baler shell (Melo sp.) (species known

locally as alup) (Figure 9c). Designated Surface Collected Item

#4 it is dated ≤500 years ago. The longitudinal margins are

shaped by subtle fl aking (unifacial and bifacial) truncated by

smoothing using abrasion action. The broader end exhibits

remains of a c.7mm-diameter circular (straight-sided) hole.

The form of the item is consistent with a broken pendant.

It is likely the hole was made with a stone awl or drill (cf.

Haddon 1912:128). Ethnographically, records of baler shell

body items in Zenadh Kes refer to pubic covers, not pendants

(e.g. Florek 2005:83). However, archaeological evidence of

rectangular sections of baler shell with drilled holes has been

found elsewhere in Zenadh Kes (e.g. Tudu – McNiven et al.

2004b).

D. A 41mm-long fragment of marine gastropod shell (baler?)

with two sets of parallel lines (straight and curved) engraved

across the outer convex surface (Figure 9d). The shell was

recovered from XU8, Square A, between dates of 700 and 950

years ago. Shell engraving is ethnographically well-known for

Zenadh Kes (Florek 2005:94; Haddon 1912:43-47).

E. The right valve of an 86mm-long Polymesoda erosa (species

known locally as akul) with a 38mm-long section of unifacial

retouch onto the outer surface (Figure 9e). Akul valves were

recorded ethnographically as cutting and scraping tools in

Zenadh Kes (e.g. Haddon 1904:262, 1912:89, 124; Moore

1984:43). Florek (2005:92) refers to one such tool (‘knife’)

with minor fl aking (‘scars’) on the edge. As such, it is plausible

the Tigershark Rockshelter akul was used also as a cutting

tool given the edge modifi cation. Collected from the surface

it could have been used anytime over the past 500 years.

F. and G. Two olive shells (Oliva sp.) – 24mm (XU4, Square

A) and 14mm (XU4, Square C) – both dating to 500–700

years ago. The shells are whole except for removal of the apex

(protoconch) and an underlying internal section of columella

such that a straight internal passage has been created from

one end of the shell to the other (i.e. from the apex to the

anterior canal) (Figure 9f and 9g). The form and modifi cation

of these shells are consistent with ethnographically-known

use of olive shells (uraz/waraz) in Zenadh Kes as beads for

body adornments in the form of necklaces and headdresses

(Florek 2005:83; Haddon 1912:41, Plate IX, 1935:183, 198,

295; Moore 1984:69, 100). It is also possible the shells had a

similar prized status – a ‘good’ necklace of waraz ‘was equal

to the highest unit of value’ – ‘a canoe’ (Haddon 1912:41).

All of the shell species used to make shell artefacts recovered

from Tigershark Rockshelter occur on shallow reefs and sand

banks (pearl, baler, olive) and mangrove environments (akul).

While potential source areas are located close (within 1km) to

the site, collection from offshore reefs via canoe should not be

discounted. Haddon (1912:3-4) recorded that ‘men sometimes

went considerable distances ... to hunt on distant reefs at low

spring tides for shells to be employed in domestic use or for the

making of ornaments’. While some shell artefact manufacture

may have taken place at Tigershark Rockshelter (see below), it is

also likely that fi nished items entered the site.

Shark Tooth ArtefactA large tiger shark tooth recovered from XU1, Square B and dated

500 years old has considerable wear that has truncated the tops

of serrations along one margin (Figure 10). As with most sharks,

tiger sharks rarely accumulate tooth wear as they have a high

turnover of teeth. While far from demonstrated, we hypothesise

that the wear on the archaeological tooth is consistent with its

use as a tool to perform cutting/scraping functions. Haddon

(1912:43) notes that shark teeth were used to inscribe designs

on pearl shell pendants across western Zenadh Kes. As such,

it is quite possible that pearl shell artefacts recovered from

Tigershark Rockshelter were cut into shape using shark tooth

tools. Furthermore, it is worth hypothesising that holes in shell

pendants may have been formed using shark tooth awls/drills.

This hypothesis follows Haddon’s (1912:128) comment that ‘I

was informed in Mabuiag that holes were pierced in pearl-shells

by means of sharks’ teeth. The anterior long, slender, simple teeth

of Crossorhinus [wobbegongs], im (W.), were used as drills when

inserted in pieces of wood’.

Stone ArtefactsAll stone artefacts recovered from Tigershark Rockshelter are

fl aked with no evidence of grinding. Quantitative insights into

stone artefact technology are available only for Square A (n=638,

88.0g). Numerically, most (94%) stone artefacts are made from



milky quartz (n=359, 62.9g) and crystal quartz (n=238, 15.1g)

(with crystal quartz defi ned as quartz where more than 50%

of the artefact area is transparent and milky quartz defi ned

as quartz where more than 50% of the artefact area is opaque

and/or white). The remaining artefacts were made from igneous

rocks: rhyolitic ignimbrite (n=16, 2.4g), granite (n=4, 5.3g),

volcanic (n=2, 0.8g) and igneous undifferentiated (n=19, 1.7g).

All of these raw materials are consistent with the known local

geology of the Mabuyag area (McNiven et al. 2006:63). The size

of artefacts in Square A is small, as indicated by the maximum

length for each raw material: milky quartz (28mm), crystal

quartz (31mm), rhyolitic ignimbrite (21mm), granite (39mm),

volcanic (14mm) and igneous undifferentiated (18mm). Many

quartz artefacts reveal evidence of bipolar (anvil) reduction. The

dominance of a quartz bipolar technology is typical of stone

artefact assemblages of western Zenadh Kes for the past 1700

years (e.g. David and Weisler 2006; McNiven 2006; McNiven et

al. 2006). While no major chronological changes were recorded

in raw material preference, the discard rate of artefacts doubles

through time from 65.3 artefacts/100 years in lower levels (XUs

8–11) dated >700–1300 years ago to 123 artefacts/100 years in

upper levels (XUs 1–7) dated 500–700 years ago.

Bottle Glass ArtefactsThe only glass items seen at the site were the fi ve fragments of

green bottle glass recovered from XU1, Square D: #1 (burnt wall

fragment, 0.9g), #2 (fi nish fragment, 7.6g), #3 (retouched fi nish

fragment, 5.4g), #4 (fl ake, 0.6g) and #5 (retouched wall fragment,

0.8g). Artefacts #1 to #4 were examined by Tom Loy (School of

Social Science, University of Queensland) for residues. The only

residues observed were shell which has ambiguous functional

signifi cance given the high potential for contamination from

the surrounding midden matrix. Finish fragments (#2 and #3)

conjoin and reveal an applied single collar or ring seal of a form

in ‘common use until c1900-1920’ (Burke and Smith 2004:364).

As such, the bottle fragments date to the late nineteenth and/

or early twentieth century given Mabuyag, like most parts of

Zenadh Kes, only came into sustained contact with Europeans

after 1870.

Rock Art and Ochre

PaintingsA total of 16 heavily deteriorated paintings (Pictures #1 to

#16) – all executed with red ochre – were recorded on a

single panel at the site in 2003 (see Brady 2005 for details).

The paintings were systematically documented using digital

photography, and given the deteriorated nature of most

paintings, computer enhancement techniques were applied to

each painting to retrieve design elements no longer visible to

the naked eye (see Brady 2006; David et al. 2001 for details of

computer enhancement techniques used in Zenadh Kes). The

12 identifiable paintings are all non-figurative motifs, and

include enclosed geometrics (#10: two vertical lines enclosed

by a circle; #12: infilled circle inside a small depression in

the granite), open geometrics (#4: one ]-shape; #15 and #16:

two curved lines; #2 and #11: two straight lines; #14: one V-

shape); linear non-figurative (#1 and #6: one L-shape; #7:

one rayed line variant consisting of curved lines emanating

from a central point; and dots (#3: a single dot). The Non-

Figurative imagery includes common, widespread motifs (e.g.

straight lines, circle variants etc) found at other sites across

western Zenadh Kes. A single distinctive design form (Picture

#7) is similar to a painting from Muralag (Brady 2006:368).

Executing an image in a shallow depression on the granite

surface (Picture #12) has only been documented at this site and

at Dauan where a single face/mask is also painted in a shallow

depression on a granite boulder. Additional comparisons

between rock paintings from the site and decorated material

culture objects collected from the region failed to identify any

further artistic links. Overall, rock paintings at Tigershark

Rockshelter conform to Western and Central Zenadh Kes rock

art. However the limited number of rare or distinctive motifs

restricts further attempts to gain information regarding

specific interisland and island-mainland links.

Ochre and Rock Art AntiquityOf the 103.7g of red/orange/yellow ochre recovered throughout

the deposit, most are tiny fragments with only four larger pieces

exhibiting ground facets. While ochre could have been used as

a pigment for any number of painting activities (e.g. artefact

decoration, body adornment), it is likely that much of it relates

to rock art production across the roof of the site. Vertical

changes in the discard rate of ochre were used as a measure

of the history of intensity of painting activity at the site. The

combined discard rate of ochre for XUs 1–7 dating between

500 and 700 years ago in Squares A to C is 28.8g/100 years and

nearly four times the combined rate of 7.7g/100 years for XUs

8–10/11 for the period >700–1300 years ago. As most ochre

and hence painting activity at the site took place 500–700 years

ago, it is likely that most rock art was similarly produced during

this period of intensive cultural activity. It is unlikely paintings

were executed within the past 500 years after occupation and

cultural discard at the site essentially ceased as such a situation

Figure 8 Flaked dugong bone implement, Square A, Tigershark Rockshelter.



would create a surface concentration of ochre for which no

evidence exists.

DiscussionIn what historical context did Tigershark Rockshelter emerge and

then cease to be a settlement place on Pulu within the broader

context of Mabuyag land- and seascapes?

Site Establishment 1300 Years AgoMask Cave, located 150m north of Tigershark Rockshelter,

demonstrates that people have been visiting Pulu for 4000

years (McNiven et al. 2006). Thus for over 2500 years people

visiting Pulu simply walked past the boulder outcrops forming

Tigershark Rockshelter. If these early visitors to Pulu used the

site they never left any traces of activity. However, 1300–1500

years ago things changed on Pulu. Use of Mask Cave essentially

ceased 1500 years ago while 1300 years ago Tigershark

Rockshelter was established. Signifi cantly, preliminary fi ndings

from excavations at the kod ceremonial complex located 500m

away at the northern end of Pulu reveal a large underlying

midden (village) which began forming 1300–1400 years ago

(McNiven et al. 2007). This succession of changes 1300–1500

years ago signals a fundamental change in the use of Pulu and the

place of the islet within the world of the Mabuyag community

and suggests a causal connection or at the very least responses

to similar cultural processes. While reasons behind these

settlement changes remain poorly understood, differing site

contexts may provide clues. Mask Cave is hidden away near the

centre of Pulu with restricted views across the narrow channel

to Mabuyag. In contrast, Tigershark Rockshelter and the village

midden are located adjacent to the southern and northern

shores of the islet respectively with panoramic views of the sea

to the horizon. Such topographical differences – the concealed

nature of Mask Cave versus the exposed nature of Tigershark

Rockshelter and the village midden – suggest changes in

settlement pattern across Pulu related to changes in site aspect

(and by extension the functional role and place of settlement

sites). That is, 1300–1500 years ago saw the introduction of a new

way people engaged with seascapes which required settlement

places to be located increasingly, if not entirely, in prominent

locations adjacent to the shoreline. This phenomenological

shift meant that people at these new shoreline settlements not

only had good views of the sea but people at sea in canoes also

had good views of these settlements. This new intervisibility

and intimacy between land and sea geographically and

conceptually positioned shoreline settlements with enhanced

liminality between land- and seascapes, perhaps signalling

the emergence of more distinctive island communities and

associated territorial marine domains. In this connection, Davis

and Prescott (1992:126) note that for the Goemulgal (Mabuyag

community), the ‘critical determinant’ of the scale of marine

territories is the extent to which one can travel out to sea and

‘maintain visual contact’ with Mabuyag. Similarly in eastern

Zenadh Kes, Haddon (1908:167) recorded that marine tenure

was a ‘spatial projection’ of terrestrial tenure out over adjacent

reefs and waters such that ‘the inhabitants of certain areas

Figure 9 Shell artefacts, Tigershark Rockshelter.



appear to have a pre-emptial right to certain distant fi shing

stations which lie off their part of the coast’. Indeed, ‘visual

surveillance’ of sea territories from land territories remains

‘central to the regulation of [marine] territorial rights and

resources’ (Scott and Mulrennan 1999:155).

Whatever the reasons for settlement rearrangements on Pulu,

an associated issue is likely to be changing demography. The

period 1300–1500 years ago saw accelerated site development

across other parts of western Zenadh Kes (Barham 2000:275;

Barham et al. 2004:37). For example, Ash and David (in press)

reveal that ‘c.1300-1400 years ago people established’ the

historically-signifi cant village site of Totalai on the north coast

of Mua. The period immediately before 1200 years ago saw

the onset of agricultural mound-and-ditch systems on Saibai

in northern Zenadh Kes (Barham 1999). These agricultural

developments imply an increasingly formalised and intimate

relationship with terrestrial resources and landscapes. That

systemically-linked settlement changes were unfolding across

eastern Zenadh Kes is suggested by a synchronous drop to

negligible rates of cultural deposition (shell and marine bone)

at Sokoli on Dauar within the past c.1400 years and a possible

cessation of occupational deposition at Ormi on Dauar after

c.1600 years ago (Carter 2004:191; Carter et al. 2004a:252).

Concomitantly, the earliest midden (shell) date for the adjacent

island of Mer is 1623±38 BP (Wk-13369) (c.1300 years ago) at

Kurkur Weid Rockshelter (Carter 2004). To what extent these

changes were shaped by reconfi gurations of resources (marine

and terrestrial) is unknown due largely to a lack of independent

palaeoenvironmental data. Whatever the situation, repositioning

of settlements in eastern and western Zenadh Kes 1300–1500

years ago would have resulted in reconfi gurations of social space

associated with new settlement patterns and group dynamics

both within and between islands.

Site Expansion 500–700 Years AgoMcNiven (2006) elaborates a chronological model for Zenadh

Kes that posits major cultural changes unfolding over the past

600–800 years linked to the development of ethnographically-

known social arrangements and ritual sites in the context of

accelerated midden development and population increase. It

was noted further that middens associated with the 600–800

year event took one of three forms – middens spanning the past

600–800 years, middens representing a single intense period of

activity between 600 and 800 years ago, and older middens

with increased use within the past 600–800 years (McNiven

2006:9). David and Weisler (2006:27, 31) documented a

variant of the fi rst and second forms at Kurturniaiwak on

Badu with a ‘burst of village activity’ sometime between 500

and 700 years ago followed by a lull and then a secondary, less

intense activity burst around 350–400 years ago. Tigershark

Rockshelter reveals further complexity with a variant of the

second and third forms – a major burst of activity between 500

and 700 years ago after which occupation of the site essentially

ceased. This activity burst has also been registered on Dauan

in the northern Strait with the lower half of deposit at Dauan

4 (XUs 10–20) accumulating rapidly over c.100 years between

500 and 600 years ago (McNiven 2006). On Mer in the eastern

Strait, Carter (2004:154, 158, 192, 257) recorded an occupation

peak in basal levels of Pitkik dated c.700–800 years ago while

Kurkur Weid Rockshelter (also on Mer) was ‘most intensively

used between 520 and 780 cal BP’. It is now clear that the 600–

800 year ‘event’ was often associated with an initial 200–300

year activity burst between 500 and 800 years ago. This activity

burst was associated with new settlement arrangements and a

broad range of more fundamental social, ritual and territorial

transformations (David and Weisler 2006; David and McNiven

2005; McNiven 2006). To what extent this activity burst was

structured also by environmental changes such as sea level

fall and new marine resource confi gurations linked to the

Little Ice Age, as hypothesised previously for Torres Strait

(McNiven 2006) and other parts of the Queensland coast and

the western Pacifi c (Nunn 2007; Nunn et al. 2007), is a matter

for future investigation.

Site Abandonment 500 Years AgoWhy Tigershark Rockshelter was closed down 500 years ago as

a regularly-used place in local Mabuyag settlement-subsistence

arrangements is the subject of ongoing research. However,

excavations of large village midden sites at Goemu on Mabuyag

(Ghaleb 1990, 1998; McNiven and Wright in press) and

underlying the kod ceremonial site at the northern end of Pulu

(McNiven et al. 2007) reveal synchronous increases in activity

and deposition commencing around 500 years ago. Thus, small

sites such as Tigershark Rockshelter may have become obsolete

as Mabuyag settlement arrangements 500 years ago required

larger more communal settlements (i.e. open villages) to

accommodate numerous people for extended periods of time.

Such demographic reconfi gurations, while consistent with

increased sedentism and perhaps population increase, must have

had socio-political dimensions.

Ritual Site Complexes of the Past 400–500 YearsMcNiven (2006:10) makes the point that cultural transformations

(along a secular-spiritual continuum) commencing 600–800

years ago ‘did not remain historically static but underwent

periods of strategic readjustment, most notably within the last

300-400 years’ with major elaborations of ritual sites such as

dugong bone mounds and bu shell arrangement in western

Zenadh Kes (cf. David and Mura Badulgal Committee 2006;

David et al. 2005; McNiven 2006; McNiven and Feldman 2003).

However, a new ∆R value for western Zenadh Kes of −32±20

(Ulm et al. 2007) has resulted in radiocarbon dates on local

marine shell/bone calibrating up to c.100 years older than

previous calibrations based on a ∆R value of 49±45 (Ulm 2002)

Figure 10 Tiger shark tooth tool, XU1 Square B, Tigershark Rockshelter.



or 50±47 (Ulm 2006). As such, ritual site transformations of

300−400 years ago associated with construction of dugong bone

mounds can now be extended back to around 400−500 years ago

(e.g. compare McNiven and Feldman 2003 with McNiven and

Bedingfi eld 2008). Certainly, David et al.’s (2005:88) conclusion

for ‘the apparent beginnings of bu shell arrangements 500-400

years ago’ remains pertinent. As dates for major elaborations of

ritual sites (dugong bone mounds and bu shell arrangements)

are pushed back to 400−500 years ago, causal relationships with

suggested demographic expansions registered at midden sites

500 years ago may well be implicated. That is, demographic

changes across western Zenadh Kes around 500 years ago may

have involved new social arrangements (e.g. clan, totemic

and territorial) that within a few generations were managed

by new communal socio-ceremonial structures incorporating

ritual installations in the form of dugong bone mounds and bu

shell arrangements (see McNiven et al. 2007 for details). These

communal socio-ceremonial structures and associated ritual

practices were a key to maintaining social cohesion and broader

shared identities in a society that ethnographically was largely

egalitarian and ‘lacked centralised authority’ (Beckett 1972:320)

as hereditary clan chiefs had no ‘marked social or political

status’ (Haddon 1904:265; cf. Lilley 1985). It is in this context

that the Goemulgal (people of Mabuyag) chose to establish the

ceremonial kod site with dugong bone mounds and bu shell

arrangements at the northern end of Pulu around 300–400

years ago (see McNiven et al. 2007), while the possible kod site

with dugong bone mounds was established on Koey Ngurtai

islet located between Mabuyag and Badu around 400–500

years ago (David and McNiven 2005). As the kod on Pulu was

located on top of a large midden that simultaneously registers

a dramatic decrease in use around 300–400 years ago (McNiven

et al. 2007), kod establishment was probably associated with the

transformation of the islet from a place of regular residential

occupation to a restricted, formalised and highly controlled

sacred landscape, as it remains today.

ConclusionTigershark Rockshelter demonstrates that the ancestors of

the Goemulgal were marine specialists just as they are today

and just as Alfred Haddon found when he undertook detailed

anthropological work on Mabuyag in 1888 and 1898 (Haddon

1901, 1904, 1935). In this sense, excavations cement an

archaeological and ethnographic pattern of marine subsistence

specialisation seen across Zenadh Kes for at least the past 4000

years (e.g. Carter et al. 2004b; Crouch et al. 2007; David and

Weisler 2006; McNiven and Hitchcock 2004; McNiven et al.

2006). Signifi cantly, Tigershark Rockshelter extends insights into

ancient fi shing practices and reveals the most diverse shark and

ray assemblages yet recovered from an Australian Indigenous

midden. While ochre at previously-excavated middens across

western Zenadh Kes indicates that people undertook a range of

decorative activities, Tigershark Rockshelter adds an embodied

dimension to such artistic pursuits with engraved shells, shell

pendants and shell beads. Using ethnographic observations as

a guide, these multifunctional shell adornments communicate

to us across time the social realities of status, prestige, gender,

aesthetics and exchange relationships. In short, the small

groups of men and women (and possibly children) camping at

Tigershark Rockshelter from time to time between 500 and 1300

years ago had broader social connections and belonged to a larger

community, the most obvious being the Mabuyag community.

The excavations also reveal that Mabuyag/Pulu society is the

result of a series of dynamic social transformations over the past

millennium. And after 500 years of site abandonment, during

which time use of Pulu focused on the ceremonial and the

sacred, archaeological research has provided an opportunity for

members of the Mabuyag community to reacquaint themselves

with Tigershark Rockshelter and to experience some familiar

and not so familiar aspects of their dynamic past.

AcknowledgementsKaima esso to the people of Mabuyag for their hospitality and

for providing the special opportunity to visit and research the

history of Pulu and for logistical support that made excavations at

Tigershark Rockshelter possible. Thanks to John Mooka, Terrence

Whap, John Bani, Harold Bani, Martha Paipai, Bessie Whap and

Tyson Gizu from Mabuyag for assistance with excavations. For

support and guidance we thank members of the Goemulgau Kod

cultural organisation, in particular Cygnet Repu, Terrence Whap,

Sophie Luffman, Holly Banasa, and Brian Whap. Special thanks

to Judith Fitzpatrick for assistance with excavations and Bruno

David for assistance with drawing the site cross-section. The

Mabuiag State Primary School kindly helped with sieving and

drying of excavated materials. For general support we thank the

Mabuiag Island Council and then chairman Fr. Bani. Vic McGrath

(Thursday Island) kindly provided information on fi shing. Steve

Van Dyck (Senior Curator of Vertebrates, Queensland Museum)

identifi ed rodent bones. The late Tom Loy (School of Social

Science, University of Queensland) examined glass artefacts for

blood residues. Steve Morton (Photography, Monash University)

and Kara Rasmanis and Gary Swinton (School of Geography and

Environmental Science, Monash University) created the fi gures.

This research was supported in part by ARC Discovery-Projects

Grant DP0344070 and AINSE grant 05/119. Helpful comments

on earlier drafts of this paper were kindly provided by Jeremy

Ash, Bruno David, Fiona Petchey, Mike Rowland, Duncan Wright

and anonymous referees.

ReferencesAllen, G. 1997 Marine Fishes of Tropical Australia and South-East Asia. Perth:

Western Australian Museum.

Ash, J. and B. David in press Mua 22: Archaeology at the old village site of Totalai. In

B. David, D. Tomsana and M. Quinnell (eds), Gelam’s Homeland: Cultural and

Natural History on the Island of Mua, Torres Strait. Memoirs of the Queensland

Museum, Cultural Heritage Series. Brisbane: Queensland Museum.

Barham, A.J. 1999 The local environmental impact of prehistoric populations

on Saibai Island, northern Torres Strait, Australia: Enigmatic evidence from

Holocene swamp lithostratigraphic records. Quaternary International 59:71-

105.

Barham, A.J. 2000 Late Holocene maritime societies in the Torres Strait Islands,

northern Australia – Cultural arrival or cultural emergence? In S. O’Connor

and P. Veth (eds), East of Wallace’s Line: Studies of Past and Present Maritime

Cultures of the Indo-Pacifi c Region, pp. 223-314. Modern Quaternary Research

in Southeast Asia 16. Rotterdam: A.A. Balkema.

Barham, A.J. and D.R. Harris 1987 Archaeological and Palaeoenvironmental

Investigations in Western Torres Strait, Northern Australia. Unpublished report

to the Research and Exploration Committee, National Geographic Society.



Barham, A.J., M.J. Rowland and G. Hitchcock 2004 Torres Strait bepotaim: An

overview of archaeological and ethnoarchaeological investigations and

research. In I.J. McNiven and M. Quinnell (eds), Torres Strait Archaeology

and Material Culture, pp.1-72 Memoirs of the Queensland Museum, Cultural

Heritage Series 3(1). Brisbane: Queensland Museum.

Beckett, J. 1972 The Torres Strait Islanders. In D. Walker (ed.), Bridge and Barrier:

The Natural and Cultural History of Torres Strait, pp.307-326. Canberra:

Australian National University Press.

Bliege Bird, R., D.W. Bird and J.M. Beaton 1995 Children and traditional subsistence

on Mer (Murray Island), Torres Strait. Australian Aboriginal Studies 1:2-17.

Brady, L.M. 2005 Painting Patterns: Torres Strait Region Rock-Art, NE Australia.

Unpublished PhD thesis, Monash University, Clayton.

Brady, L.M. 2006 Documenting and analyzing rock paintings from Torres Strait,

NE Australia, with digital photography and computer image enhancement.

Journal of Field Archaeology 31(4):363-379.


OxCal program. Radiocarbon 37(2):425-430.


OxCal. Radiocarbon 43(2A):355-363.

Burke, H. and C. Smith 2004 The Archaeologist’s Field Handbook. Crows Nest: Allen

& Unwin.

Carter, M.J. 2004 North of the Cape and South of the Fly: The Archaeology of

Settlement and Subsistence on the Murray Islands, Eastern Torres Strait.

Unpublished PhD thesis, School of Anthropology, Archaeology and Sociology,

James Cook University, Townsville.

Carter, M., P. Veth, A. Barham, D. Bird, S. O’Connor and R. Bird 2004a Archaeology

of the Murray Islands, eastern Torres Strait: Implications for a regional

prehistory. In R. Davis (ed.), Woven Histories Dancing Lives: Torres Strait

Islander Identity, Culture and History, pp.234-258. Canberra: Aboriginal

Studies Press.

Carter, M., A.J. Barham, P. Veth, D.W. Bird, S. O’Connor and R.B. Bird. 2004b The

Murray Islands Archaeological Project: Preliminary results of excavations on

Mer and Dauar eastern Torres Strait. In I.J. McNiven and M. Quinnell (eds),

Torres Strait Archaeology and Material Culture, pp.163-182 Memoirs of the

Queensland Museum, Cultural Heritage Series 3(1). Brisbane: Queensland

Museum.

Crouch, J., I.J. McNiven, B. David, C. Rowe and M. Weisler 2007 Berberass: Marine

resource specialisation and environmental change in Torres Strait over the

past 4000 years. Archaeology in Oceania 42(2):49-64.

David, B., J. Crouch and U. Zoppi 2005 Historicizing the spiritual: bu shell

arrangements on the Island of Badu, Torres Strait. Cambridge Archaeological

Journal 15(1):71-91.

David, B. and I.J. McNiven 2005 Archaeological Survey and Salvage Programme

of Koey Ngurtai, Torres Strait (Stage 2): Final Report. Unpublished Cultural

Heritage Report Series 9. Clayton: Programme for Australian Indigenous

Archaeology, School of Geography and Environmental Science, Monash

University.

David, B., I.J. McNiven, J. Brayer and A. Watchman 2001 Why digital enhancement

of rock-art works: Rescaling and saturating colours. Antiquity 75:781-792.

David, B. and Mura Badulgal Committee 2006 What happened in Torres Strait 400

years ago? Ritual transformation in an island seascape. Journal of Island and

Coastal Archaeology 1:123-143.

David, B. and M.I. Weisler 2006 Kurturniaiwak (Badu) and archaeological evidence

of villages in Torres Strait. Australian Archaeology 63:21-34.

Davis, S.L. and J.R.V. Prescott 1992 Aboriginal Frontiers and Boundaries in

Australia. Carlton: Melbourne University Press.

Draffan, R.D.W., S.T. Garnett and G.J. Malone 1983 Birds of Torres Strait: An

annotated list and biogeographical analysis. The Emu 83(4):207-234.

Edmundson, A. and C. Boylan 1999 Adorned: Traditional Jewellery and Body

Decoration from Australia and the Pacifi c. Sydney: University of Sydney.

Fink, D., M. Hotchkis, Q. Hua, G. Jacobsen, A.M. Smith, U. Zoppi, D. Child, C.

Mifsud, H. van der Gaast, A. Williams and M. Williams 2004 The ANTARES

AMS facility at ANSTO. Nuclear Instruments and Methods in Physics

Research Section B: Beam Interactions with Materials and Atoms 223-

224:109-115.

Fitzpatrick, J., J. Cordell and I.J. McNiven 1998 Torres Strait Culture Site

Documentation Project. Unpublished report to the Island Coordinating

Council, Thursday Island.

Fleming, M.A. 1986 The Scaridae Family in Pacifi c Prehistory. Unpublished M.A.

thesis, Department of Anthropology, University of Otago, Dunedin.

Florek, S. 2005 The Torres Strait Islands Collection at the Australian Museum.

Technical Reports of the Australian Museum 19. Sydney: Australian Museum.

Ghaleb, B. 1990 An Ethnoarchaeological Study of Mabuiag Island, Torres Strait,

Northern Australia. Unpublished PhD thesis, Institute of Archaeology,

University College London, London.

Ghaleb, B. 1998 Fish and fi shing on a western Torres Strait Island, northern

Australia: Ethnographic and archaeological perspectives. In A.K.G. Jones

and R. Nicholson (eds), Fish Remains and Humankind: Part Two. Internet

Archaeology 4. http://intarch.ac.uk/journal/issue4/ghaleb_index.html

Grant, E. 1995 Fishes of Australia. Redcliffe, QLD: E.M. Grant P/L Publishers.

Haddon, A.C. 1890 The ethnography of the western tribe of Torres Straits. Journal

of the Anthropological Institute of Great Britain and Ireland 19:297-446.

Haddon, A.C. 1901 Head-Hunters: Black, White and Brown. London: Methuen.

Haddon, A.C. (ed.) 1904 Reports of the Cambridge Anthropological Expedition to

Torres Straits: Sociology, Magic and Religion of the Western Islanders. Vol. 5.

Cambridge: Cambridge University Press.


Torres Straits: Sociology, Magic and Religion of the Eastern Islanders. Vol. 6.



Torres Straits: Arts and Crafts. Vol. 4. Cambridge: Cambridge University Press.

Haddon, A.C. 1935 Reports of the Cambridge Anthropological Expedition

to Torres Straits: General Ethnography. Vol. 1. Cambridge: Cambridge

University Press.

Harris, A., G. Dews, J. Kerr and I. Poiner 1995 Monitoring the Traditional and

Island-Based Catch of the Torres Strait Protected Zone: Report on CSIRO

Research 1991-1993. Brisbane: CSIRO Division of Fisheries.

Harris, D.R., A.J. Barham and B. Ghaleb 1985 Archaeology and Recent

Palaeoenvironmental History of Torres Strait, Northern Australia. Unpublished

report to the Research and Exploration Committee, National Geographic

Society.

Johannes, R.E. and J.W. MacFarlane 1991 Traditional Fishing in the Torres Strait

Islands. Hobart: CSIRO Division of Fisheries.

Kwan, D. 2005 Traditional use in contemporary Ailan (Island) ways: The

management challenge of a sustainable dugong fi shery in Torres Strait. In

N. Kishigami and J.M. Savelle (eds), Indigenous Use and Management of

Marine Resources, pp.281-302. Senri Ethnological Studies 67. Osaka: National

Museum of Ethnology.

Last, P.R. and J.D. Stevens 1994 Sharks and Rays of Australia. Melbourne: CSIRO.

Leung, L.K-P. 1995 Cape York Melomys. In R. Strahan (ed.), The Mammals of

Australia, pp.634-635. Sydney: Reed New Holland.

Lilley, I. 1985 Chiefs without chiefdoms: Comments on prehistoric socio-political

organisation in western Melanesia. Archaeology in Oceania 20(2):60-65.

MacGillivray, J. 1852 Narrative of the Voyage of H.M.S. Rattlesnake. 2 vols. London:

T. & W. Boone.


Reimer 2004 SHCAL04 southern hemisphere calibration, 0-11.0 cal kyr BP.

Radiocarbon 46:1087-1092.



McNiven, I.J. 2006 Dauan 4 and the emergence of ethnographically-known social

arrangements across Torres Strait during the last 600-800 years. Australian

Archaeology 62:1-12.

McNiven, I.J. 2008 Inclusions, exclusions and transitions: Torres Strait Islander

constructed landscapes over the past 4000 years, northeast Australia. The

Holocene 18(3):449-462.

McNiven, I.J. and A.C. Bedingfi eld 2008 Past and present marine mammal hunting

rates and abundances: Dugong (Dugong dugon) evidence from Dabangai Bone

Mound, Torres Strait. Journal of Archaeological Science 35:505-515.

McNiven, I.J., J. Crouch and M. David 2004b Tudu (Warrior Island), Torres

Strait: Historical and Archaeological Survey. Unpublished Cultural

Heritage Report Series, 4. Clayton: Programme for Australian Indigenous


University, Australia.

McNiven, I.J., B. David, Goemulgau Kod and L.M. Brady 2007 The Great Kod of

Pulu: Emergence of a Sacred Totemic Ceremonial Complex in Western Zenadh

Kes (Torres Strait). Manuscript on fi le, Programme for Australian Indigenous


University.

McNiven, I.J., W.R. Dickinson, B. David, M. Weisler, F. von Gnielinski, M. Carter and

U. Zoppi 2006 Mask Cave: Red-slipped pottery and the Australian-Papuan

settlement of Zenadh Kes (Torres Strait). Archaeology in Oceania 41(2):49-81.

McNiven, I.J. and R. Feldman 2003 Ritually orchestrated seascapes: Bone

mounds and dugong hunting magic in Torres Strait, NE Australia. Cambridge

Archaeological Journal 13(2):169-194.

McNiven, I.J., J. Fitzpatrick and J. Cordell 2004a An Islander world: Managing the

archaeological heritage of Torres Strait. In I.J. McNiven and M. Quinnell (eds),

Torres Strait: Archaeology and Material Culture, pp.73-91. Memoirs of the


Museum.

McNiven, I.J. and G. Hitchcock 2004 Torres Strait marine subsistence specialisation

and terrestrial animal translocation. In I.J. McNiven and M. Quinnell (eds),

Torres Strait: Archaeology and Material Culture, pp.105-162. Memoirs of the


Museum.

McNiven, I.J. and D. Wright in press Ritualised marine midden formation in

Western Zenadh Kes (Torres Strait). In G. Clark, F. Leach and S. O’Connor (eds),

Islands of Inquiry: Colonisation, Seafaring and the Archaeology of Maritime

Landscapes. Terra Australis. Canberra: ANU E Press.

Moore, D.R. 1984 The Torres Strait Collections of A.C. Haddon. London: British

Museum Publications Ltd.

Nunn, P.D. 2007 Climate, Environment and Society in the Pacifi c during the Last

Millennium. Developments in Earth and Environmental Science 6. Elsevier:

Amsterdam.

Nunn, P.D., R. Hunter-Anderson, M.T. Carson, F. Thomas, S. Ulm and M.J. Rowland

2007 Times of plenty, times of less: Last-millennium societal disruption in the

Pacifi c basin. Human Ecology 35:385-401.

Randell, J.E., G.R. Allen and R.C. Steene 1997 Fishes of the Great Barrier Reef and

Coral Sea. Bathurst: Crawford House Publishing.

Ray, S.H. 1907 Reports of the Cambridge Anthropological Expedition to Torres

Straits: Linguistics. Vol. 3. Cambridge: Cambridge University Press.

Rowe, C. 2006 Landscapes in Torres Strait prehistory. In B. David, B. Barker, and

I.J. McNiven (eds), The Social Archaeology of Indigenous Societies, pp.270-286.

Canberra: Aboriginal Studies Press.

Rowland, M.J. 1985 Archaeological investigations on Moa and Naghi Islands,

western Torres Strait. Australian Archaeology 21:119-132.

Scott, C. and M. Mulrennan 1999 Land and sea tenure at Erub, Torres Strait:

Property, sovereignty and the adjudication of cultural continuity. Oceania

70:146-176.

Smith, E.A. and R.L. Bliege Bird 2000 Turtle hunting and tombstone opening: Public

generosity as costly signalling. Evolution and Human Behavior 21:245-261.

Stanisic, J. in press Land snails of Mua Island, Torres Strait. In B. David, D. Tomsana

and M. Quinnell (eds), Gelam’s Homeland: Cultural and Natural History on

the Island of Mua, Torres Strait. Memoirs of the Queensland Museum, Cultural

Heritage Series. Brisbane: Queensland Museum.

Stein, J.K. (ed.) 1992 Deciphering a Shell Midden. San Diego: Academic Press.

Telford, R.J., E. Heegaard and H.J.B. Birks 2004 The intercept is a poor estimate of

a calibrated radiocarbon age. The Holocene 14(2):296-298.

Ulm, S. 2002. Calibrating marine radiocarbon dates: A guide to Australian ∆R

values. AACAI Newsletter 89:10-14.

Ulm, S. 2006 Australian marine reservoir effects: A guide to Australian ∆R values.

Australian Archaeology 63:57-60.

Ulm, S., A.J. Barham, B. David, G. Jacobsen, I.J. McNiven, F. Petchey and M. Rowland

2007 Marine carbon reservoir variability in Torres Strait: Preliminary results

of AMS dating of live-collected shell specimens. Quaternary International

167-168:426.

Weisler, M.I. 2001 On the Margins of Sustainability: Prehistoric Settlement of

Utrok Atoll, Northern Marshall Islands. BAR International Series 967. Oxford:

Archeopress.


DATING OF BUSH TURKEY ROCKSHELTER 3 in the Calvert Ranges establishes Early Holocene Occupation of the Little Sandy Desert, Western AustraliaPeter Veth1,3, Jo McDonald2,3 and Beth White2

1 National Centre for Indigenous Studies, Australian National University, Canberra, ACT 0200, Australia

2 Jo McDonald Cultural Heritage Management, 6 Supply Place, Red Hill, ACT 2603, Australia and 77 Justin Street, Lilyfi eld, NSW 2040, Australia

3 Research School of Humanities, Australian National University, Canberra, ACT 0200, Australia

AbstractSystematic excavation of occupied rockshelters that occur in

ranges along the Canning Stock Route of the Western Desert

has seen the establishment of both a Pleistocene signal

(c.24ka BP) as well as the fl eshing out of a Holocene sequence.

Recent dating of a perched rockshelter in the Calvert Ranges,

east of the Durba Hills, has provided a Holocene record fi lling

in previous occupational gaps from the Calvert Ranges. The

extrapolated basal date of the site is in the order of 12,000

BP. Assemblages from this site illustrate repeated occupation

through the Holocene with a notable shift in raw materials

procured for artefact production and their technology of

manufacture in the last 1000 years. Engraved and pigment art

is thought to span the length of occupation of the shelter. The

site illustrates a signifi cant increase in the discard of cultural

materials during the last 800 years, a trend observed at other

desert sites. Much of the pigment art in this shelter seems

likely to date to this most recent period.

IntroductionIn this paper we describe the dating of a cultural assemblage

from Bush Turkey Rockshelter 3 in the Calvert Ranges (Kaalpi),

Western Australia (Figures 1-2). The Calvert Ranges are located

some 80km east of Well 15 on the Canning Stock Route, well

within the linear dunefi elds of the Little Sandy Desert. This is

the second rockshelter excavated and dated in these very remote

ranges of the Western Desert (Veth et al. 2001). Bush Turkey

Rockshelter 3 is signifi cant in that it:

• extends the known occupation of the interior of the Little

Sandy Desert back to the early Holocene;

• has several phases of lithic production illustrating changes in raw

material use and implement types through the Holocene; and

• provides a dated occupation sequence that may be

correlated to different phases of pigment and engraved art

found both within and adjacent to the shelter (McDonald

and Veth in press).

Eight dates were obtained from a test excavation which

reached a maximum depth of 86cm below surface level (bsl).

These dates, combined with evidence for recent Martu occupancy,

reveal repeated occupations from 8202–8414 cal BP through to

the contact period.

Figure 1 Location of Bush Turkey Rockshelter 3 at Kaalpi (Calvert Ranges), Little Sandy Desert, Western Australia.

Figure 2 Bush Turkey Rockshelter 3 at completion of excavation. Engravings are found on the slabs to the left of frame. Pigment art covers the back wall and overhanging ceiling (Photograph: Peter Veth).


Dating of Bush Turkey Rockshelter 3 establishes Early Holocene Occupation of the Little Sandy Desert

The shelter contains sediments and a human occupation

record from the early Holocene. Artefacts were recorded in spits

adjacent to bedrock from below the lowest dated charcoal sample,

meaning that this date is a minimum age for occupation of the

Calvert Ranges and indeed the Little Sandy Desert.

Excavation Context and MethodThe previous absence of mid-to-late Holocene sequences from

shelters located within valleys of the Calvert Ranges may be due

to the scouring out of their deposits during heavy and localised

rain storms. While average annual precipitation for this area is

low (<300mm), rain storms are signifi cant and swollen creeks

have etched into lower elevation shelters and their deposits. Bush

Turkey Rockshelter 3 is a perched shelter and not subject to these

geomorphic processes.

The shelter is 8m above and some 30m distant from a

creekline to the west that had slow-fl owing water in it from the

Kaalpi spring at the time of excavation in July 2005. The mouth

of the shelter faces west.

Excavation at the Bush Turkey Rockshelter 3 site involved two

adjacent 50cm x 50cm test pits (BTA and BTB) (Figure 3). Both

squares were initially excavated to a depth of c.50cm (spit 14) at

which time the intrusion of roof fall made further excavation in

BTA diffi cult. Excavation continued in square BTB (spits 15-24)

as a column measuring 50cm x 30cm.

All excavated deposit was wet-sieved through 1mm mesh with

initial sorting carried out in the fi eld. Volumetric measurements

were made by using spit depths and all non-cultural material

larger than 5cm diameter was separately measured for volume

and weight. Below the scuff zone, faunal and fl oral remains were

negligible. At the time of excavation the deposit was quite moist

due to recent rains; faults in the metamorphosed sandstone tend

to seep waters for months after rainfall. The extreme wet and dry

cycles in these range uplands would be highly aggressive to the

preservation of macro-fl oral and faunal remains and charcoal.

Stratigraphy and DatingBedrock of the shelter is metamorphosed sandstone with

pebble-sized quartzite clasts. Large roof fall boulders have

entrapped sediments from several sources including via a chute

in the northwest of the chamber, in situ weathering of the

metamorphosed sandstone and aeolian sands. Four stratigraphic

layers are identifi ed (Figure 4):

Layer I – Compacted red-brown sediment with brown/grey

laminae. At its surface, the scuff zone is loose red-brown and

charcoal-rich with scats and macro-fl oral remains.

Layer II – Compacted red-brown-grey sediment with brown/

grey laminae, same texture as Layer I, but with more ash.

Layer III – Fine red-brown sand with increasing roof fall/

boulders.

Layer IV – Dark red indurated sand with increased proportion of

grit and roof fall fragments.

Charcoal was present in variable amounts with little surviving

below 50cm depth. The eight charcoal samples submitted for

Figure 3 Plan of Bush Turkey Rockshelter 3, showing location of excavation and art panels.

Sample # Layer Spit Description Lab. No. 14C Age (years BP)


(probability)1 I 2 In scuff zone Wk-15271 197±34 0*-30 (9.5%)

58-121 (17.6%)135-296 (73%)

2 I/II 8 Hearth at boundary NZA-20627 386±39 320-491 (100%

3 II 5 Upper Layer II NZA-20626 422±33 326-409 (37.5%)436-505 (62.5%)

4 II 9 Middle of Layer II NZA-20629 793±32 654-732 (100%)

5 III 14 5cm below boundary with Layer II

NZA-20630 3013±36 2985-3259 (100%)

6 III 16 Middle of Layer III NZA-20631 2942±35 2880-2910 (4.6%)2919-3161 (95.2%)3194-3196 (0.2%)

7 III/IV 17 Boundary Layer III/IV Wk-15277 3665±63 3720-4090 (99.4%)4130-4138 (0.6%)

8 IV 20 Middle of Layer IV NZA-20632 7584±44 8202-8265 (17.6)8282-8414 (82.4%)

Table 1 Radiocarbon dates for Bush Turkey Rockshelter 3. Conventional radiocarbon ages were calibrated using CALIB (v.5.0.2) (Stuiver and Reimer 1993; Stuiver et al. 2007) and the Southern Hemisphere calibration dataset (McCormac et al. 2004). Ranges marked with a ‘0*’ are suspect owing to impingement on the end of the calibration dataset (Stuiver et al. 2007). NZA-codes are AMS dates.


Peter Veth, Jo McDonald and Beth White

radiocarbon determination were mostly recovered during

excavation in situ or from the south section (Table 1, Figure 4).

The lowest sample (from spit 20, 65-70cm bsl) was collected

from the sieves. Most samples weighed less than 10g and were

thus dated by AMS.

Only one sample was recovered from Layer IV (in Spit 20). A

further 16cm of deposit was recovered from beneath this spit

(yielding nine artefacts). These deeper artefacts were deposited prior

to the earliest radiocarbon date of 8202–8414 cal BP. The radiocarbon

dates generally occur in correct chronostratigraphic order, with a

minor inversion between the conventional ages of NZA-20630 and

NZA-20631, although the calibrated ages overlap at two standard

deviations. This reinforces the fi eld observation that the deposits

appeared to have a good level of integrity. There is no evidence for a

disconformity between Layers III and IV. Sedimentation was more

or less continuous and with increasing compaction with depth and

age giving an extrapolated basal date for the site of c.12ka BP.

Analysis of Lithic AssemblageThe lithic analysis addressed the following issues:

1. Identifi cation of artefacts. Some of the raw materials from

which artefacts have been made occur naturally in the shelter

walls. Identifi cation of lithics as artefacts rather than naturally-

fractured stone required special consideration. A total of 466

lithic pieces have been identifi ed as artefacts.

2. Raw material use and technological strategies. A variety of

stone material types have been used to manufacture artefacts

and an analysis of technological variables can lead to the

identifi cation of different strategies relating to raw material

use (e.g. conservation and rationing of non-local chalcedony

and more expedient use of locally available quartzite pebbles;

cf. Hayden 1977).

3. Changes in artefact assemblages over time. The use of different

raw material types, especially chalcedony and quartzite, show

change over time and this can be understood as change in

technological strategies.

Artefact Identifi cationStone artefacts were identifi ed using technological criteria

(Cotterell and Kamminga 1987; Holdaway and Stern 2004; Speth

1972). Many fl akes have been broken, either during fl aking or

afterwards by trampling, burning in hearths or by natural

weathering processes. Where lateral or distal portions have

been broken the artefact is classifi ed as a broken fl ake. Flakes

broken longitudinally are classifi ed as cone-split broken fl akes.

Sometimes only a fragment or piece of a fl ake was recovered (e.g.

a distal or medial piece) and these are classifi ed as fl ake fragments.

As some of the raw material types used for artefact manufacture

occur naturally in the shelter, fractured pieces without signs of

fl aking were not identifi ed as artefacts, even though it is possible

that some may be fragments of broken artefacts.

Conservation Strategies and General Comments on Artefact UseSubstantial research has been carried out on the behavioural

aspects of stone artefact production, particularly in relation

to settlement organisation and mobility (cf. Andrefsky 1998;

Hiscock 1988). Mobility is a particularly important consideration

when studying stone artefacts left by people who frequently

changed their place of residence, and who regularly moved from

residential sites into surrounding foraging areas. Such groups

would only carry limited equipment with them, especially if they

also had to carry small children, food and other resources. People

clearly developed strategies – often referred to as conservation

or rationing strategies – for dealing with the constraints

posed by mobility. Such strategies include light-weight and

multifunctional tools (Hayden 1977; Hook 1999; Mulvaney

1975:73), use of locally available stone (Byrne 1980; McNiven

1993), heat-treatment (Lurie 1989) and transportation of high

quality stone with predictable fl aking qualities (Morwood and

L’Oste-Brown 1995). Available stone supplies may have been

conserved by discarding cores and tools less often. Groups

could make more tools than usual from available supplies, by

fashioning smaller tools (Morwood and L’Oste-Brown 1995).

These strategies could result in fewer artefacts, lower densities

of artefacts on archaeological sites and smaller artefacts. More

artefacts which might otherwise have been discarded as debitage

could have been used as tools; so the ratio of tools to debitage

might be high. People could have extended use-life via greater

core rotation, fl aking cores to exhaustion, or by adopting bipolar

fl aking at the fi nal stages of core reduction (cf. Summerhayes

and Allen 1993).

The practical limitations of transporting stone are clear from

some of the arid zone literature. Sharp blades were diffi cult to

carry as the edges could easily be dulled. Blades were wrapped

in bark, and paper-bark pouches were carried under the arms

or in string belts or string bags (Binford and O’Connell 1989;

Jones and White 1988; Paton 1994). Flake tools and/or women’s

knives were sometimes carried by women in their hair (Binford

1989:164). Large choppers were not usually transported (Gould

1977:164). Large grinding stones were particularly ineffi cient to

carry long distances (Hamilton 1987), although exceptions to

this have been noted.

Gould (1977:163) notes that stone materials were obtained from

either specifi c quarries where usable stone was known and used

Figure 4 Bush Turkey Rockshelter 3, south sections, squares BTA and BTB, showing large intrusive boulder in BTA and column taken down to bedrock in BTB.



repeatedly, or from non-quarried stone which was obtained from

the surface at or near where it was needed for a particular task:

[M]uch more quarried than non-quarried raw material appears

in habitation campsites, despite the fact that quarried stone

represents only a minute fraction of stone used in the total

cultural system. The only exception to this is when surface scatters

of stone are found occurring naturally near the campsite [such

as is found on the sand plain outside Bush Turkey Rockshelter 3]

... Adze fl akes and adze slugs tend to be common in habitation

sites, where they are most often manufactured, used, maintained,

and replaced. Along with these one may also expect to fi nd some

hand-held fl ake scrapers and spokeshaves ... ‘Chopper-planes’

are left where they were used and can generally be found lying

near the base of any mulga tree which shows a scar on its trunk

to indicate removal of a slab of wood ... Flake knives are also

left where they were used, most often in close proximity to earth

ovens, where they were used in butchering and dividing meat

after a successful hunt (Gould 1977:166).

The utility of the quarry/non-quarry dichotomy has been

tested with data from the Pilbara (Veth 1982) and found to be a

useful explanatory approach when considering lithic procurement,

reduction and discard strategies. Other archaeologists have

reported variations on this general distinction between quarried

and local stone. Binford and O’Connell (1989:144) also note

that stone from quarries was transported to habitation sites and

could also be used in an expedient fashion for making tools for

immediate tasks.

Raw Material Use at Bush Turkey Rockshelter 3Analysis of the Bush Turkey Rockshelter 3 artefact assemblage

indicates that several raw material types were used − chalcedony

and quartzite being dominant (37% and 36% respectively).

Around 12% of artefacts have been made from silcrete and

8% from quartz. Five minor lithic categories contribute the

remaining 10%.

Quartzite is available at and near the site as both pebble-sized

inclusions in the shelter walls and in the creek bed. A source for

the chalcedony is not known locally although large outcrops are

known from the Canning Stock Route some 80–100km to the

north and west (Veth 1993). Strong difference between the uses

of these raw materials is found at this site.

The assemblage includes a broken quartzite cobble which

appears to have been used as a combination anvil and hammer

stone (Table 2). A small fragment of a grindstone (in an

unidentifi ed material) has grinding on one surface. A tula adze

slug and three other retouched artefacts of chalcedony (including

one cuspate retouched tool) were recovered. There were no

chalcedony cores. Conversely, bipolar cores and debitage in both

quartzite and quartz were recovered.

While chalcedony dominates the assemblage by count,

most of these artefacts are small: all but four are <20mm-long.

The chalcedony artefacts combined weigh less than 30g, with

a very low mean weight (<0.2g/artefact). In contrast, artefacts

made from quartzite, silcrete and quartz vary more widely in

size and have much higher average weights (≥2g/artefact). This

stark difference in the size of chalcedony artefacts compared

to those of quartzite, silcrete and quartz points to a marked

difference in the strategy of reducing chalcedony. This is

consistent with its prized isotropic qualities and endurance

for hardwood processing tasks and the assumed distance of

its supply zone.

Chalcedony and silcrete artefacts at this site seldom retain

cortex. In contrast, almost one-third of the quartzite artefacts

and nearly half of the quartz artefacts have cortex (Table S1,

supplementary information). The dorsal faces of some of

the quartzite artefacts are entirely covered with cortical and

weathered surfaces and it is likely that some were struck from

locally weathered pebbles. Conversely, it appears that cortex was

removed from most of the chalcedony and silcrete artefacts off-

site. This pattern is consistent with the assumed local and ‘exotic’

provenience of these raw materials.

It is also likely that variation in the fl aking quality of different

stone materials has contributed to different transportation and

reduction strategies. Most (86%) chalcedony artefacts are of high

quality isotropic stone, as are over half of the silcrete artefacts

(61%). Conversely only a small proportion of the quartz (14%)

and quartzite (7%) artefacts are of unfl awed stone (Table S2,

supplementary information).

Chalcedony artefacts are generally very small, mostly of high

quality and largely without cortex: a typical assemblage for a well-

reduced transported stone, prepared off-site and already well-

utilised before discard at this site. The tula adze slug and small

chalcedony cuspate retouched fl ake (both <5g) are a predictable

aspect of this strategy (Hiscock and Veth 1991).

Silcrete artefacts, while slightly less common, are often made

on high quality stone. These are generally larger in size than

chalcedony and often lack cortex. Silcrete artefacts were probably

also prepared before transport to the site.

Quartzite is almost as common as chalcedony. It is of poorer

isotropic quality and often retains cortex. Most quartzite bipolar

cores have cortex and can be identifi ed as fl aked pebbles or

pebble pieces (Table 3). It is likely that quartzite, possibly also

quartz, were local adjuncts to higher quality imported stone.

Bipolar fl aking of quartz and quartzite was part of the repertoire

for exploiting local pebbles.

A total of 14 (mostly quartzite) bipolar cores were recovered,

of which 12 were unbroken. Ten of these unbroken cores have a

striking axis of between 18mm and 26mm. Only two unbroken

cores have shorter lengths (Table 3). While the bipolar cores vary

considerably in weight (from c.20g to <2g) there may have been

a threshold beyond which the size of most fl akes struck from

these cores were not considered useful.

Platforms on fl akes and broken fl akes are predominantly

plain (Table 4). Focal, crushed or partly crushed specimens are

also present in high proportions. Platforms on chalcedony fl akes

and broken fl akes are predominantly plain (42%) – as is the

platform on the tula adze slug. These plain platforms, along with

the few cortical and ridged platforms, probably resulted from

unifacial (unidirectional) fl aking. Also present are many small

focal platforms, probably resulting from attempts to detach small

fl akes in the conservation of this prized material.

A number of chalcedony fl akes and broken fl akes also have

crushed or partly crushed platforms, indicating that some

platforms were too small to support force applications. No

chalcedony bipolar artefacts were identifi ed in the assemblage,

and nor were any chalcedony cores.



Mat

eria

l

An

vil-

Ham

mer

Gro

un

d F

rag

men

t

Tula

Slu

g

Ret

ou

ch/ U

se-W

ear

Po

ssib

le U

se-W

ear

Co

re

Bip

ola

r C

ore

Bip

ola

r D

ebit

age

Deb

itag

e

Tota

l

Tota

l W

eig

ht

(g)

Mea

n W

eig

ht

(g)

Chalcedony 0 0 1 3 0 0 0 0 168 172 26.5 0.15

Quartzite 1 0 0 1 1 0 7 9 147 166 474.6 2.86

Silcrete 0 0 0 1 1 0 0 0 52 54 106.7 1.98

Quartz 0 0 0 1 1 0 5 1 29 37 94.2 2.55

FGS 0 0 0 0 0 0 0 0 11 11 5.4 0.49

Chert 0 0 0 0 0 1 2 0 7 10 35.4 3.54

Sandstone 0 0 0 0 0 0 0 0 4 4 3.6 0.90

Silicifi ed Wood 0 0 0 0 0 0 0 0 1 1 2.6 2.60

Unidentifi ed 0 1 0 1 0 0 0 0 9 11 219.4 19.95

Total 1 1 1 7 3 1 14 10 428 466 968.4 2.08

Table 2 Bush Turkey Rockshelter 3: Raw materials and artefact types. FGS = Fine-grained siliceous.

Table 3 Bush Turkey Rockshelter 3: Bipolar cores.

Spit Material Quality Size (mm) Weight (g) Body Cortex Comments8 Chert Medium-high 21-25 6.7 Indeterminate <30% Striking axis = 18mm

8 Chert High 21-25 5.2 Pebble 31-69% Striking axis = 23mm

9 Quartzite Poor-medium 26-30 9.3 Pebble frag. 31-69% Striking axis = 20mm

9 Quartzite Medium 21-25 6.8 Pebble frag. 31-69% Striking axis = 19mm

9 Quartzite Medium 21-25 4.8 Pebble frag. 31-69% Striking axis = 21mm

9 Quartz Poor-medium 26-30 7.9 Indeterminate <30% Broken. Heavy battering along ridge

10 Quartzite Poor-medium 31-35 19.8 Indeterminate <30% Blocky core with bipolar fl aking. Striking axis = 26mm

10 Quartz Medium-high 31-35 14.2 Pebble 31-69% Striking axis = 26mm

10 Quartz Medium-high 26-30 4.1 Indeterminate 0 Broken. Striking axis = 27mm

11 Quartzite Medium-high 21-25 7.5 Pebble >70% Striking axis = 23mm

11 Quartzite Medium-high 21-25 3.0 Indeterminate <30% Striking axis = 22mm

11 Quartzite Medium-high 16-20 2.0 Indeterminate <30% Striking axis = 13mm

11 Quartz Poor-medium 25-30 7.3 Pebble >70% Striking axis = 24mm

20 Quartz Medium-high 16-20 1.4 Indeterminate 31-69% Striking axis = 14mm

Material Cortex Plain Ridged Scars Focal Bipolar Crushed & Partly Crushed

Total Ind.

Chalcedony

Debitage 5 40 3 0 32 0 15 95 1

Tula adze 0 1 0 0 0 0 0 1 0

Retouched Flake

0 0 1 0 0 0 0 1 0

Quartzite 5 16 2 2 3 4+3* 16 51 0

Silcrete 0 10 1 1 2 0 8 22 1

Quartz 0 2 0 0 1 1 1 5 0

Others 0 6 1 0 1 0 4 12 2

Total 10 75 8 3 39 8 44 187 4

Table 4 Bush Turkey Rockshelter 3: Platforms on fl akes and broken fl akes >5mm in size. * Three quartzite bipolar platforms have been fl aked directly from cortical surfaces. Ind.=Indeterminate.



The quartzite platform sample is small but plain platforms

and crushed/partly crushed platforms are most frequent.

Unifacial fl aking was an important fl aking technique along

with bipolar fl aking. Some of the crushed platforms may have

resulted from this latter technique. Two quartzite fl akes have

scarred platforms which indicate that they may have been struck

from bifacial cores.

The small sample of silcrete platforms shows they are mostly

plain and crushed. As with the chalcedony, silcrete appears to

have been fl aked unifacially, with some attempt to conserve the

material. The sample of platforms for other raw material types is

too small for meaningful comment.

Technological StrategiesThe assemblage from Bush Turkey Rockshelter 3 demonstrates

two different technological strategies. These are temporally

discrete (see below). One of these has been used for chalcedony

(and to a lesser extent silcrete) while the other has been used on

quartzite.

Conservation strategies were in place for the high quality

chalcedony assemblage. Special long-life tools (Gould 1977, 1980;

O’Connell 1977; Veth 1993) such as the tula adze and a cuspate

retouched fl ake were made from this material. A low cortex

incidence indicates extensive previous fl aking. Small artefact

size and detached fl akes with small platforms indicate extensive

knapping and restricted discard. The absence of cores from the

assemblage suggests the removal of (already small) chalcedony

cores for further use elsewhere.

Non-artefactual quartzite pebbles and weathered pieces are

common in the deposit and the largest of these weighs c.80g. The

quartzite artefact assemblage shows expedient use of this locally

available material, which fl akes less predictably than other stone

materials and was not used for long use-life implements. Overall,

larger artefact sizes indicate less intensive reduction. Cortical

rates in local raw materials are higher than in imports indicating

less preparation of the stone before transport to the shelter.

There is evidence for primary fl aking of weathered pebbles. Both

unifacial and bipolar fl aking have been employed. Bipolar fl aking

was probably used here as a practical technique for reducing

round pebbles into cores. Most bipolar cores were discarded

when their striking axes were >20mm-long and they weighed

>5g. The bipolar cores in this assemblage indicate that viable

raw materials were discarded here rather than being removed

for use elsewhere.

Chronological Change in Technological StrategiesThe sample of artefacts decreases with depth (Figure 5, Table

S3, supplementary information). Artefact numbers are highest

between spits 4 and 11; but the volume of excavated deposit per

spit diminished signifi cantly below spit 15.

The site appears to have been occupied most intensively

between c.400–1200 BP (i.e. spits 6–11). Occupation of the site

just prior to contact seems to be lower than the peak period,

while before c.1200 BP occupation was sporadic (Figure 5).

A rate of artefact discard per 100 years has been calculated

using the depth age curve (Table 5, Figure 6). These data suggest

that the highest artefact discard rate occurred in the last 1000

years and that artefact discard rates before this time, certainly

pre-3600 BP, were extremely low.

Are these variations in artefact deposition rates related to

changes in technology? Debitage was found throughout the

deposit except in spit 22 (Table 6). Retouched and bipolar

artefacts were recorded in spits 7–11 (Layer II) and in spits

19–23 (Layer IV). Retouched/utilised artefacts represent only

a minor component of the overall assemblage: a combination

anvil-hammer was recovered from spit 19; the tula adze slug was

recovered from spit 6; the ground fragment from spit 2.

Layers I and II represent the last 1200 years of occupation and

appear to be part of a single technological phase (Table 6). Layers III

and IV have been combined to create a more viable artefact sample

for these discussions, despite the obvious long timeframe, because

of similarities in stratigraphy and assemblage characteristics.

The vertical distribution of artefacts of different raw

material types (Tables 6-7) indicates change over time in

raw material use. Chalcedony is dominant in Layers I and II

(51%), while locally available quartzite is prevalent in Layers

III and IV (62%). Technological analysis of these two raw

material types indicates different reduction strategies for

chalcedony and quartzite. Change over time in the use of

these two raw material types may have resulted from shifts

in exchange networks, changing residential mobility and

more logistical provisioning linked with the use of changing

technological strategies.

Figure 5 Bush Turkey Rockshelter 3: Vertical distribution of artefacts.

0 200 400 600 800 1000 1200 1400

123456789

101112131415161718192021222324

spits

Artefacts/m³

Layer I

Layer II

Layer III

Layer IV

197±34 BP

386±39 BP

793±32 BP

3013±36 BP

2942±35 BP3655±63 BP

7584±44 BP

Figure 6 Bush Turkey Rockshelter 3: Age-depth curve. Samples 2 and 6 are omitted from this graph. Conventional ages for the samples are used.

0

10

20

30

40

50

60

70

80

90

100

Years BP

Dep

th b

elow

sur

face

(cm

)

0 1000

3

57

8

2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000

Layer I Layer II Layer III Layer IV



Layer Total Artefacts Years Accumulated Years Artefacts/100 yearsI 109 170-400 BP 270 18.8

II 306 400-1200 BP 800 47.8

III 33 1200-3600 BP 2400 5.9

IV 18 3600-12000 BP 8400 2.9

Table 5 Bush Turkey Rockshelter 3: Calculated artefacts per 100 years based on the age-depth curve and calculated volume of artefacts retrieved from each layer. NB: Layer IV may have taken even longer to accumulate.

Laye

r

Sp

its

An

vil-

Ham

mer

Gro

un

d F

rag

men

t

Tula

Slu

g

Ret

ou

ched

/Use

-Wea

r

Po

ssib

le U

se-W

ear

Co

re

Bip

ola

r C

ore

Bip

ola

r D

ebit

age

Deb

itag

e

# % # % # % # % # % # % # % # % # %I 1-5 0 0 1 0.9 0 0 0 0 0 0 0 0 0 0 0 0 108 99.1

II 6-11 0 0 0 0 1 0.3 6 1.9 2 0.7 1 0.3 13 4.2 7 2.3 276 90.2

III 12-17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 33 100

IV 18-24 1 5.6 0 0 0 0 1 5.6 1 5.6 0 0 1 5.6 3 16.7 11 61.1

Table 6 Bush Turkey Rockshelter 3: Summary distribution of artefact types.

Layer Spits Chalcedony Quartzite Silcrete Quartz Others TotalI 1-5 51 35 8 6 9 109

II 6-11 116 99 38 26 27 306

III 12-17 2 21 6 3 1 33

IV 18-24 3 11 2 2 0 18

Table 7 Bush Turkey Rockshelter 3: Summary distribution of raw materials.

Laye

r

Sp

its

0.0

-0.5

0.6

-1.0

1.1

-1.5

1.6

-2.0

2.1

-2.5

2.6

-3.0

3.0

-3.5

3.6

-4.0

4.1

-4.5

4.6

-5.0

5.1

-5.5

5.6

-6.0

6.1

-6.5

6.6

-7.0

Tota

l N

o.

Tota

l W

eig

ht

Mea

n

Wei

gh

tI 1-5 5 67 29 6 2 0 0 0 0 0 0 0 0 0 109 15.7 0.1

II 6-11 7 112 80 41 26 22 7 5 2 1 2 0 0 1 306 674.3 2.2

III 12-17 0 2 10 11 3 2 3 0 0 0 1 1 0 0 33 109.9 3.3

IV 18-24 0 1 4 5 2 2 0 2 0 1 0 0 1 0 18 168.5 9.4

Table 8 Bush Turkey Rockshelter 3: Artefact weight (g) by layer.

Layer Spits 0% <30% 40-60% >70% Total % Artefacts with Cortex

I 1-5 103 3 1 2 109 5.5

II 6-11 237 30 18 21 306 22.5

III 12-17 24 2 3 4 33 27.3

IV 18-24 10 3 1 4 18 44.4

Table 9 Bush Turkey Rockshelter 3: Proportion of artefacts with cortex.

Layer Spits Poor Poor-Medium

Medium Medium-Good

Good Poor-Good

Total N/A % Good Quality

I 1-5 1 5 15 28 57 1 107 2 53.3

II 6-11 8 27 51 72 142 5 305 1 46.6

III 12-17 0 8 12 10 3 0 33 0 9.1

IV 18-24 0 2 3 9 3 0 17 1 17.6

Table 10 Bush Turkey Rockshelter 3: Flaking quality.



Change over time in artefact size is also indicated. In Layer

I, all artefacts are <25mm in size and have a very low average

weight (Table 8). While artefact size in these upper spits may be

infl uenced by the small size of chalcedony artefacts, artefacts of

other raw material types are also relatively small in these spits.

In Layer II the modal artefact size is larger (11–15mm)

than in Layer I, and almost the full range of size categories are

represented by this larger sample of artefacts. Average artefact

weight is signifi cantly greater than in Layer I (2.2g/artefact). In

Layers III and IV there are very few artefacts <10mm in size.

Modal size increases to 11–20mm, and average artefact weight

exceeds 3g/artefact in Layer III and 9g/artefact in Layer IV. The

larger number of small artefacts in the upper layers indicates

more intensive fl aking of lithic materials in the past 800 years

or so.

The proportion of artefacts with cortex is lowest in Layer I in

all raw material types (Table 9). In both Layers II and III slightly

less than a third of all artefacts have cortex. Here, the higher

numbers of artefacts with cortex correlate with the increased use

of local quartzite pebbles.

Analysis of the fl aking quality of artefact stone indicates a

greater reliance on high quality stone in Layers I and II, while

local materials which dominate the earlier two layers, generally

have a lower fl aking quality (Table 10).

The vertical distribution of artefact types (Table 6) illustrates

increased use of bipolar reduction below spit 12 with greater

reliance on plain and focal platforms in the upper layers (Tables

11-12). Plain platforms possibly relate to tula technology and the

production of fl akes with wide broad platforms.

Summary of Assemblage CharacterisationOn the assumptions that the chalcedony recovered from Bush

Turkey Rockshelter 3 was procured from a distant quarry source

(such as Kaalpa at Well 23 on the Canning Stock Route) and that

quartzite is an abundant local stone found within the shelter

wall and the local creek, then the observed shift through time

in raw material use and reduction strategies suggests changes in

the way the site was occupied though time. Occupation from

the early Holocene to just before the last millennium refl ects

longer duration occupations and lower mobility at Bush Turkey

Rockshelter 3 and the Calvert Ranges, during which time local

stone was used predominantly for a range of tasks. Within the last

millennium, site usage changed exhibiting a greater proportion

of exotic lithics and increased reduction of all raw material classes.

During this time quarried stone was procured elsewhere, curated,

rejuvenated and then discarded here in much greater proportions.

A chalcedony tula adze slug and an arenaceous grindstone

fragment were discarded during this more recent time period.

The presumed strategic importance of the Calvert Ranges as an

aggregation centre with reliable water might explain this increase

in the proportion of exotics, higher discard rates and increased

reduction of all lithics. A shift from an embedded to a logistical

procurement strategy is suggested as is the greater frequency

of occupation visits (of shorter duration) especially during the

last millennia (cf. Holdaway and Stern 2004; Sellet et al. 2006).

The later signal suggests an increased use of the ranges as an

aggregation locale (sensu Conkey 1980), with various groups of

people coming from afar for a range of domestic and ceremonial

activities and focusing on this landscape more intensively. This

pattern is supported by the Calvert Ranges rock art which reveals

increasing complexity in its most recent art phases. While our

rock art analyses are still only preliminary, both the occupation

and art evidence supports increasing and repeated use of the

dunefi elds and the resource-rich ranges in the recent past (see

McDonald et al. 2008; Veth 1995).

The Rock ArtOn bedrock at the front of the shelter and on large boulders

outside the shelter (site P14) a series of Panaramitee-style

(sensu Maynard 1977) and more recent fi gurative engravings

are present. These are mostly heavily weathered and patinated

engravings of tracks, circles, complex non-fi gurative and

geometric motifs as well as fi gurative motifs such as life-size

bush turkeys and a human. These clearly post-date the shelter’s

formation through roof collapse, but pre-date the smaller

roof fall event(s) documented in the excavation. On its back

wall and ceiling are c.100 motifs, mostly monochrome (87%)

and bichrome (11%) paintings. There is also one red hand

print. The pigment assemblage comprises several phases of

production and includes anthropomorphs, concentric circles

and other geometric motifs (Table 13). A single faded red and

white anthropomorph with distinctive headdress is painted

in isolation high on the wall at the north of the shelter. The

pigment art includes black (charcoal), red, white, orange and

pink coloured ochres: white and red are the most common

(Table 14).

Layer Spits Cortex Plain Ridged Scars Focal Bipolar Crushed & Partly Crushed

Total Indeterminate

I 1-5 2 24 4 0 17 0 6 53 1

II 6-11 8 48 2 3 21 6 30 118 2

III 12-17 0 3 1 0 1 0 4 9 0

IV 18-24 0 0 1 0 0 2 5 8 1

Table 11 Bush Turkey Rockshelter 3: Type of platforms on fl akes and broken fl akes.

Layer Spits Cortex Plain Ridged Scars Focal Bipolar Crushed & Partly Crushed

Total

I 1-5 3.8 45.3 7.5 0 32.1 0 11.3 53

II 6-11 6.8 40.7 1.7 2.5 17.8 5.1 25.4 118

III/IV 12-24 0 17.6 11.8 0 5.9 11.8 52.9 17

Table 12 Bush Turkey Rockshelter 3: Type of platforms on fl akes and broken fl akes (% frequency).



A superimposition sequence has been identifi ed in this

pigment art assemblage which starts with a fi ne (brush-

painted) cherry-red set of motifs (including phytomorphs

and complex pole designs and concludes with an assemblage

of thickly painted (fi nger application) motifs in orange and

white of anthropomorphs, lizards, tracks and concentric circles

(Table 15). Not all art components within the shelter occur in

superimposition relationships, meaning that these cannot be

placed in this relative sequence. A bichrome headdress fi gure,

located in isolation high on the shelter’s back wall is one such

example. Its style and general condition (and superimpositions

noted elsewhere in the Calvert Ranges) suggest that this could be

amongst some of the older art in the pigment assemblage (but

possibly not as old as the defi ned Phase I which has been painted

with a fi ne brush/twig).

Intuitively we would propose that the majority of the

pigment art’s production (Art Phases 2–4 in Table 15) correlates

with the major phase of shelter occupation (i.e. within the last

millennium). The oldest pigment art is relatively sparse and

much more weathered. It seems reasonable to suggest that this

art pre-dates the site’s main occupation period, and could be

correlated with either the early or late Holocene occupation

of the site. The engravings could have been produced at any

time in the shelter’s occupation: unfortunately, no engraved

art is directly or indirectly dateable in this site. Samples from

several of the site’s red motifs have been collected (McDonald

and Veth 2006a) but dating results for these motifs are not

yet available. The dating techniques being employed are

reported elsewhere (McDonald and Veth in press; McDonald

et al. 2008).

Motif Monochrome Bichrome Print Engraved TotalAbraded Grooves 0 0 0 1 1

Bird Track 1 0 0 1 2

Complex-Non-Figurative 0 2 0 1 3

Scats 0 0 0 1 1

Oval 2 0 0 2 4

Anthropomorph 8 0 0 0 8

Arc 1 0 0 0 1

Barred Circle 1 0 0 0 1

Barred Oval 2 0 0 0 2

Circle 1 0 0 0 1

Concentric Circle 5 0 0 0 5

Fern 0 1 0 0 1

Hand 0 0 1 0 1

Headdress Figure 0 1 0 0 1

Lizard 1 0 0 0 1

Lizard Man 1 0 0 0 1

Phytomorph 1 0 0 0 1

Macropod Track Trail 1 0 0 0 1

Snake 1 0 0 0 1

Simple-Non-Figurative 5 1 0 0 6

Solid Unidentifi ed 7 0 0 1 8

Unidentifi ed Line 3 0 0 50 53

% Pigment Assemblage 87.2 10.6 2.1 0 100

Total 41 5 1 57 104

Table 13 Bush Turkey Rockshelter 3: Motif and technique information.

Black Orange Pink Red Red/White White TotalMotifs 3 6 7 12 5 14 47

% 6.4 12.8 14.9 25.5 10.6 29.8 100

Table 14 Bush Turkey Rockshelter 3: Proportions of different coloured pigments used in the pigment art assemblage (note, the site’s other motifs are engraved).

Phase Technique Motifs1 (Oldest) Fine cherry-red painted (brush) outline and

infi ll (including geometric infi ll)Phytomorph, complex pole design, concentric circles

2 White (and pink) painted outline Anthropomorph, simple-non-fi gurative

3 Black washy painted outline Simple-non-fi gurative, circle

4 (Most recent) Orange and red thick painted outline (fi nger) Anthropomorph, lizard man, bird track, concentric circle

Table 15 Bush Turkey Rockshelter 3: Superimposition sequence for pigment art.



Discussion and Conclusions

Research Problems in the Sandy DesertsThe initial occupation of the Western Desert has been placed at

c.24,000 BP at Serpents Glen (Katjarra) in the Carnarvon Ranges

to the southwest of the Calvert Ranges (O’Connor et al. 1998;

Veth 2000, 2005; Veth et al. 2000). Serpents Glen is located on

the margin of the Little Sandy Desert dunefi elds and contains

a major multiphase painted art corpus. There were no cultural

materials registered at Serpents Glen for most of the duration

of the Last Glacial Maximum (after Lambeck and Chappell

2001). Reoccupation occurred during the mid-Holocene with

low discard rates until the last several hundred years, when

occupational intensity increased signifi cantly: nearly 97% of

the artefacts deposited at Serpents Glen date to the last several

hundred years. A similar pattern of intensive occupation with

higher implement diversity was noted at Kaalpi Rockshelter

(Veth et al. 2001).

Kaalpi Rockshelter is located in a major valley southeast

of Bush Turkey Rockshelter 3. This site registered a signifi cant

increase in its discard rates of fl aked and ground artefacts

during the last millennium and a shift towards ‘formal’ wet-

milling grindstones made in arenaceous sandstone from an

earlier assemblage of amorphous basal grindstones fashioned

from quartzite (Veth et al. 2001). At this time there was also an

increase in ochre fragments. This pattern is essentially the same

as found in another seven shelters excavated in the wider region

including the Rudall River, Mackay Range and the Durba Hills

in the Little Sandy Desert (Veth 1993). Signifi cantly, Bush Turkey

Rockshelter 3 fi lls a chronological ‘gap’ by providing well-dated

cultural materials spanning the Holocene while reinforcing the

wider trend of an increase in the discard rates of artefacts and

sedimentation during the last millennium.

In a recent paper on the artefact assemblages from Puritjarra

Rockshelter, at the eastern margin of the Western Desert, Smith

(2006:406) makes two conclusions regarding changes occurring

there during the last 800 years. He identifi es a major increase

in the use of the rockshelter with increased artefact discard

rates, more features and ochre, more formal grindstones, and

the highest indices for the diversity and richness of tools. This

is interpreted as the archaeological expression of Puritjarra

becoming a core residential site linked to improved summer

rainfall between 1000 to 1500 years ago. During this period

Smith posits that there was a dual pattern of occupation with

frequent cycling of people through the site intercalated with

extended periods of occupation at the site.

A switch in provisioning strategies, from embedded

procurement (for both stone and ochre) to a logistical pattern

of direct procurement is also suggested as use of the shelter

intensifi ed (Binford 1989; Holdaway and Stern 2004; Smith

2006:404; Veth 1993). Smith favours an explanation whereby

a ‘dual’ system of occupation occurs during the emic cycle:

with core habitation sites being the focus of more extended

occupation as well as being frequently visited outside of these

‘aggregation’ events (Veth 1993). By calculating rank order scores

for residential mobility at the Puritjarra sequence, Smith (2006:

Table 19) illustrates a steady decrease in mobility in the last 800

years. While we believe a similar dual system operated in the

Calvert Ranges, a combination of increased exotics, their heavy

reduction and increased discard rates argue for an increase in

mobility and likely more frequent site visits.

The Contribution of Bush Turkey Rockshelter 3Our recent excavation of Bush Turkey Rockshelter 3 has

filled a major gap by establishing a record of early Holocene

occupation in the Little Sandy Desert and resolving a long

debate about the antiquity of occupation in this region. As

well as demonstrating occupation at 8202–8414 cal BP, the

major occupation phase here is older than the main phase

at Katjarra (98% of the artefacts at Bush Turkey Rockshelter

3 were deposited in the last 1200 years but only 24% were

deposited in the last 500 years or so). It has also refined and

extended the sequence found earlier at the Kaalpi Rockshelter

located around 5km south in a separate valley of the Calvert

Ranges (Veth et al. 2001).

Bush Turkey Rockshelter 3 has two distinct phases of

lithic production which illustrate changing raw material use

and technological organisation through time. Changes in

residential mobility patterns and the nature of site occupation

are predicted to correlate with different phases of pigment

art production.

We propose that from c.8400–1200 cal BP (effectively

Layers III and IV) this rockshelter was used for longer periods

of time by small groups exploiting an extensive territory. The

majority of stone used throughout this time is local quartzite.

There is no evidence for core exhaustion or intensive stone

curation/rejuvenation. Non-local materials such as chalcedony

are present but only in very low proportions. In Layer II, post-

dating 1200 cal BP and peaking around 735 cal BP, there is

evidence for curation and increased conservation strategies

with an appreciable shift towards:

• predominant use of chalcedony and other non-local high

quality raw materials;

• a signifi cant decrease in mean size and weight of artefacts;

• a major decrease in the proportion of artefacts with cortex;

• a high proportion of plain and small focal platforms on

chalcedony fl akes;

• the presence of a hafted artefact; and

• a signifi cant increase in artefact deposition rates.

These changes reinforce the fi ndings made earlier at Kaalpi,

where there was a notable infl ux in ochre and formal grindstones

and mullers in the deposit after 1300 cal BP. Interestingly at

that site, the artefact discard rate halved in the most recent

millennium (Veth et al. 2001:13). We believe that these combined

changes over time refl ect shifts in exchange networks, increased

artistic activity, increasing residential mobility patterns and more

logistical provisioning of tool kits.

The nine sites excavated within and adjacent to the Little Sandy

Desert over the last 20 years (Kaalpi Rockshelter, Calvert Range

Rockshelter, Durba Hills Rockshelter, Karlamili Rockshelter,

Yulpul Rockshelter, Jalpiyari Rockshelter, Winakurijuna Cave,

Serpents Glen (Katjarra) Rockshelter and Katampul Rockshelter)

display a broadly similar trend with a greater proportion of exotic

lithics and increasing conservation strategies for all raw materials

over time (cf. Smith 1989, 2006; Smith et al. 1997, 1998). This

inter-regional pattern is best explained by:



• establishment of homelands and relative circumscription

of territory (with a narrowing of ochre catchments and the

development of regionally distinct graphic vocabularies in

the pigment art);

• intensifying trade/exchange networks refl ecting increasingly

complex social relations (with ramifi ed kinship systems

underwriting reciprocity and exchange networks both as a

social lubricant and a risk-minimising strategy);

• a shift from an embedded to a logistical pattern of direct

procurement of stone and ochres (as estates and their resource

zones became socially proscribed);

• higher residential mobility with lower duration of

occupations;

• a dual system of land-use embracing frequent domestic visits

by bands as well as aggregation events including multidialect

meetings; and

• economising of lithics as these took on multivalent values

(long use-life, multifunctionality, standardisation and sacred

elements linked with the Jukurrpa).

Western Desert rock art provides evidence that the widely

ramifi ed open social networks functioning at contact, have

changed through time (McDonald 2005). Arid art systems

appear to have functioned to provide broad-scale intergroup

cohesion over vast periods of time as well as demonstrating,

more recently, distinctive group-identifying and bounding

behaviour. We have argued elsewhere (McDonald and Veth

2006b) that the high degree of stylistic variability displayed

in the Calvert Ranges art province strongly suggests that this

place has acted as an aggregation locale over a considerable

period of time.

At Bush Turkey Rockshelter 3 we have demonstrated a

changing pattern of site use from the early Holocene through

to a time just before contact. This reinforces the patterning

found at Kaalpi, while showing a slightly different signature of

lithic behaviour (i.e. only one ground fragment was retrieved

from the top layer of Bush Turkey Rockshelter 3 compared

with the c.50 grindstones found in Kaalpi; see Haley 1999).

Interestingly the rock art assemblages in these two shelters

also are widely dissimilar, while sharing broad stylistic traits.

The Kaalpi Rockshelter contains a complex assemblage of

370 (mostly pigment) motifs (cf. Table 13). No doubt the two

sites represent different nodes in a complex local occupation

pattern. Both would appear to have represented residential

bases, especially after c.1300 BP – although the intensity and

nature of occupation in these two shelters in this time period

varies. Our dating work on the Calvert Ranges art phases has

involved the collection of samples from Kaalpi as well as Bush

Turkey Rockshelter 3 (McDonald and Veth in press; McDonald

et al. 2008). These rock art assemblages, with their complex

graphic vocabularies in the recent past, add to this picture of

increased dynamism.

The arid landscapes of the Western Desert provide a strong

archaeological signature for intensifying social relations in

the recent past. It is perhaps ironic that the strongest case for

an archaeological expression of intensification (Lourandos

1985; Veth 2006) comes not from the fertile plains or tropical

seascapes of eastern Australia, but rather Australia’s arid

zone, where effective social strategies for successful survival

were crucial. The stone-working strategies and highly

diverse graphic vocabularies in these isolated but resource-

rich ranges continue to demonstrate the complexities of

these strategies.

Supplementary InformationSupplementary information for this article is available online at

www.australianarchaeologicalassociation.com.au.

AcknowledgmentsThis fi eldwork was done in association with the Birrilibulu Native

Title claim, coordinated by Dr Bill Kruse of the Ngaanyatjarra

Land Council. Drs Kathryn Przywolnik and June Ross assisted

JM recording rock art. The site was excavated by PV with

occasional assistance from KP and JM (under a Section 16

permit granted by DIA). The artefact analysis was done by BW.

The radiocarbon dates were paid for by Jo McDonald Cultural

Heritage Management Pty Ltd.

ReferencesAndrefsky, W. 1998 Lithics: Macroscopic Approaches to Analysis. Cambridge:

Cambridge University Press.

Binford, L.R. 1989 An Alyawara day: Flour, spinifex gum, and shifting perspectives.

In L.R. Binford (ed.), Debating Archaeology, pp.147-171. San Diego: Academic

Press.

Binford, L.R. and J.F. O’Connell 1989 An Alyawarra day: The stone quarry. In L.R.

Binford (ed.), Debating Archaeology, pp.121-146. San Diego: Academic Press.

Byrne, D. 1980 Dynamics of dispersion: The place of silcrete in archaeological

assemblages from the lower Murchison, Western Australia. Archaeology and

Physical Anthropology in Oceania 15:110-119.

Conkey, M.W. 1980 The identifi cation of prehistoric hunter-gatherer aggregation

sites: The case of Altamira. Current Anthropology 21:609-630.

Cotterell, B. and J. Kamminga 1987 The formation of fl akes. American Antiquity

52(4):675-708.

Gould, R.A. 1977 Ethno-archaeology; or, where do models come from? In R.V.S.

Wright (ed.), Stone Tools as Cultural Markers: Change, Evolution and

Complexity, pp.162-168. Canberra: Australian Institute of Aboriginal Studies.

Gould, R.A. 1980 Living Archaeology. Cambridge: Cambridge University Press.

Haley, M. 1999 Kaalpi: Investigation of Archaeological Assemblages from the

Calvert Ranges Western Desert, Western Australia. Unpublished BA (Hons)

thesis, Department of Anthropology and Archaeology, James Cook University,

Townsville.

Hamilton, A. 1987 Dual social system: Technology, labour and women’s secret rites

in the eastern Western Desert of Australia. In W.H. Edwards (ed.), Traditional

Aboriginal Society: A Reader, pp.34-52. South Melbourne: MacMillan.

Hayden, B. 1977 Stone tool functions in the Western Desert. In R.V.S. Wright (ed.),

Stone Tools as Cultural Markers: Change, Evolution and Complexity, pp.178-

188. Canberra: Australian Institute of Aboriginal Studies.

Hiscock, P. 1988 Prehistoric Settlement Patterns and Artefact Manufacture at

Lawn Hill, Northwest Queensland. Unpublished PhD thesis, Department of

Anthropology and Sociology, University of Queensland, Brisbane.

Hiscock, P. and P. Veth 1991 Change in the Australian desert culture: A reanalysis of

tulas from Puntutjarpa Rockshelter. World Archaeology 22(3):332-345.

Holdaway, S. and N. Stern 2004 A Record in Stone: The Study of Australia’s Flaked

Stone Artefacts. Melbourne: Museum Victoria and Canberra: Aboriginal

Studies Press.

Hook, F. 1999 An Aboriginal Archaeological Heritage Assessment of Proposed Road

Alignments Along the Great Central Road, East of Laverton, Western Australia.

Unpublished report to Ngaanyatjarra Council (Aboriginal Corporation).



Jones, R. and N. White 1988 Point blank: Stone manufacture at the Ngilipitji

quarry, Arnhem Land, 1981. In B. Meehan and R. Jones (eds), Archaeology

with Ethnography: An Australian Perspective, pp.51-87. Canberra:

Department of Prehistory, Research School of Pacifi c Studies, Australian

National University.

Lambeck, K. and J. Chappell 2001 Sea level change throughout the Last Glacial

cycle. Science 292:679-686.

Lourandos, H. 1985 Intensifi cation and Australian prehistory. In T.D. Price and

J.A. Brown (eds), Prehistoric Hunter-Gatherers: The Emergence of Cultural

Complexity, pp.385-423. New York: Academic Press.

Lurie, R. 1989 Lithic technology and mobility strategies: The Koster site Middle

Archaic. In R. Torrence (ed.), Time, Energy and Stone Tools, pp.46-55.


Maynard, L. 1977 Classifi cation and terminology in Australian rock art. In P.J. Ucko

(ed.), Form in Indigenous Art, pp.387-402. Canberra: Australian Institute of

Aboriginal Studies.

McCormac, F.G., A.G. Hogg, P.G. Blackwell, C.E. Buck, T.F.G. Higham and P.J. Reimer

2004 SHCal04 southern hemisphere calibration 0-11.0 cal kyr BP. Radiocarbon

46(3):1087-1092.

McDonald, J.J. 2005 Archaic faces to headdresses: The changing role of the rock

art across the arid zone. In P.M. Veth, M. Smith and P. Hiscock (eds), Desert

Peoples: Archaeological Perspectives, pp.116-141. Boston: Blackwell.

McDonald, J.J., K. Steelman and P.M. Veth 2008 Radiocarbon Dating of Western

Desert Pigment Samples: Dating Rock Art in the Calvert and Carnarvon

Ranges. Report to AIATSIS and DIA in support of Section 16 Permit and

AIATSIS Research Grant.

McDonald, J.J. and P.M. Veth 2006a S16 Report: Rock Art Dating Work in the Calvert

and Carnarvon Ranges, Western Desert. Unpublished report to Western

Australian Department of Indigenous Affairs, Perth.

McDonald, J.J. and P. Veth 2006b Rock art and social identity: A comparison of

graphic systems operating in arid and fertile environments in the Holocene. In

I. Lilley (ed.), Archaeology of Oceania: Australia and the Pacifi c Islands, pp.96-

115. Oxford: Blackwell.

McDonald, J.J. and P.M. Veth in press Stylistic variability in the art of the Western

Desert and Pilbara: New perspectives on the role of art in the Australian arid

zone. Australian Aboriginal Studies.

McNiven, I. 1993 Raw material proximity and bevel-edged tool use, Teewah Beach,

southeast Queensland. Archaeology in Oceania 28(3):138-143.

Morwood, M.J. and S. L’Oste-Brown 1995 Chronological changes in stone artefact

technology. In M.J. Morwood and D.R. Hobbs (eds), Quinkan Prehistory: The

Archaeology of Aboriginal Art in S.E. Cape York Peninsula, Australia, pp.161-

177. Tempus 3. St Lucia: Anthropology Museum, University of Queensland.

Mulvaney, D.J. 1975 The Prehistory of Australia. Middlesex, England: Penguin

Books.

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.

O’Connor, S., P. Veth and C. Campbell 1998 Serpent’s Glen: A Pleistocene

archaeological sequence from the Western Desert. Australian Archaeology

46:12-22.

Paton, R. 1994 Speaking through stones: A study from northern Australia. World

Archaeology 26(2):172-184.

Sellet, F., R. Greaves and P-L. Yu 2006 (eds) Archaeology and Ethnoarchaeology of

Mobility. Tallahassee, FL: University Press of Florida.

Smith, M.A. 1989 The case for a resident human population in the central Australian

Ranges during full glacial aridity. Archaeology in Oceania 24(3):93-105.

Smith, M. 2006 Characterising late Pleistocene and Holocene stone artefact

assemblages from Puritjarra Rock Shelter: A long sequence from the Australian

desert. Records of the Australian Museum 58:371-410.

Smith, M.A., B. Fankhauser and M. Jercher 1998 The changing provenance of red

ochre at Puritjarra rockshelter, central Australia: Late Pleistocene to present.

Proceedings of the Prehistoric Society 64:275-292.

Smith, M.A., J.R. Prescott and M.J. Head 1997 Comparison of 14C and luminescence

chronologies at Puritjarra rock shelter, Central Australia. Quaternary Science

Reviews 16(3-5):299-320.

Speth, P. 1972 Mechanical basis of percussion fl aking. American Antiquity 37(1):34-

60.

Stuiver, M. and P.J. Reimer 1993 Extended 14C data base and revised CALIB 3.0 14C

age calibration program. Radiocarbon 35(1):215-230.

Stuiver, M., P.J. Reimer and R. Reimer 2007 CALIB Manual. Retrieved 24 March

2008 from http://calib.qub.ac.uk/calib/manual/.

Summerhayes, G.R. and J. Allen 1993 The transport of Mopir obsidian to Late

Pleistocene New Ireland. Archaeology in Oceania 28(3):144-148.

Veth, P. 1982 Testing the Behavioural Model: The Use of Open Site Data.

Unpublished BA (Hons) thesis, Department of Anthropology, University of

Western Australia, Perth.

Veth, P.M. 1993 Islands in the Interior: The Dynamics of Prehistoric Adaptations

within the Arid Zone of Australia. Archaeology Series 3. Ann Arbor, MI:

International Monographs in Prehistory.

Veth, P. 1995 Aridity and settlement of northwest Australia. Antiquity 69:733-746.

Veth, P. 2000 Origins of the Western Desert language: Convergence in linguistic and

archaeological space and time models. Archaeology in Oceania 35(1):11-19.

Veth, P. 2005 Cycles of aridity and human mobility: Risk-minimization amongst

late Pleistocene foragers of the Western Desert, Australia. In P. Veth, M.A. Smith

and P. Hiscock (eds), Desert Peoples: Archaeological Perspectives, pp.100-115.

Oxford: Blackwell Publishing.

Veth, P. 2006 Social dynamism in the archaeology of the Western Desert. In B.

David, I.J. McNiven and B. Barker (eds), The Social Archaeology of Indigenous

Societies, pp.242-253. Canberra: Australian Aboriginal Studies Press.

Veth, P., S. O’Connor and L.A. Wallis 2000 Perspectives on ecological approaches in

Australian archaeology. Australian Archaeology 50:54-66.

Veth, P., M. Smith and M. Haley 2001 Kaalpi: The archaeology of a sandstone outlier

in the Western Desert. Australian Archaeology 52:9-17.


GLEN THIRSTY: The History and Archaeology of a Desert WellM.A. Smith1 and J. Ross2

AbstractThe archaeology of Glen Thirsty, a desert well in the Amadeus

Basin, Central Australia, illustrates the changing relationship

between the ranges and desert lowlands during the last 1500

years. Historical records and Aboriginal accounts of the site

document the regional importance of Glen Thirsty as one

of the few wells in this part of the desert. Archaeological

excavations and rock art research show that despite its

proximity to Puritjarra with its long, late Pleistocene record

of occupation, Glen Thirsty only became an important

focus of occupation after 1500 BP. Several lines of evidence

independently suggest the establishment and consolidation

of a new cultural and economic landscape in the Glen Thirsty

area around this time. Growing population pressure and shifts

in patterns of land-use and economy in the Central Australian

ranges may have provided the impetus for more intensive

use of the Glen Thirsty area, although the timing of this was

constrained by climatic factors. As a rain-fed well in the lower

part of the Amadeus Basin, Glen Thirsty is sensitive to shifts

in palaeoclimate and its history refl ects changes in regional

rainfall patterns during the late Holocene.

IntroductionThe fi rst European exploring party to enter western Central

Australia reached a desert well called ‘Glen Thirsty’ in October

1872, and then turned back. Led by Ernest Giles, the expedition

had set out to fi nd a route from the newly-built Overland

Telegraph Line in Central Australia, across the desert, to the west

coast of the continent (Giles 2000[1875]). Mounted on horses,

Giles and his two companions failed in their attempt to push

west, and turned south from Mt Udor through spinifex and sand

hill country, fi nding a shady waterhole at Tjungkupu (Giles’ ‘Tarn

of Auber’) (Figure 1). Further south, and two and a half months

into their expedition, they reached a series of quartzite ridges on

the northern rim of Lake Amadeus. Here Giles and Carmichael

found a small soakage well (latitude: 24º28'34.9" S, longitude

131º01'55.2" E), which Giles named ‘Glen Thirsty’, ‘for whenever

we returned to it ourselves and horses were choking for water’

(Giles 2000[1875]:74). The water was ‘of a thick and muddy and

rather nauseous fl avour’ and only seeped slowly into its base

(2000[1875]:65). Although they substantially enlarged the native

well – digging a trench about 6m-long and 2m-deep – there was

still barely enough water for the party and their horses.

Attempts to travel west were foiled by lack of water. To the

south, Giles was drawn by the highpoints of Mt Olga (Katatjuta)

visible on the horizon, but found his way blocked by an impassable

salt lake, which he named Lake Amadeus. Here his party ‘sank

up to our knees (when once the crust was broken) in hot salt

mud’ and his horses ‘fl oundered about in the bottomless bed

of this dreadful lake’ (2000[1875]:67). Desperate for water, he

fell back to Glen Thirsty three times between 12 and 27 October

1872. Finally, ‘not having met with another place in the whole

of our travels in this part of the country where another drop of

water was to be found’ (2000[1875]:75), Giles retreated 80km to

Tjungkupu. Looking for a route back to the Overland Telegraph

Line and the Finke River, he turned east towards the high country

of the George Gill Range, where he found numerous spring-fed

waterholes and well-grassed valleys in the area that today forms

Watarrka-Kings Canyon National Park.

While camped at Glen Thirsty, Giles explored the surrounding

ridges, describing the rock art he saw nearby:

We took a stroll up into the rocks and gullies of the ridges,

and found a cave ornamented with the choicest specimens of

aboriginal art ... One hieroglyph was most striking. It consisted

of two Roman numerals, a V and an I placed together, and

representing our fi gure VI. They were both daubed over with

spots, and were painted with red ochre (Giles 2000[1875]:65).

Because of its inaccessible location, Glen Thirsty has had

few European visitors since 1872. In 1999, Ross relocated the

distinctive ‘Roman VI’ panel of paintings described by Giles and

showed that Glen Thirsty had a wide range of archaeological

remains in addition to rock paintings, including stone

artefact scatters, rock engravings and several rockshelters with

occupation deposit. There were also numerous stick-nest rat

(Leporillus sp.) middens on rocky ledges in the area. We saw the

1 Centre for Historical Research, National Museum of Australia, GPO Box 1901, Canberra, ACT 2601, Australia

² Discipline of Archaeology and Palaeoanthropology, School of Humanities, University of New England, Armidale, NSW 2351, Australia

Figure 1 Western Central Australia, showing Glen Thirsty and places mentioned in the text.


Glen Thirsty: The History and Archaeology of a Desert Well

potential, therefore, to reconstruct the history of an outlying

well in the desert lowlands, integrating an analysis of rock art

with archaeological excavations at several rockshelter sites and a

record of palaeovegetation from the stick-nest rat middens.

One of the key questions concerns the changing relationship

between the ranges and desert lowlands during the last 1000–1500

years. This is a period of major change in the Western Desert and

Central Australia (Smith 1996; Smith and Ross 2008; Thorley

1999, 2001; Veth 1993, 2005) during which, as Thorley (2001)

suggests, the use of marginal or outlying areas (such as Glen

Thirsty) may have played an integral adaptive role in sustaining

higher regional populations in focal areas of occupation (such

as in the George Gill Range, Palmer catchment and Cleland

Hills areas). To the north of Glen Thirsty, Tjungkupu (another

desert well and waterhole) was only used intensively after 1000

BP, despite its proximity to Puritjarra rockshelter with its longer,

late Pleistocene record of occupation. In this context, we set out

to establish how events during the late Holocene had affected use

of the Glen Thirsty area – and whether or not there was an earlier

record of occupation here.

Physical SettingToday, Glen Thirsty is much as Giles found it in 1872. The

area centres on an anticlinal fold of quartzite ridges set in

dunefi eld and low scrub, 20–25km north of Lake Amadeus

and approximately 60km southwest of the George Gill Range

(Figures 1-2). The complex of ridges forms an isolated rocky

catchment, capturing and concentrating run-off and directing

this into several creeklines. A narrow pass, called ‘Worrill’s Pass’

by Giles, provides the only convenient passage through a high

quartzite ridge that rises out of the surrounding sand plain to

a height of 20m. Just beyond Worrill’s Pass, to the south of

the ridge, the various creeklines coalesce and fl ood out into

the sand plain creating ‘a small and well-grassed plain’ with

a bright patch of fresh grass and acacia shrubs (Giles’ ‘glen’)

(Giles 2000[1875]:60). Further to the south, spinifex and sand

dunes again dominate the landscape.

There is no longer any sign of the 1872 well, but the most

likely location is 1.6km southwest of Worrill’s Pass (Giles

(2000[1875]:75) notes the pass was ‘a mile from the tank’),

where the creek is deeply entrenched and the sand today holds

some moisture. Rainwater seepage into shallow sandy aquifers

is reflected by the growth of several large eucalpyts along the

channel in this area. This is also the case, to the north of

Worrill’s Pass, where the creek forms a gutter running along

the base of a quartzite ridge, and several ephemeral waterholes

(Figure 2C). Along the upper reaches of the western tributary,

some 2.6km west of the pass, there are several rock holes

(Figure 2B). Giles (2000[1875]:60) commented on these:

‘upon searching further up the gully we found some good-

sized rock holes, but unfortunately they were all dry’. The

presence of mud rather than sand in the channels, rock holes

and waterholes in the upper reaches of these drainage lines

suggests only gentle flows of storm water, even after heavy

rain. When Giles visited the area it was late in the dry season

and some heavy rain had fallen, but not sufficient to recharge

local rock holes.

The whole system at Glen Thirsty appears to be an isolated

rain-fed catchment focused on the ‘glen’. There is no evidence

of groundwater discharge or of local springs. In any case,

groundwater in the lower part of the Amadeus Basin is mostly

saline and undrinkable (Jacobson et al. 1989).

Figure 2 Map of the Glen Thirsty area. Open circle shows probable position of the 1872 well. A: Sites GT1-3, 7-9; B: Sites GT4-6, 10-13 and location of rock holes mentioned by Giles; C: Open sites and quarried outcrop of silcrete.


M.A. Smith and J. Ross

Aboriginal HistoryAboriginal people knew the Glen Thirsty area as ‘Yatajirra’, or

perhaps ‘Nyatajirri’ (Smith 2005:17, Footnote 48). When Giles

visited the area in October 1872, rain had allowed people to

disperse out into the spinifex and sand hill country from their

fallback waters, and they were busy fi ring the surrounding

country. Giles (2000[1875]:69) notes ‘the whole atmosphere was

thick with a smoky haze’ and that ‘the natives were ever busy

at their grass fi res’. Aboriginal people also continued to use the

soakage between Giles’ visits ‘as their tracks were visible in the

sand around the tank’ (Giles 2000[1875]:73) though people

avoided direct contact with the exploring party.

By 1880, the pastoral frontier in western Central Australia

was represented by a line of outposts from Glen Helen and

Hermannsburg Mission in the north, to Tempe Downs and

Erldunda in the south. The sand hill country north of Lake

Amadeus remained well outside this zone. In 1889 a police patrol,

which entered the area a day’s walk to the north of Glen Thirsty,

encountered a party of 17 Aboriginal people, including old women

and children, travelling south towards the salt lake (Willshire

1895; see also Smith 2005:23-25). The onset of severe drought

from 1895 to 1906 prompted many Kukatja and Matuntara

people to migrate to pastoral stations or Hermannsburg Mission

and it seems likely that the area north of Lake Amadeus was only

lightly or intermittently occupied after this time. Although the

area was increasingly depopulated, an exploring party led by R.T.

Maurice, met a small group of Aboriginal people (a man, two

women and two children) in July 1902 about 50km northwest

of Glen Thirsty (Murray 1904). Some people continued to live

in the sand hill country north of Lake Amadeus into the 1920s.

Peter Bullah (or Bulla) a Luritja/Matuntara man (the senior

custodian of Glen Thirsty into the 1990s) was born around 1920

at Katulkira or Katulykira, a day’s walk north of Glen Thirsty, an

area that he always described as his ‘borning country’ (though

his father’s country was actually Katiti, on the southern side of

Lake Amadeus – Maurice 1989:30). His elder brother, Napula

Jack, guided Michael Terry’s prospecting party through the

area in 1932 (Terry 1937; see also Smith 2005:63-65) and Syd

Coulthard (grandson of his eldest sister Anyinina) was our guide

after Peter Bullah’s death in 2001.

By the 1940s, the country on the northern side of Lake Amadeus

was becoming more diffi cult to reach and intergenerational

transfer of knowledge was also becoming more diffi cult to

manage. Syd Coulthard recalls visiting the Glen Thirsty area

with a string of camels in 1955, shortly after he was initiated at

Hermannsburg Mission by Tiger Tjalkalyiri (Peter Bullah’s elder

brother), and again after the exceptional rains of 1974. In each

case, it was an ‘emergency trip’, prompted by the age and infi rmity

of senior men and the need to learn the country before they died.

In 1955, their travelling route with camels was from Palm Valley

to ‘Unturu, Wiputa, and Yatajirri’, and then through to ‘Yirarka

and Mutitjulu’. ‘My grandfather brought me out here to show all

the things, for secret things, to learn more, to get it into my mind’

(Syd Coulthard in Smith 2005:72-73). Towards the end of his life,

Peter Bullah attempted a similar trip in about 1988, with a car-

load of young men, but they ran out of fuel in sand hill country.

Some then walked out to Watarrka-Kings Canyon; others were

rescued by a helicopter organised by Northern Territory Parks

and Wildlife Service staff.

The recent Aboriginal history of Glen Thirsty thus emphasises

links with the better-watered range country to the north and

northeast, to the Ulpanyali-Watarrka-Lilla area. Historical

records alone do not allow us to say whether these links refl ect

the contingencies of the last hundred years, or whether they

accurately refl ect the pre-contact situation (although the

archaeology suggests the latter – see below). But whether or

not Glen Thirsty was a discrete clan estate, or part of one that

included the better-watered country in the ranges, its use must

always have been dependent on seasonal rainfall.

Fieldwork 1999–2004Initial archaeological appraisal of the sites was limited to

a short helicopter and ground reconnaissance by Ross in

October 1999 (Ross 2003), organised by Northern Territory

Parks and Wildlife Service staff at the request of Peter Bullah.

Tapes of Bullah singing songs and relating stories associated

with Glen Thirsty were recorded by Ross and now represent

a key anthropological document for this area (on file with

AIATSIS). In August 2003, Ross and Smith, accompanied by

Syd Coulthard, visited Glen Thirsty (again by helicopter) to

make a more detailed assessment of the potential of Glen

Thirsty for future archaeological research. Syd Coulthard’s

account of Glen Thirsty has also been lodged on audio CD

with AIATSIS.

In August 2004, with better ground access to the area

(courtesy of Northern Territory Parks and Wildlife), we were able

to make a more extended search of Glen Thirsty. This showed

that archaeological remains were concentrated in three main

areas (Figure 2), all in the upper reaches of local drainage lines:

1. The fi rst is a cluster of sites around the margins of a sandy fl at,

between quartzite ridges, 500m west of Worrill’s Pass (Figure

2A). There are six rockshelters with rock art here (GT1-3, 7-

9). Three contain substantial assemblages of painted, printed

and stencilled art, and occupation deposit. These include the

painting Giles described as ‘Roman numerals’ (see Figure

10). Chipped stone artefacts and fragments of grindstones,

including seed-grinding implements, litter the sandy fl at.

2. The second cluster of sites (GT 4-6, 10-13) (Figure 2B)

is 2.6km west of Worrill’s Pass in the upper reaches of the

western-most channel. There are four small rockshelters

with painted and stencilled motifs here (GT1, 4, 6, 10, 11).

GT5 and GT12 are open engraving sites. GT13, a shallow

overhang, contains a small assemblage of engraved track

motifs and two hand stencils. The engravings at GT12 are on

ledges and open rocky terraces around a rock hole (44m-long,

6m-wide and 0.75m-deep), which would hold a considerable

volume of water. At the time of our visit in August 2004, just

a month after heavy local rains, the waterhole was already

dry (although the muds were still damp, attracting masses of

butterfl ies). We also noted extensive artefact scatters on rocky

terraces above the waterhole.

3. The third cluster of sites (Figure 2C) is 3km northwest of

Worrill’s Pass, along the upper reaches of the creek that fl ows

south through the pass. There are several waterholes in the

bed of the creek here, a worked outcrop of pedogenic silcrete,

and scatters of chipped stone artefacts on the terraces and

nearby fl ats.



There were few signs of Aboriginal occupation on the sand plain

adjacent to Giles’ well, or on dunes to the south.

Our fi eldwork in 2004 focused on four aspects of Glen Thirsty:

• archaeological excavation of two rockshelters (GT1 and

GT3);

• further recording of rock paintings and rock engravings,

especially in the western cluster;

• sampling of stick-nest rat middens to provide palaeo-

environmental data on the pre-European vegetation of the

area; and

• a search for remains of Giles’ 1872 camp.

A third rockshelter with archaeological deposit (GT2) was not

excavated at the request of the senior Aboriginal custodian (the late

Syd Coulthard) who was a member of the fi eld party. Excavation

methods followed Johnson (1979) and are described more fully in

Smith (1988). Recording of the paintings and engravings followed

the approach developed by Rosenfeld and Smith (2002), where

motif counts include compositions made up of multiple graphic

elements. This gives lower motif counts than methods that break

these down into individual graphic elements.

Archaeological Excavations

Glen Thirsty 1Glen Thirsty 1 is the largest of the rockshelters in this locality

(Figure 3). It faces north, opening directly onto the sand plain.

Structurally, the shelter is a cavity developed along joint lines

beneath horizontally bedded quartzite in the Winnall Beds (a

white or brown sandstone), to create a habitable shelter 10m-

long by 7m-wide, with a roof height of 2.5m (Figure 4). Several

large panels of paintings cover the walls and roof of the shelter.

Occupation debris, visible on the surface of the fl oor deposits,

includes small pieces of dark red ochre, large pieces of charcoal,

chipped stone artefacts, seed-grinding implements (including

a broken millstone and several intact mullers) and numerous

small fragments of broken grindstones. Outside the overhang,

occupation debris extends out from the shelter for about 10m.

Excavation of a small trench (2 x 1m² pits, designated F6

and F7 on a nominal site grid) showed a sedimentary sequence

approximately 1m-thick, consisting of an upper unit (20cm-

thick) with stone artefacts, fragmented bone, ochre, charcoal and

interleaved ashy lenses, overlying a unit of colluvial sand and roof

fall (60-80cm-thick), and bedrock. Tables S1-S3 (supplementary

information) provide details of sieve and granulometric data,

and depths of individual excavation units.

Stratigraphy and Sedimentary HistoryThe stratigraphy at Glen Thirsty 1 shows a mantle of late

Holocene occupation deposit covering a sandy soil, in which the

upper part of the profi le contains a few stone artefacts together

with evidence of widespread burning of local vegetation.

The deposits consist of three layers (Table 1, Figure 5). Layer

I is a loose light brown sand with large charcoal pieces (up to

20mm) and other occupation debris. This overlies a more

consolidated unit of ashy grey sand with intact hearths (Layer

II), representing the major occupation of the rockshelter. Layer

II overlays colluvial sands (Layer III), which grade from a black

sand (Munsell 7.5YR 2/3, pH 4.5) rich in fi nely comminuted

charcoal to brown (Munsell 7.5YR 3/3), becoming lighter with

depth (Layer V). The major source of sediments appears to be

weathering of local quartzite, both within the rockshelter and

along the foot of the strike ridge. The fi ne red aeolian sand

that mantles the sand plain in front of the rockshelter has not

signifi cantly contributed to the rockshelter deposits. A band of

roof spall and rock fall marked the top of the colluvial sands,

protecting deeper deposits from disturbance.

The black sand is unusual in Central Australia and requires

some explanation. At 0.35% the net organic carbon content of

Layer III (at 30cm below surface) is within the normal range

for arid zone soils (Jackson 1962:44). However, fl otation and

fi ltration of these sediments shows the carbon is in the form of

abundant fi ne charcoal particles (<100µm), suggesting burning

of local grassland immediately prior to major occupation of

the rockshelter.

Our excavations reached bedrock in F7. The F6 pit came

down on a large rock slab that stopped further excavation. Both

Figure 3 Excavations at Glen Thirsty 1, August 2004, looking southeast (L to R: Mike Smith and Sharon Overend) (Photograph: M.A. Smith).



Figure 4 (A) Plan of Glen Thirsty 1. A-A1 indicates position of cross-section shown in Figure 4B. (B) Site profi le.

A B

Site Layer Rocks/Rubble

(%)

Fine Gravel

(%)

Charcoal >3mm

(g/10kg)

6mm Lithics

(#/100g)

3mm Lithics

(#/100g)

Ochre >6mm

(g)

Bone >3mm (g/kg)

GT1 I & II: Main occupation unit 26 2 61.0 220 30 13.5 5.3

GT1 III: Black sand 66 4 1.3 30 1 – 0.2

GT1 V: Brown sand & rubble 53 7 – – – – –

GT3 I & II: Occupation unit 4 2 67.3 160 52 1.1 18.5

Table 1 Composition of the rockshelter deposits, Glen Thirsty 1 and 3. Figures for charcoal, lithics and bone are standardised by gross weight of excavated sediment.

Figure 5 Stratigraphic cross-section, west face of Trench F6/F7, Glen Thirsty 1. Layers I-VI are described in the text.

Site Layer Context Depth(cm)

Lab. No. 14C Age (years

BP)

Calibrated Age BP

1σ

Calibrated Age BP

2σ

Material Dated

GT1 II: Main occupation unit F7/3-1 16 Wk-16359 1300±38 1270-1090 1280-1070 Charcoal from hearth

GT1 III: Top of black sand F7/5 27-37 Wk-18696 1715±32 1610-1520 1700-1410 Detrital charcoal (6 & 3mm)

GT3 II: Occupation unit E2/3 6-13 Wk-18697 272±30 320-150 440-150 Detrital charcoal (>6mm)

GT3 II: Occupation unit E3/3-1 16 Wk-16358 235±35 300-150 320-10 Detrital charcoal in hollow on bedrock (6 & 3mm)

Table 2 Radiocarbon dates for Glen Thirsty 1 and 3. Radiocarbon ages were calibrated using OxCal 4.0 (Bronk Ramsay 1995, 2001) and the SHCal04 (McCormac et al. 2004) dataset.



bedrock and the rock slab were friable and deeply weathered,

so much so that excavation cut through their original exterior

surfaces into an orange sand representing in situ disintegration

of the sandstone (IV and VI, Figure 5). The rock slab at the

base of F6 can be matched to a prominent scar on the lip of

the rockshelter, providing a terminus post quem for the rock

paintings that now cover the scar.

ChronologyTwo radiocarbon samples were submitted from this site (Table

2): Wk-16359 (1300±38 BP) on charcoal from an intact hearth

(F7/3-1) in the middle of the ashy occupation unit (Layer II),

and Wk-18696 on detrital charcoal recovered by fl otation from

F7/5 to date the top of the black sand (1715±32 BP). The two

radiocarbon samples bracket the base of Layer II, showing that

it began to accumulate about 1500 years ago, consistent with the

presence of late Holocene implement types such as seed-grinding

implements, tula adze slugs, and geometric microliths. The base

of the rock slab in F6 rests on deposits that are estimated to date

between 2000–3000 BP.

Chipped Stone ArtefactsThe distribution of chipped stone artefacts is shown in Tables

1 and 3. The concentration of artefacts in Layers I and II is

striking. There are fewer artefacts in the underlying sands

(Layer III). Although most of these are in the top of the black

sand (F6/7 and F7/5), the presence of a band rocks and roof

spall seals the deposit at this level, and suggests that these are

likely to be in situ. Some corroboration of this is provided by

the presence of small amounts of bone and fi ne chipping debris

in these levels. The lowest artefacts are from the top of F7/6

(36cm below surface) where the black sand grades into lighter

sediments. Artefacts recovered from F6/6 represent material

from pockets on the surface of the rock slab, or displaced by

treadage into the orange sand that results from in situ decay

of this rock.

Chert or cherty silcrete is the dominant raw material,

making up 52% of this assemblage (Table 4). The most common

variety is mottled buff chert. Pedogenic silcrete (a yellow-grey

cherty silcrete) is available in worked outcrops 3km northeast

of the rockshelter, but is not represented signifi cantly in the

excavated assemblage. Local quartzite makes up the balance of

the artefacts (47%). The white chalcedony, and red or yellow

nodular cherts found in the Puritjarra assemblage (Smith 2006)

are rare at Glen Thirsty.

Tables 5 and 6 show the distribution of artefact classes and

tool types. The assemblage is dominated by small fl akes (up to

15mm) with comparatively few cores or debitage. Retouched

artefacts are expediently retouched (amorphous) implements

or are broken pieces of larger implements, often elements

snapped off a retouched edge. Tool types are characteristic of late

Holocene assemblages elsewhere in Central Australia, with tula

adzes (or tula slugs), geometric microliths, thumbnail scrapers

and endscrapers as the main formal artefacts.

Layer XU Chipped Stone Artefacts Grind-stones

Ochre Bone Nuts Charcoal

Total Weight

n Mean Weight

Artefact Density

Flakelet Density (3mm)

6mm 3mm 6mm 3mm 6 & 3mm

(g) (#) (g) (#/kg) (#/100g) (#) (g) (g/kg)

(g) (g/kg)

(g) (g/10kg)

Layer I F6/1 296.2 120 2.5 3.3 21 2 0 0 3.4 9.1 0.7 308

F6/2 233.6 155 1.5 3.0 23 0 1.6 0 9.5 10.7 0.5 147

Layer II F6/3 112.8 85 1.3 1.4 26 1 0 0 4.0 14.4 0 112

F6/4 171.5 122 1.4 1.7 25 0 0.4 0.2 3.5 10.6 0 85

F6/5 194.4 305 0.6 2.8 20 0 0 0 0.3 1.2 0 80

Layer IV F6/6 151.1 74 2.0 0.5 2 0 0 0 0 0 0 14

Layer III F6/7 7.9 2 4.0 0.1 1 0 0 0 0 0 0 2

Layer I F7/1 291.6 151 1.9 1.7 26 1 0.8 0 2.4 2.6 0 12

F7/2 441.0 301 1.5 2.8 32 0 0 0 1.1 1.5 0 10

Layer II F7/3 687.7 75 9.2 0.7 58 0 10.1 0.2 0 0.2 0 5

F7/4 645.5 277 2.3 2.6 34 0 0.6 0 0 0 0 3

Layer III F7/5 194.4 68 2.9 0.4 11 0 0 0 0 0.2 0 1

Layer V F7/6 56.9 1 56.9 0 1 0 0 0 0 0 0 3

F7/7 0 0 0 0 0 0 0 0 0 0 0 0

Layer VI F7/8 0 0 0 0 0 0 0 0 0 0 0 0

Table 3 Distribution of artefacts and occupation debris at Glen Thirsty 1. Data are for 6mm fraction unless indicated otherwise.

Site Layer n % Chert % Quartzite % Ironstone % ChalcedonyGT1 I & II Main occupation 557 52.2 47.4 0.2 0.2

GT3 I & II Occupation unit 294 66.0 29.9 0 4.1

Table 4 Raw material use at Glen Thirsty 1 and 3. Data are percentage frequency of artefacts of each raw material.



GrindstonesThe excavation recovered four pieces of grindstone, all except

one from the loose surface sand (Layer I). None of these can

be securely identifi ed as seed-grinding implements, although

millstones and mullers are found on the surface of the site, as

well as on the sand plain in front of the rockshelter.

OchrePieces of ochre were found throughout the ashy grey occupation

unit, with most coming from F6/4 and F7/3 (for details see Table S4,

supplementary information), a level directly dated by Wk-16359.

Several different ochres are represented. The most common (8 of

the 10 pieces recovered) is a fi ne-grained, bright red ochre. This

relatively homogenous pigment is found either as small subangular

fragments, or as intact pisoliths (up to 15mm diameter). The latter

suggests a source in Tertiary laterites or ferricretes, such as those

formed over the Goyder formation on the Watarrka plateau (Bagas

1988). F7/3-2 is a piece of ground ochre with the distinctive coarse-

grained texture, purple colour and metallic lustre of material from

the Ulpanyali quarry, 50km northeast of Glen Thirsty. A single

piece of fi ne-grained dark red micaceous ochre (possibly Karrku

ochre) was recovered from F7/1.

SeedsFragments of quondong nuts (Santalum acuminatum) are

present in the loose surface sand (Layer I), suggesting occupation

during September or October (when the fruits ripen, depending

on rain)(Table 3).

BoneTable 3 shows that bone is restricted to Layers I and II (except

for a single small piece in F7/5). The bone mainly consists of

fragments of postcranial bone <30mm-long, ranging from the

hard compact bone of large macropods to the epiphyses of small-

medium mammals, to complete bones of small rodents and

lizards. Identifi able elements include tooth fragments of medium-

large macropods (including tooth enamel from molars and lower

incisors), the maxilla of a rodent, reptile vertebral elements, and

the dentaries of small agamid (5) and scincid lizards (1). Of the

larger bone, some pieces are scorched, charred or calcined.

Glen Thirsty 3Excavations at rockshelter GT3 revealed a similar assemblage to

GT1, but one with a more limited chronology, which contrasts

with its comparatively rich assemblages of paintings and chipped

stone artefacts.

Glen Thirsty 3 is 370m northwest of GT1, on the northern

side of a low quartzite strike ridge. It is a low rockshelter, 12m-

long by 3m-wide, with a sloping roof 1–2m high (Figure 6),

facing out onto a mulga flat. The rear wall of the shelter has a

prominent frieze of paintings, now covered in dust. Exterior

deposits around the shelter are subject to gentle slope-wash,

and because of this have a relatively high concentration of

chipped stone artefacts (~500/m²). The only identified

types are tula adze slugs of which several were noted. Seed-

grinding implements and other grindstones are rare at

this site.

Table 5 Composition of the fl aked stone assemblage at Glen Thirsty 1. Data are numbers of artefacts (6mm sieve fraction) in each category. Wall cleaning is material recovered from F6/1 and /2 and F7/1 and 2.

Laye

r

XU

Geo

met

ric

Mic

roli

ths

Tula

Ad

ze S

lug

s

Th

um

bn

ail

Scr

aper

s

En

dsc

rap

ers

Am

orp

ho

us

Ret

ou

ched

Im

ple

men

ts

Layer I F6/1 0 2 0 0 10

F6/2 1 3 0 0 6

Layer II F6/3 0 1 0 1 2

F6/4 0 1 0 0 6

F6/5 1 1 1 0 9

Layer IV F6/6 1 0 0 0 2

Layer III F6/7 0 0 0 0 0

Layer I F7/1 0 2 0 0 6

F7/2 0 2 1 0 9

Layer II F7/3 1 4 1 2 12

F7/4 2 2 0 0 12

Layer III F7/5 0 0 0 0 3

Layer V F7/6 0 0 0 0 0

Layer I Wall cleaning

0 2 0 0 8

Total 6 20 3 3 85

Table 6 Temporal distribution of fl aked tools at Glen Thirsty 1. Data are numbers of artefacts (6mm sieve fraction) in each category.

Laye

r

XU

Ret

ou

ched

Art

efac

ts

Un

mo

difi

ed

Fla

kes

Co

res

Deb

itag

e

Tota

l

Layer I F6/1 12 38 0 75 125

F6/2 10 65 0 76 151

Layer II F6/3 4 36 0 43 83

F6/4 7 32 1 77 117

F6/5 12 86 0 182 280

Layer IV F6/6 3 39 0 30 72

Layer III F6/7 0 2 0 0 2

Layer I F7/1 9 56 0 80 145

F7/2 12 113 1 172 298

Layer II F7/3 20 122 1 157 300

F7/4 16 107 1 154 278

Layer III F7/5 3 46 0 19 68

Layer V F7/6 0 2 1 2 5

Layer I Wall cleaning

10 13 0 23 46

Total 118 757 5 1090 1970



Excavation of two 1m² pits (E2 and E3) showed that the

archaeological deposit forms a shallow wedge of sandy sediment

(20cm-thick) resting on bedrock. Tables S5-S7 (supplementary

information) provide details of sieve and granulometric data,

and depths of individual excavation units.

Stratigraphy and Sedimentary HistoryThe deposits consist of two layers (Table 1, Figure 7). Layer

I is a loose red aeolian sand (Munsell 5YR 4/8) (6cm-deep),

abundant in chipped stone artefacts, fragmented bone, charcoal

pieces and macropod scats. Layer II is a more compact form of

the same sediment becoming darker and more grey with depth,

with some small discrete lenses of red sand or white ash. In

both layers, the sediment matrix is aeolian sand. The sandy

loam on the mulga fl at in front of the shelter does not appear

to have contributed signifi cantly to the rockshelter deposits.

ChronologyTwo radiocarbon samples were submitted from this site (Table

2). Wk-16358 (235±35 BP) dates charcoal resting in a hollow on

bedrock (E3/3-1) at the base of the deposit. Wk-18697 (272±30

BP) dates a larger sample of detrital charcoal from E2/3, 4cm

above bedrock (6-13cm below surface). Calibration of these

radiocarbon ages indicates the deposit began to accumulate

between AD 1630-1800, suggesting that use of this shelter

overlaps Giles’ visit in 1872.

Chipped Stone ArtefactsThe distribution of chipped stone artefacts is shown in

Tables 1 and 7. The highest concentration is in Layer I, but

artefact density is high throughout the deposit, with more

fine debitage than at GT1. As at GT1, chert or cherty

silcrete is the dominant raw material, making up 66% of

this assemblage (Table 4). Again, the most common variety

is mottled buff chert. Local quartzite makes up 30% of the

artefacts, and there is also some chalcedony (4%). Tables 8

and 9 show the distribution of artefact classes and tool types.

The assemblage is dominated by small flakes (>15mm), with a

higher proportion of debitage than GT1. Retouched artefacts

are mainly expediently retouched (amorphous) implements

or are broken pieces of larger implements, often elements

snapped off a retouched edge. Tula adzes (or tula slugs),

geometric microliths, thumbnail scrapers and endscrapers

are the main formal artefacts.

GrindstonesOnly one grindstone was recovered during excavation. E3/1-1 is

a small fl ake off a grindstone.

OchreOnly three pieces of ochre were recovered from GT3 (Table S7,

supplementary information), all from unconsolidated surface

sediments (Layer I). Two of these are a fi ne-grained, bright red

ochre, matching the most common ochre at GT1. The remaining

piece is a friable orange sandy ochre, probably from local

sandstone beds.

SeedsFragments of quondong nuts (Santalum acuminatum) are

present in the upper 13cm of the profi le.

Figure 7 Stratigraphic cross-section, west face of Trench E2/E3, Glen Thirsty 3. Layers I and II are described in the text.

A B

Figure 6 (A) Plan of Glen Thirsty 3. A-A¹ indicates position of cross-section shown in Figure 6B. (B) Site profi le.



BoneFinely fragmented bone is common throughout this deposit

(Table 7). As with GT1, this mainly consists of fragments of

postcranial bone from small-medium-large mammals. The

few identifi able elements include tooth fragments of medium-

large macropods (including the crowns of molars), the claw of

a large macropod, rodent incisors, and parts of mandibles of

small mammals. Two bones, both segments of long bones from

medium-sized mammals, have distinct cut-marks.

The Rock ArtGiles’ descriptions of the rock art at Glen Thirsty (and at the ‘Tarn

of Auber’) provide the earliest written records of the rich corpus

of rock art in Central Australia. Glen Thirsty, he wrote, was ‘the

aboriginal national gallery of paintings and hieroglyphics’ (Giles

2000[1875]:69) and he was particularly struck by the large

emblemic motifs at these sites, commenting that ‘rude fi gures

of snakes were the principal objects, but hands, and devices

for shields were also conspicuous’ (2000[1875]:65). (His use of

the phrase ‘devices for shields’ suggests he read some of these

paintings as heraldic motifs).

Our survey identified 13 rock art sites: 11 with painted,

drawn, stencilled or printed art, and three with engravings.

Figure 8 illustrates the range of painted motifs. Table 10

gives the relative density of art at these sites, showing that

the eastern sites (especially GT1-3) have been the focus of

pigment art production, while the western sites contain all

the engravings.

Layer XU Chipped Stone Artefacts Grind-stones

Ochre Bone Nuts Charcoal

Total Weight

n Mean Weight

Artefact Density

Flakelet Density (3mm)

6mm 3mm 6mm 3mm 6 & 3mm

(g) # (g) #/kg #/100g # (g) (g/kg)

(g) (g/kg)

(g) g/10kg

Layer I E2/1 251.1 141 1.8 2.6 56 0 0 0.2 2.1 14.2 1.2 81.9

E2/2 97.3 86 1.1 1.4 39 0 0 0.2 1.3 18.5 0.6 92.2

Layer II E2/3 33.3 47 0.7 0.6 62 0 0 0 0.9 21.2 2.0 53.1

E2/4 21.7 17 1.3 0.5 18 0 0 0 0 9.3 0 15.6

Layer I E3/1 332 1.2 3.1 63 1 0.9 0 13.9 18.3 4.5 85.0

Layer II E3/2 132.9 86 1.5 1.0 66 0 0 0 5.2 23.6 0 99.6

E3/3 58.0 51 1.1 1.3 38 0 0 0 0.6 17.4 0 3.8

Table 7 Distribution of artefacts and occupation debris at Glen Thirsty 3. Data are for 6mm fraction unless indicated otherwise.

Layer XU Retouched Artefacts

Unmodifi ed Flakes

Cores Debitage Total

Layer I E2/1 9 33 0 99 141

E2/2 3 34 0 49 86

Layer II E2/3 1 23 0 23 47

E2/4 1 12 0 4 17

Layer I E3/1 11 99 0 222 332

Layer II E3/2 1 47 0 38 86

E3/3 2 25 0 24 51

Total 28 273 0 459 760

Table 8 Composition of the fl aked stone assemblage at Glen Thirsty 3. Data are numbers of artefacts (6mm sieve fraction) in each category.

Layer XU Geometric Microliths

Tula Adze Slugs

Thumbnail Scrapers

Endscrapers Amorphous Retouched Implements

Layer I E2/1 0 3 0 0 6

E2/2 1 0 0 1 1

Layer II E2/3 0 0 1 0 0

E2/4 0 1 0 0 0

Layer I E3/1 0 0 0 0 10

Layer II E3/2 0 0 1 1 0

E3/3 0 1 0 1 1

Total 1 5 2 2 18

Table 9 Temporal distribution of fl aked tools at Glen Thirsty 3. Data are numbers of artefacts (6mm sieve fraction) in each category.



Overall, this rock art complex provides a strong comparative

assemblage for those at Watarrka to the northeast (Frederick

1999; Rosenfeld 1990; Ross 2003; Smith and Rosenfeld 1992)

and Puritjarra to the north (Rosenfeld and Smith 2002; Ross

2003). Although the Glen Thirsty assemblage has a broader

range of motifs than either Puritjarra or Watarrka, the greater

proportion of the motif vocabulary, the colour range utilised

and the method of applying the pigment is similar. The context

and physical organisation of the sites at Glen Thirsty is also

similar to other rock art complexes in Central Australia (Gunn

1995, 2000) where each complex is associated with a water

source, however ephemeral, and each is made up of one or two

major sites with a number of smaller satellite sites containing

fewer motifs.

Glen Thirsty 1In addition to hand stencils and hand prints, the main motifs

at Glen Thirsty 1 include trails of macropod or possum tracks

(track lines or meanders), monochrome sinuous or bichrome/

white outline ‘snake’ motifs, and concentric circles (Table 11,

Figure 8). Most motifs are painted, though other techniques such

as stenciling, printing and drawing are also evident. Red (22)

or dark red (21) ochre were the most frequently used colours,

although white, cream, black, pink, yellow and grey pigments

were also recorded.

Large ‘snake’ motifs (~2m-long) dominant the painted

frieze at Glen Thirsty 1. Three large monochrome snakes,

painted as either white outline (silhouette) or solid red, occupy

the back wall of the shelter. Although anatomical details are

not shown, Peter Bullah identifi ed these as ‘kuniya’ or large

python. At the front of the shelter, on the overhang above the

excavation, a monochrome snake in dark red ochre has been

painted over an earlier bichrome white-outline/red infi ll snake

and appears to have been repainted on at least one occasion (as

it is interlaced with a bichrome white-outline ‘inverted T-shape’

motif) (Figure 9).

Other prominant paintings in white pigment include a large

concentric circle (58cm diameter) with four irregular rings

and a white-outline ‘lizard’ motif (identifi ed by Peter Bullah

as relating to the blue-tongue lizard [Tiliqua multifasciata], his

personal totem). The similarity in both the method of pigment

application and the colour and consistency of the ochre used

to produce these paintings suggests they represent a single

painting episode.

Macropod trails and tracks are represented by paired

parallel dashes, which meander along the rear wall of the

shelter following natural features. Peter Bullah identifi ed the

most prominent of these as mala (Lagorchestes hirsutus). He

identifi ed others made up of diagonally opposed feather-like

motifs as possum (Trichosurus vulpecula) tracks. Several longer

trails have been extended using different pigment, indicating

either repainting of these motifs or use of discrete batches of

pigment during their creation.

Hand stencils and hand prints have been placed individually

on panels, rather than in pairs or clusters. No hand prints or

stencils of children were recorded here or at other sites.

On several panels, fi ne drawings in charcoal form clusters of

short irregular vertical lines, or concentric arcs.

Relative SequenceThe superposition of motifs indicates changes in the composition

of this frieze over time, although there is no consistent temporal

sequence in the use of different coloured pigments.

Figure 8 Composite drawing illustrating the range of painted motifs at Glen Thirsty. (1-3, 6) White outline or bichrome silhouette motifs. (4-5) Anthropomorphic motifs. (7-12) Sinuous line (‘snake’) motifs, either white outline (7-8, 11), or monochrome. (13-15) Other complex motifs. (16-21) Trails of tracks, including bird (16), possum (19-20) and macropod (18, 21). (22-26) Circles including barred (22), concentric (23-25) and conjoined variants (26). (27-34) Fern motifs. (35-37) Irregular bichrome motifs. (38) Giles’ ‘Roman numerals’. (39-40) Y motifs.

Site 1 2 3 4 5 6 7 8 9 10 11 12 13Hand stencils 12 7 1 0 0 0 2 3 1 0 4 0 2

Hand prints 2 1 17 0 0 4 0 0 0 0 1 0 0

Paintings 63 116 76 2 0 0 2 0 3 21 17 0 0

Drawings 3 7 9 0 0 0 0 0 0 0 9 0 0

Engravings 0 0 0 0 2 0 0 0 0 0 0 272 3

Total 80 131 103 2 2 4 4 3 4 21 31 272 5

Table 10 Composition of rock art assemblages at Glen Thirsty. Data are motif counts by composition (ordered arrangements of several graphic elements).



• Yellow or white hand prints, and trails of macropod or possum

tracks in red or yellow ochre, represent the most recent art

at this site. Individual tracks within some trails are outlined

in charcoal, a practice associated with very recent rock art

assemblages in Central Australia (Frederick 1999; Ross 2003,

2005). The rough wash technique used to apply the pigment

on two large white motifs, and their placement over a pair

of yellow macropod tracks indicates that these motifs also

belong to this phase.

• Hand stencils and individual or paired macropod tracks

repeatedly occur under other motifs so are likely to form part

of the earliest assemblage at this site.

• The large ‘snake motifs’ represent a phase of art production lying

between the earliest and latest art at this site. Snake motifs overlie

hand stencils or large arc motifs but consistently occur under

hand prints and trails or lines of tracks. Some of these motifs

infl uenced the placement of later art. On the rear wall, a trail

of possum tracks runs parallel with a monochrome red snake,

suggesting this retained signifi cance into the recent past.

Glen Thirsty 2This shelter contains the greatest concentration of pigment art

at Glen Thirsty (Tables 10-11) and there are several striking

features of the art assemblage. The fi rst is the high proportion

of motifs that have been painted using two or more colours

(20 out of 131). Most have an outline, frequently painted in

thick white pigment with an infi ll of red, red/purple, or most

commonly, dark grey ochre. Most are large motifs; the largest of

these (~1m²) consists of six linked concentric circles painted in

black and white pigment (Figure 8:26).

Giles’ ‘Roman numerals’ are included in this group and

comprise a large v-shape and line, outlined in white and

infilled with parallel lines of white and dark grey pigment

(Figures 8:38, 10). Giles described this motif as ‘daubed over

with spots’ and ‘painted with red ochre’ (2000[1875]:65). As

there is no evidence of repainting since 1872, we think Giles

must have added these details from memory when revising

his journal for publication in his 1875 report, conflating other

details of the art panel (his original field journals are lost). For

instance, the entire surface of a lower panel is daubed with

spots of white paint.

The second feature is the presence of a diverse suite of

large, irregular, white motifs (>50cm at the highest or widest

point) (Figure 8:4-7, 10, 25, 35-37) not commonly found in art

assemblages further east (cf. Gunn 1995, 2000). The majority of

these have no recognisable iconic form and are best described as

irregular enclosed shapes (though there is one ‘snake’ motif, and

two others are anthropogenic in form).

White is the most frequently used pigment at this site (44

motifs). Dark red (40) and red (14) were also widely used, and

we recorded a variety of other colours.

Figure 9 Art panel at Glen Thirsty 1, showing a white-outline, black infi ll motif overlying a red monochrome snake. The latter has been painted over an earlier bichrome white-outline/red infi ll snake (lower right).

Category GT1 GT2 GT3 GT12Tracks

Macropod (1) 10 11 1 75

Bird 0 15 2 36

Trails (2) 28 37 31 29

Formal motifs: Basic forms

Sinuous line 3 2 7 0

Arc 2 0 0 6

Circle (3) 9 31 8 71

Fern 2 4 6 2

Amorphous shape 0 3 0 4

Y-shape 0 3 0 0

Linear 2 9 13 4

Formal motifs: Composite forms

Bichrome snake silhouettes

1 1 0 0

Bichrome amorphous shape

4 8 0 0

Bichrome linked circle 0 1 0 0

Other complex 0 1 1 3

Table 11 Recurring motifs in paintings and engravings at Glen Thirsty. Data are motif counts by composition rather than graphic element. (1) Includes single or paired tracks. (2) Track lines, clusters or meanders, including bird, macropod, possum and other. (3) Includes plain circles, concentric circles, barred ovals, conjoined circles and spirals.



Trails of tracks, including roughly executed trails of dots,

extend across the shelter wall. Diagonal or short vertical trails of

bird tracks contrast with longer meandering trails of macropod

tracks. Characteristically, the macropod tracks were produced

with fi ngers, using wet pigment to create a series of parallel

tracks, whereas some possum tracks are notable in the delicate

manner in which pigment has been applied.

Hand stencils are uncommon at Glen Thirsty 2, although

there are traces of seven, adult, dark red hand stencils on an

elevated panel.

Relative SequenceSuperposition analysis indicates that the suite of large bichrome

paintings and white outline motifs are amongst the most recent

produced at this site. Trails of tracks were added to the assemblage

over time, but generally (9 cases) underlie the large white motifs.

The active exfoliation on the surface of the rock art panels argues

against any great antiquity for the assemblage at this site.

Glen Thirsty 3This site has few formal motifs and an abundance of track lines,

hand prints, and charcoal drawings (Table 11).

• Trails of tracks constitute 33% of this assemblage. Bird tracks

typically form vertical trails, while trails of macropod tracks

meander diagonally or horizontally across the shelter wall.

• Bichrome concentric circles and white fern-like motifs

(averaging 30cm-high) are visually dominant amongst the

formal motifs. One of these motifs has been repainted using

the distinctive red/purple Ulpanyali ochre. Another larger

motif consists of an oval with fern attached, painted in yellow

pigment and outlined in black.

• We recorded a single faded hand stencil and 17 dark red hand

prints (14 left hand, three right hand), all produced using

the same dark red pigment and all representing the same (or

similar) adult hand.

• This assemblage also includes a series of fine vertical

lines, drawn in charcoal. These are widespread in western

Central Australia (Ross 2003:223-224, 2005) and although

often dismissed as graffiti by rock art recorders in the

past, there is growing evidence that they reflect a more

structured practice.

Relative SequenceHand prints and charcoal drawings are the most recent motifs

added to the assemblage at Glen Thirsty 3 as they are superposed

on all motifs with which they overlap. Fine charcoal lines have

also been used to outline older possum and bird tracks. Most

tracks and trails are earlier in the superposition sequence and in

turn overlay fern motifs and concentric circles.

The Engravings at Glen Thirsty 12The engraved assemblage at Glen Thirsty 12 consists of 272

shallow pecked engravings on rock surfaces surrounding

the adjacent waterhole. While most are placed on horizontal

slabs on terraces running for 150m along the northern side of

the waterhole, other engravings occur on the vertical faces of

boulders on the southern side.

All motifs are relatively small in size and most rock faces have

just a few motifs engraved into them. The range of motif classes

at Glen Thirsty 12 (Table 11) is similar to that found at other

Central Australian engraving sites. At Glen Thirsty, there is a

predominance of bird and macropod tracks – either as individual

tracks, paired tracks, or short trails, or as series of parallel dots.

Pecked circles are small (<15cm diameter) and include plain

circles, barred circles, concentric circles, spirals, in-fi lled circles,

ovals and barred ovals. No fi gurative motifs are present but more

complex designs include a small amorphous maze.

Several factors indicate that the engravings were produced

over a relatively short period of time (possibly representing a

discrete pulse of production): style and technique are consistent

across the site; there is no superpositioning of engraved motifs

on these panels; they are uniformly and only lightly patinated.

In addition to six shallow seed-grinding grooves, there

are numerous small abraded patches (28) amongst the art

panels (Figure S1, supplementary information), similar to

those found at recent rock art complexes in the central ranges

(Ross 2003:156-157). At Glen Thirsty, the abraded patches are

comparatively small (averaging 44cm x 29cm) and are often

juxtaposed with engravings. In most cases, the superposition

order could not be reliably determined: some abraded patches

were superimposed on engravings (3 examples) and one

appeared to underlie an engraving. Mountford (1976:127; see

also Strehlow 1947:17) recorded Aboriginal men in the Western

Desert rubbing particular rock surfaces with hand-held stones

as part of the reactivation of totemic sites. At Glen Thirsty,

the size, placement and general context of abraded patches

point to a similar practice, perhaps associated with both initial

production and subsequent activation of the engravings. The

clusters of peck marks (20) on many engraved surfaces may

refl ect a similar practice.

Chronology of Art ProductionThe occupation history of the two excavated rockshelters provides

a broad temporal framework for the pigment art at Glen Thirsty,

suggesting a maximum age of ~1500 BP for the art panels we

see today. Correlation with sedimentary records is provided by

several lines of evidence:

Figure 10 The late Syd Coulthard at Glen Thirsty 2, August 2003. The painted motifs include those described by Giles in October 1872 as

‘Roman numerals’ (Photograph: M.A. Smith).



• Fragments of red or dark red, or Ulpanyali ochre, similar to

the pigments used in the paintings and stencils are found

throughout the occupation deposit at GT1 and 3.

• The fallen rock slab exposed in the excavations at GT1

provides an estimated terminus post quem of 2000–3000 BP

for paintings and stencils on the related scar on the overhang.

These include hand stencils that are amongst the earliest

motifs produced at the site.

• There is broad agreement regarding the relative age of the sites,

comparing occupation histories and phases of art production.

For example, the rock art frieze at GT1 has the greatest

intrinsic time depth of any of the 11 sites we examined – and

this site also has the longest 14C chronology.

Within the last 1500 years, the Glen Thirsty sites show an

internally-consistent series of changes in style and composition

in the pigment art. Hand stencils and trails (mainly track lines

or meanders) appear to have been the earliest compositions.

These were followed by production of large emblemic motifs,

including the white-outline silhouette designs at Glen Thirsty

1 and 2. The latest phase of art production at the Glen Thirsty

sites consists of hand prints, charcoal drawings and paintings

in white ochre. This sequence parallels similar changes

at Puritjarra (to the north) during the last millennium

(Rosenfeld and Smith 2002), especially the shift towards large

emblemic motifs.

The engravings at Glen Thirsty are of no great antiquity

and may be contemporary with the pigment art. The circles

and tracks use the same stylistic conventions as the pigment

art at Glen Thirsty. The shallow pecking technique used to

produce motifs at Glen Thirsty 12 is similar to that used to

produce many of the more recent engraved motifs across

Central Australia (Ross 2002, 2003). Similarly, the range of

motif classes recorded, the lack of patination, the absence of

pecked pits, and the abundance of abraded areas, all suggest a

recent age for this assemblage.

Historical ArchaeologyNo surviving trace of Giles’ 1872 encampment was located.

There are few trees of suffi cient age and size to carry an old blaze.

Extensive transects with a metal detector in the area south and

southwest of Worrill’s Pass, and on the fl at south of the rock

holes, failed to locate any metal items. It seems likely that Giles’

use of Glen Thirsty was so light that little repair or discard of non-

perishable objects such as horseshoes, or nails occurred. No metal

objects were detected in the sandy fl oors of local rockshelters.

Vegetation HistoryDirect palaeoenvironmental evidence for the region is sparse

(especially for the last few millennia). The phytolith record at

Puritjarra (Bowdery 1995, 1998) shows that the mid-Holocene

saw a decline in regional grass levels after 5000 BP, which only

recovered again after about 1500 BP as the modern vegetation

took shape. Assuming that periods of high representation of

Poaceae in the phytolith record refl ect better summer rainfall

(which is likely), the last millennium was a comparatively

favourable period for human settlement in Central Australia

(Bowdery 1995:Figure 6.1). This is supported by a study of

palaeofl oods on the Todd River, Central Australia, which shows

frequent high-magnitude fl oods during the last 1500 years, with

identifi ed fl ood phases at 400 BP, 600 BP, 1000 BP and 2100 BP

(Bourke 1998:131-134).

In Central Australia, stick-nest rat (Leporillus sp.) middens

have also been used as a source of palaeoecological data. Berry

(1991) sampled a midden at Kathleen Springs in Watarrka-Kings

Canyon National Park, with results indicating little local vegetation

change between 3500–1700 BP. Other Leporillus sp. middens, in

the eastern MacDonnell Ranges show a marked late Holocene

increase in tree and shrub taxa beginning sometime between

2510 and 850 BP, at the expense of herbs and grasses (Webeck

and Pearson 2005). To test if there were comparable changes at

Glen Thirsty, Webeck sampled four local Leporillus sp. middens

in 2004. This work has yet to be completed, but preliminary

results indicate that none of these middens date earlier than 430

BP and that over this period there is little apparent change in

local vegetation.

Discussion and Interpretation

The History of Glen ThirstyOur archaeological investigations give a picture of people using

a small soakage well on the northern rim of Lake Amadeus.

Despite our initial expectation that there would be several

phases of major occupation at a site in this location, separated

by periods of intermittent use or abandonment, we found that

substantive occupation of Glen Thirsty only took place during

the last 1500 years.

The presence of chipped stone artefacts in earlier deposits

points to intermittent use of the area prior to this, but as visitation

rather than colonisation. Some early use of the Glen Thirsty

locale would be expected given its proximity to the George Gill

Range with its springs and waterholes, and to Puritjarra with its

long record of late Pleistocene occupation to the north. But Glen

Thirsty itself only appears to have become an important focus of

occupation after 1500 BP.

Analysis of the rock art assemblage – both engravings and

paintings – leads us to similar conclusions: the rock engravings

surrounding the waterhole are uniformly late prehistoric in their

style and condition. There are no older weathered engravings to

suggest that an earlier pulse of occupation in the Glen Thirsty

area has been missed in the excavations. Other evidence supports

an association between the paintings at Glen Thirsty 1 and 3 and

the dated occupation deposits at these sites.

Several lines of evidence independently suggest establishment

of a new cultural and economic landscape at Glen Thirsty around

1500 BP, although much necessarily remains speculative:

• Major use of Glen Thirsty 1 is preceded by widespread burning

of local vegetation, coinciding with the fi rst appearance of

stone artefacts in these deposits. We have no data on whether

this refl ects human fi ring of local grassland (as Giles saw in

1872), or whether it represents large wildfi res associated with

increased fuel build up during years of exceptional rainfall (as

Central Australia experienced in 2002-2003). But increased

wildfi res and increased occupation of Glen Thirsty would

both be consequences of stronger summer rainfall.

• The rock engravings appear to represent a short-lived pulse of

art production, consistent with the establishment of a more



sustained presence of people in the area and the assertion of

new rights and associations to this place. This phase of art

production appears to have been followed by an ‘operational’

phase during which the engravings were rubbed and abraded,

presumably to reactivate them.

• An initial generic phase of painting and stenciling is followed

by the painting of large site-specifi c iconic motifs at Glen

Thirsty 1 and 2.

• Both the rock art and occupation histories of these sites

suggest a gradual expansion of use of the area, with this use

extending outwards from the initial focal areas around the

waterhole (Glen Thirsty 12) and Glen Thirsty 1.

Once established, occupation of Glen Thirsty over the last

millennium represents seasonal use of these desert lowlands,

whenever rain reactivated the well or recharged local rock

holes. Giles saw evidence of people using the well and firing

the surrounding country in October 1872, after heavy rain

had allowed people to disperse out into the spinifex and sand

hill country. The presence of quondong nuts in excavations

at Glen Thirsty also indicates some use of these sites in

September or October.

Regional ContextChanges during the last millennium in Central Australia provide

the regional context for developments at Glen Thirsty. To the

north of Glen Thirsty, on the rim of the Amadeus Basin, both

Puritjarra rockshelter and the Tjungkupu sites were occupied

more intensively after 800–1000 BP. To the northeast, the major

phase of use of Wanmara, a spring site in the George Gill Range,

began c.1300 BP. In the main ranges, these changes also involved

an expansion of Aboriginal settlement into the lower reaches of

catchments, with increasing use of fl oodplains, ephemeral waters

and sites on sand plain and valley fl oors (Thorley 1998a:317,

2001), greater use of grass and acacia seeds in the subsistence

economy, and the budding off of clan estates into marginal areas

(Smith 1988:325-332, 1996:70) (as also may have been the case

at Glen Thirsty).

Although growing population pressure and shifts in patterns

of land-use and economy would have provided much of the

impetus for settlement of the Glen Thirsty area, the timing of this

move is likely to have been constrained by climatic factors. Small

shifts in the amplitude, frequency or seasonality of rainfall are

important in determining levels of access to these desert lowlands

(as the historical record for Glen Thirsty shows). As a rain-fed

well in the lower part of the Amadeus Basin, Glen Thirsty is also

more sensitive to shifts in palaeoclimate than archaeological

sites on the Amadeus rim (such as Puritjarra and Tjungkupu)

or in the main ranges (such as Kulpi Mara, see Thorley 1998b).

In this context, any improvement in summer rainfall after 1500

BP (evident in vegetation and palaeofl ood records) would have

been important in recharging ephemeral wells and waterholes,

underpinning more sustained use of this area. Poor as it was, the

native well found by Giles was the key to development of Glen

Thirsty as a focal point for rock art and Aboriginal occupation in

this part of the Amadeus Basin.

At any one time, we can expect there to be core and peripheral

areas of occupation in the Australian arid zone, with the

boundaries changing as colonisation proceeds and people’s

ability to use the desert landscape improves (cf. Veth 1993). The

archaeology of Glen Thirsty exemplifi es changing relationships

between the ranges and desert lowlands during the last 1500

years, and shows that colonisation of the desert was not a discrete

event but a long-term process that also involved consolidation of

settlement in marginal areas in the last few thousand years.

Supplementary InformationSupplementary information for this article is available online at

www.australianarchaeologicalassociation.com.au.

AcknowledgementsThis project was supported by grants from AIATSIS (G2003/6832)

and the Fund of the Dean of Arts and Humanities, University of

New England, and was undertaken with permits from the Central

Land Council and the Northern Territory Heritage Branch. In

the fi eld, we enjoyed the companionship of Peter Bullah and

Syd Coulthardt (both now deceased). Logistical support for

the 2004 excavations was provided by the Northern Territory

Parks and Wildlife Service at Watarrka-Kings Canyon National

Park and we thank Darren Larcombe and Andrew Bridges for

organising this. Scott Mitchell and the Strehlow Research Centre

generously provided bench space for preliminary analysis of

excavated fi nds. Kim Webeck undertook analysis of the stick-

nest rat middens and AMS 14C dates for these were provided

by Stuart Pearson (Land and Water Australia). Stephen Cotter

determined the organic carbon content of the sediments at Glen

Thirsty 1. Figures were drawn by Kay Dancey (Cartography,

Australian National University) and the authors. We thank all of

the above and especially our fi eld and support crew during the

2004 fi eldwork: Danny Barrow, Tracy Blackney, Glenn Blackney,

Shannon Carne, Syd Coulthard, Rodney Coulthard, Phillip

Driffi n, Sharon Overend, Sandy Walters and Kim Webeck.

ReferencesBagas, L. 1988 Geology of Kings Canyon National Park, Northern Territory.

Geological Survey Report 4. Darwin: Department of Mines and Energy.

Berry, S.L. 1991 The potential of fossil mammal middens as indicators of vegetation

history in Central Australia. Australian Journal of Botany 39:305-313.

Bourke, M.C.A. 1998 Fluvial Geomorphology and Paleofl oods in Arid Central

Australia. Unpublished PhD thesis, Australian National University, Canberra.

Bowdery, D.E. 1995 Phytolith Analysis Applied to Archaeological Sites in the

Australian Arid Zone. Unpublished PhD thesis, Department of Archaeology

and Anthropology, Australian National University, Canberra.

Bowdery, D.E. 1998 Phytolith Analysis Applied to Pleistocene-Holocene

Archaeological Sites in the Australian Arid Zone. British Archaeological

Reports Series 695. Oxford: Archaeopress.


OxCal. program. Radiocarbon 37(2):425-430.


OxCal, Radiocarbon 43(2A):355-363.

Frederick, U.K. 1999 At the centre of it all: Constructing contact through the rock

art of Watarrka National Park, Central Australia. Archaeology in Oceania

34(3):132-144.

Giles, E. 2000[1875] Giles’s Explorations 1872-76: South Australian Parliamentary

Papers 1872-76. Australian Parliamentary Editions 2. Adelaide: Friends of the

State Library of South Australia, Adelaide.

Gunn, R.G. 1995 Regional patterning in the Aboriginal rock art in Central Australia:

A preliminary report. Rock Art Research 12:117-128.



Gunn, R.G. 2000 Central Australian rock art: A second report. Rock Art Research

17:111-126.

Jacobson, G., G.C. Lau, P.S. McDonald and J. Jankowski 1989 Hydrogeology and

Groundwater Resources of the Lake Amadeus and Ayers Rock Region, Northern

Territory. Bureau of Mineral Resources, Geology and Geophysics Bulletin 230.

Canberra: Australian Government Publishing Service.

Jackson, E.A. 1962 Soil Studies in Central Australia: Alice Springs-Hermannsburg-

Rodinga Areas. Melbourne: CSIRO.

Johnson, I. 1979 The Getting of Data: A Case Study from the Recent Industries of

Australia. Unpublished PhD thesis, Department of Prehistory, Research School

of Pacifi c Studies, Australian National University, Canberra.

Maurice, Justice 1989 Lake Amadeus Land Claim, Report by the Aboriginal Land

Commissioner to the Minister of Aboriginal Affairs and the Administrator of

the Northern Territory. Canberra: Australian Government Publishing Service.


Reimer 2004 SHCAL04 southern hemisphere calibration, 0-11.0 cal kyr BP.

Radiocarbon 46(3):1087-1092.

Mountford, C.P. 1976 Nomads of the Australian Desert. Adelaide: Rigby Limited.

Murray, W.R. 1904 Explorations by R. T. Maurice – Fowlers Bay to Cambridge Gulf.

South Australian Parliamentary Paper 43.

Rosenfeld, A. 1990 Rock Art in Watarrka National Park. Unpublished report to the

Conservation Commission of the Northern Territory, Alice Springs.

Rosenfeld, A. and M.A. Smith 2002 Rock-art and the history of Puritjarra rock

shelter, Cleland Hills, Central Australia. Proceedings of the Prehistoric Society

68:103-124.

Ross, J. 2002 Rocking the boundaries, scratching the surface: An analysis of the

relationship between paintings and engravings in the Central Australian

arid zone. In S. Ulm, C. Westcott, J. Reid, A. Ross, I. Lilley, J. Prangnell and

L. Kirkwood (eds), Barriers Borders Boundaries: Proceedings of the 2001

Australian Archaeological Association Annual Conference, pp.83-90. Tempus

7. Brisbane: Anthropology Museum, University of Queensland.

Ross, J. 2003 Rock Art, Ritual and Relationships: An Archaeological Analysis of

Rock Art from the Central Australian Arid Zone. Unpublished PhD thesis,

School of Human and Environmental Studies, University of New England,

Armidale.

Ross, J. 2005 Rock art of the Red Centre. In M.A. Smith and P. Hesse (eds), 23°

South: Archaeology and Environmental History of the Southern Deserts, pp.217-

230. Canberra: National Museum of Australia Press.

Smith, M.A. 1988 The Pattern and Timing of Prehistoric Settlement in Central

Australia. Unpublished PhD thesis, Department of Prehistory and Archaeology,

University of New England, Armidale.

Smith, M.A. 1996 Prehistory and human ecology in Central Australia: An

archaeological perspective. In S.R. Morton and D.J. Mulvaney (eds), Exploring

Central Australia: Society, the Environment and the 1894 Horn Expedition,

pp.61-73. Chipping Norton, NSW: Surrey Beatty and Sons.

Smith, M.A. 2005 ‘Peopling’ the Cleland Hills: Aboriginal History in Western

Central Australia, 1850-1980. Aboriginal History Monograph 12. Canberra:

Aboriginal History Inc.

Smith, M.A. 2006 Characterising late Pleistocene and Holocene stone artefact

assemblages from Puritjarra rock shelter: A long sequence from the Australian

desert. Records of the Australian Museum 58:371-410.

Smith, M.A. and A. Rosenfeld 1992 Archaeological Sites in Watarrka National Park:

The Northern Sector of the Plateau. Unpublished report to the Conservation

Commission of the Northern Territory, Alice Springs.

Smith, M.A. and J. Ross 2008 What happened at 1500-1000 BP in Central Australia?

Timing, impact, and archaeological signatures. The Holocene 18(3):387-396.

Strehlow, T.G.H. 1947 Aranda Traditions. Melbourne: Melbourne University Press.

Terry, M. 1937 Sand and Sun: Two Gold-Hunting Expeditions with Camels in Dry

Lands of Central Australia. London: Michael Joseph.

Thorley, P.B. 1998a Shifting Location, Shifting Scale: Towards an Archaeology of

Place in the Palmer River Catchment, Central Australia. Unpublished PhD

thesis, Northern Territory University, Darwin.

Thorley, P.B. 1998b Pleistocene settlement in the Australian arid zone: Occupation

of an inland riverine landscape in the central Australian ranges. Antiquity

72:34-45.

Thorley, P.B. 1999 Regional archaeological research in the Palmer River catchment.

Australian Aboriginal Studies 2:62-68.

Thorley, P.B. 2001 Uncertain supplies: Water availability and regional archaeological

structure in the Palmer River catchment, Central Australia. Archaeology in

Oceania 36(1):1-14.

Veth, P.M. 1993 Islands in the Interior: The Dynamics of Prehistoric Adaptations

within the Arid Zone of Australia. Archaeology Series 3. Ann Arbor, MI:

International Monographs in Prehistory.

Veth, P. 2005 Between the desert and the sea: Archaeologies of the Western

Desert and Pilbara regions, Australia. In M.A. Smith and P. Hesse (eds) 23°S:

Archaeology and Environmental History of the Southern Deserts, pp.132-141.

Canberra: National Museum of Australia Press.

Webeck, K. and S.G. Pearson 2005 Stick-nest rat middens and a late-Holocene

record of White Range, Central Australia. The Holocene 15:466-471.

Willshire, W.H. 1895 A Thrilling Tale of Real Life in the Wilds of Australia by

Mounted Constable, First-Class, Offi cer in Charge of Native Police. Adelaide:

Frearson and Brother.


IntroductionThe use of geophysical techniques as an aid to archaeological

investigations has become common-place, however these methods

have only occasionally been applied in Indigenous Australian

archaeology. This is despite recognition (and recommendations)

since the 1970s that such approaches have the potential to yield

positive results in such contexts (e.g. Connah et al. 1976; Stanley

1983; Stanley and Green 1976). Australian archaeologists have

perhaps been reluctant to embrace these techniques because of their

perceived high cost (both of equipment and specialist staff) and the

subtle nature of subsurface Indigenous sites as geophysical targets.

Nevertheless, there have been a number of recent applications of

these techniques in Australia, particularly in relation to burial and

hearth sites. We report the results of a pilot study conducted in

northwest Queensland. This study aimed to test the applicability of

geophysical methods being routinely employed to locate a variety

of open site features (particularly hearths and middens) as part of

reconnaissance surveys. While not being entirely successful, this

study demonstrated that certain archaeological features can be

readily identifi ed using geophysical techniques, though further

research and trials should be carried out to refi ne the uses of these

techniques to allow their more widespread applicability.

Characteristics of Hearth and Midden SitesHeat retainer hearths are ubiquitous in many parts of Australia,

typically appearing in surface exposures as small mounded

features with a locally available raw material – typically stone, clay

or termite mound – used as the heat retaining source. While the

majority of dated hearths have proved to be Holocene in age (e.g.

Holdaway et al. 2005; Robins 1996; Wallis et al. 2004) hearths of

greater age have been dated (e.g. Allen 1998; Smith et al. 1991; Veth

et al. 1990). Their widespread occurrence and ease of dating means

that hearths can be extremely useful for establishing chronologies

in parts of Australia where few other such possibilities exist.

Unfortunately, many decades of cattle and sheep grazing have had

a negative effect on the integrity of hearths, with heat retainers

sometimes so dispersed that the primary site location can no

longer be ascertained. Increased erosion rates caused by ungulates

have also accelerated exposure of such sites.

Historical documents reveal that middens composed

predominantly of freshwater mussel shells were once another

relatively common site type along watercourses in inland Australia.

Such sites also afford archaeologists abundant opportunities to

establish regional chronologies, as well as to examine questions

related to subsistence strategies, seasonality and resource use. Like

hearths, middens exposed at the ground surface are vulnerable

to physical destruction through stock treadage, and the chances

of fi nding such sites intact is therefore substantially reduced

even when other extensive evidence of Indigenous occupation

occurs (e.g. Crothers 1997; Wallis 2007; Wallis et al. 2004). Where

middens are found, their surface expression is often minimal,

with the shell being highly fragmented and dispersed and only

the subsurface shell material appearing to be intact (e.g. Wallis

2007). Consequently, such sites are often diffi cult to identify

using standard surface survey methods.

Both hearths and middens are important components of the

inland archaeological record, but both site types are prone to

destruction when exposed at the ground surface. This ongoing

destruction means they are an urgent contemporary heritage

management concern. The ability to identify such sites in an intact

subsurface context before they have been exposed and disturbed

or destroyed would greatly assist archaeologists and heritage

managers. Geophysical techniques such as electromagnetic

induction (EMI) and magnetometry provide a possible means

of accomplishing this.

EMI and Magnetometry TechniquesMagnetometry has a long history of use in European and

North American archaeology. It measures local perturbations in

the earth’s magnetic fi eld caused by accumulations of ferrous

material which may be from an anthropogenic or a geological

origin (Reynolds 1997). There are a variety of sensor types, sensor

confi gurations and survey methodologies for magnetometry

which are variously employed depending on the target material

and survey budget.

Electromagnetic induction (EMI) typically measures two

components of an induced magnetic fi eld: the quadrature

phase, which is linearly related to the ground conductivity, and

the in-phase component of the induced magnetic fi eld, which

measures magnetic susceptibility. Because each measurement

is not solely dependent on ferrous material, EMI is capable of

detecting a wide range of features including soil type, sediment

type, bedrock location or presence of cultural material, and has

been applied with success at archaeological sites for a variety of

tasks (Kvamme 2003). Furthermore, EMI can identify changes to

soil conductivity caused by both buried objects and associated

sedimentary disturbance (Nobes 2000:716; Nobes and Tyndall

1995:266).

SHORT REPORTS

Trialing Geophysical Techniques in the Identifi cation of Open Indigenous Sites in Australia: A Case Study from Inland Northwest QueenslandIan Moffat1,2, Lynley A. Wallis2, Alice Beale3 and Darren Kynuna4

1 Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia

2 Department of Archaeology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia

3 Department of Anthropology, Western Australian Museum, Perth Cultural Centre, James Street, Perth, WA 6000, Australia

4 Woolgar Valley Aboriginal Corporation, c/- 11 Harris Street, Richmond, QLD 4822, Australia


Short Reports

Detecting Hearths and MiddensThe results of fi res, including hearths, have been a particular

target of magnetometer investigations as burning creates

magnetic anomalies either through the enhancement of the

soil’s magnetic susceptibility (e.g. Dalan and Banerjee 1998)

and/or the contribution of wood ash (Linford and Canti 2001;

McClean and Kean 1993; Peters et al. 2001). The type of anomaly

produced by an intense, constrained fi re such as associated with

a hearth is quite different from that associated with bushfi res,

where ‘the very low thermal conductivity of the ground usually

results in uniform and insignifi cantly weak magnetizations’

(Stanley and Green 1976:55). Hence there is little possibility that

the anomalies likely to be produced by bushfi res and hearths

could be confused.

Likewise, the physical properties of middens make them

amenable to subsurface detection using non-invasive geophysical

techniques. Direct detection of midden material is based on

its difference from background levels of response using such

techniques as magnetometry, EMI, ground penetrating radar,

direct current resistivity or seismic methods (Steinberg et al.

2007; Whiting et al. 2001). Furthermore, the relationship that

often exists between middens and burning has been exploited

to locate middens in the same manner as is used for hearths

(e.g. Frederick and Abbot 1992).

Fanning et al. (2005) recently reported a pilot study in

western New South Wales using a gradiometer (a multisensor

magnetometer) to confi rm that hearths produced a magnetic

signal. They demonstrated that these features produce a

magnetic response substantially higher than that of the

surrounding ground surface, confi rming the experimental

fi ndings of others (e.g. Linford and Canti 2001). Their results

suggested a wider application for these methods; however, the

gradiometer was used only to assess the presence and magnitude

of the magnetic response from known hearths rather than as a

tool to locate them. Because the methodology employed did not

interrogate areas away from known hearths or incorporate any

positioning information it is unsuitable to test the application

of magnetometry as a reconnaissance tool. Nevertheless the

results suggested that with an alternative survey methodology

a gradiometer might well be a useful tool to detect previously

unlocated hearths.

A Pilot Study in Inland Northwest QueenslandIn order to assess whether EMI and magnetometry techniques

could be successfully applied in an open survey context, a pilot

study was undertaken in an area known to contain hearths

and middens (Domett et al. 2006; Wallis 2007), as well as the

partial skeletal remains of an ‘Old Person’ (the preferred term

for a burial) wrapped in paperbark which had been recently

reinterred in the vicinity. These archaeological features, along

with non-cultural mudstone eroding at the surface and

substantial erosion gullies, provided an ideal study site to test

the potential of geophysical techniques. The hypothesis was

that if these techniques could successfully relocate and identify

such known features, they might also be employed in the search

for such sites below the surface where a surface expression was

not visible.

The study area is located approximately 100km south of

Richmond in inland northwest Queensland, in a region of

gently fl at to undulating plains of low relief. The regional

geology is dominated by surface exposures of Allaru Mudstone,

with small outcrops of siliceous pebbles, gravelly deposits and

silcrete occurring on terraces and low ridges. Major watercourses

and creeks are dominated by deep, fi ne-textured Quaternary

alluvium consisting of sand, silt and clay, with the dominant soil

type being calcareous black clay.

MethodologyA 50m x 50m survey grid was established over an area

encompassing a range of archaeological and non-cultural

features, using an automatic level and survey tapes (Figure 1).

Survey lines were located within this grid using tapes, with station

locations collected using a Garmin 12XL navigation GPS with an

external antenna. The geophysical surveys were conducted over

the course of a single day using a line spacing of 1m orientated

in an east-west direction.

Both magnetometry and EMI techniques were employed,

chosen on the basis of their inexpensive nature, wide availability,

ease of execution and the nature of the anomalies we expected to

encounter. Data were collected with:

• A Geometrics G-856 proton precession single sensor

magnetometer tuned to a background level of 40000 nT.

Data were automatically collected along each survey line at

intervals of 5 seconds while the operator walked at a slow,

constant rate, with the sensor oriented north at a consistent

height.

• A Geophex GEM-2 EMI instrument. Data were collected at a

rate of 10 per second at frequencies of at 7875 Hz, 17575 Hz,

26275 Hz, 35275 Hz and 47975 Hz (with higher frequencies

representing shallower depths of penetration for the same

ground conductivity).

All collected data were gridded with MagPick software using

a spline interpolation (Smith and Wessel 1990) with an X and

Y interval of 0.1, a tension of 0.25 for 4000 iterations with a

convergence limit of 0.1 using the highest and lowest data values

as data limits. Results are displayed as simple contour maps with

250 non-equalised colour points and overlain contours.

Figure 1 Plan showing recorded surface features within the 50m x 50m survey grid.


Short Reports

ResultsNo in situ archaeological material, such as known hearths or

middens, were reliably detected using either of the techniques

employed in the survey area (Figures 2-3). However, the

reburied skeletal material was detected using both techniques

(although only at some frequencies of EMI) suggesting a

significant disturbance to the physical properties of the soil

in this area caused by the digging of the reburial grave. Non-

archaeological features were somewhat better identified using

the geophysical techniques. The eroded area in the northeast

of the survey area is shown in all EMI frequencies, but is

most evident in the higher frequency maps. Exposed bedrock

appears detectable, although poorly spatially resolved in the

higher EMI frequencies. Several magnetometer anomalies not

associated with any obvious causes at ground surface were

also observed. These anomalies have not yet been ground-

truthed and therefore may or may not represent subsurface

archaeological features.

In summary, using a routine field survey strategy, the

trialled methods did not successfully locate surface hearths

and middens in the study area despite there being a previously

established correlation between these features and a detectable

geophysical response. One possible reason is that the survey

methodology was not sufficiently robust to locate these

features reliably.

DiscussionAs noted elsewhere (Connah et al. 1976:153), successfully

identifying targets of this nature will depend largely on

the distance between survey transects. We have shown that

attempts to identify hearths using a survey grid spacing of

1m will not result in the universal detection of the features.

To be confident of identifying hearths in open contexts

survey transects would need to be carried out every 10–20cm

(Connah et al. 1976:153; Stanley 1983:84). This suggested

survey density was not used in our survey because decreasing

the line and station spacing increases survey time required over

a given area (i.e. a halving of line and station spacing would

result in a four-fold increase in survey duration). Geophysical

surveys are not likely to be employed as a reconnaissance tool

if they are excessively time intensive and therefore expensive.

As hearths and middens are extremely subtle the survey needs

to be both expedient and comprehensive, a balance we hoped

to achieve with 1m line spacings. This has been demonstrated

to be erroneous, since the surveys, while conducted relatively

quickly with this methodology, were not successful in detecting

the known targets.

Similarly we thought the increased survey speed available

from using a handheld GPS rather than a tightly controlled

survey grid would overcome the disadvantages of decreased

positioning accuracy. Despite the strong correlation between

the reburial location and recorded anomalies – suggesting

the positioning system used was not entirely ineffective – this

does not appear to be the case. We therefore think another

positioning system (e.g. differential GPS, submetre GPS

or survey tapes) may have yielded a better result, due to the

ability of these techniques to provide a higher degree of spatial

accuracy to located anomalies.

We suggest that further investigation of the application of

such techniques to hearths and midden sites in the Australian

context should be pursued, to develop a robust methodology

that can be rapidly deployed with a high level of confi dence and

success. The authors intend to pursue further research on the

applicability of these techniques to such sites with a focus on the

most appropriate survey methods to achieve a robust result in

the least amount of fi eld time.

AcknowledgementsThanks to the Woolgar Valley Aboriginal Corporation and

Matt Kersh (station owner) for granting us permission to carry

out this research. We acknowledge the contribution of Toni

Massey, Ant Timms and Jane Simons to the geophysical survey.

Ecophyte Technologies Pty Ltd is thanked for providing the

GEM-2 EMI instrument and other in-kind support. The initial

archaeological surveys and excavations on Bora Station were

generously funded through an AIATSIS Research Grant (Grant

Number G2004/6898). We thank reviewers Jo McDonald and

Scott L’Oste-Brown for comments which signifi cantly improved

the manuscript; however, any errors remain the responsibility of

the authors.

Figure 2 Geophysical survey results: Singe sensor proton precession magnetometer.

Figure 3 Geophysical survey results: Electromagnetic Induction, Quadrature Response 47975 Hz.


Short Reports

ReferencesAllen, H. 1998 Reinterpreting the 1969-1972 Willandra Lakes archaeological

surveys. Archaeology in Oceania 33(3):207-220.

Connah, G., P. Emmerson and J. Stanley 1976 Is there a place for the proton

magnetometer in Australian fi eld archaeology? Mankind 10:151-155.

Crothers, L. 1997 A Documentation of Aboriginal Cultural Heritage Sites Located in

the Richmond Region, Northwest Queensland. Unpublished report to Central

Queensland Aboriginal Land Council.

Dalan, R.A. and S.K. Banerjee 1998 Solving archaeological problems using

techniques of soil magnetism. Geoarchaeology 13:3-36.

Domett, K.M., L.A. Wallis, D. Kynuna, A. Kynuna and H. Smith 2006 Late Holocene

human remains from northwest Queensland, Australia: Archaeology and

palaeopathology. Archaeology in Oceania 41(1):25-36.

Fanning, P., S. Holdaway and R. Phillips 2005 Heat Retainer Hearth Identifi cation as

a Component of Archaeological Survey in Western New South Wales, Australia.

Unpublished paper presented to the Australasian Archaeometry Conference,

Australian National University, 12-15 December 2005, Canberra.

Frederick, C.D. and J.T. Abbott 1992 Magnetic prospection of prehistoric sites in an

alluvial environment: Examples from NW and west-central Texas. Journal of

Field Archaeology 19(2):139-153.

Holdaway, S.J., P.C. Fanning and J. Shiner 2005 Absence of evidence or evidence of

absence? Understanding the chronology of Indigenous occupation of western

New South Wales, Australia. Archaeology in Oceania 40(2):33-49.

Kvamme, K.L. 2003 Multidimensional prospecting in North American Great Plains

village sites. Archaeological Prospection 10:131-142.

Linford, N.T. and M.G. Canti 2001 Geophysical evidence for fi res in antiquity:

Preliminary results from an experimental study. Archaeological Prospection

5(3):128-138.

McClean, R.G. and W.F. Kean 1993 Contributions of wood ash magnetism to

archaeological properties of fi re pits and hearths. Earth and Planetary Science

Letters 119:387-394.

Nobes, D.C. 2000 The search for “Yvonne”: A case example of the delineation of a

grave using near-surface geophysical methods. Journal of Forensic Sciences

45(3):715-721.

Nobes, D.C. and A. Tyndall 1995 Searching for avalanche victims: Lessons from

Broken River. The Leading Edge 14(4):265-268.

Peters, C., M.J. Church and C. Mitchell 2001 Investigation of fi re ash residues using

mineral magnetism. Archaeological Prospection 8:227-237.

Reynolds, J.M. 1997 Introduction to Applied and Environmental Geophysics.

Chichester: John Wiley and Sons.

Robins, R. 1996 A report of archaeological investigations of open hearth sites in

southwest Queensland. Queensland Archaeological Research 10:25-35.

Smith, M.A., E. Williams and R.J. Wasson 1991 The archaeology of the JSN site:

Some implications for the dynamics of human occupation in the Strezlecki

Desert during the late Pleistocene. Records of the South Australian Museum

25:175-192.

Smith, W.H.F. and P. Wessel 1990 Gridding with continuous curvature splines in

tension. Geophysics 55(3):293-305.

Stanley, J.M. 1983 Subsurface survey: The use of magnetics in Australian

archaeology. In G. Connah (ed.), Australian Field Archaeology: A Guide to

Techniques, pp.82-86. Canberra: Australian Institute of Aboriginal Studies.

Stanley, J.M. and R. Green 1976 Ultra-rapid magnetic surveying in archaeology.

Geoexploration 14:51-56.

Steinberg, J.M., A.M. Gontz, H. Trigg, A. Lahey and L. Aragon 2007 Identifi cation of

small, near-surface buried shell middens with conductivity on Shelter Island,

NY. Geological Society of America: Abstracts with Programs 39(1):64.

Veth, P., G. Hammand and R.J. Lampert 1990 The archaeological signifi cance of the

Lower Cooper Creek. Proceedings of the South Australian Museum 24(1):43-

66.

Wallis, L.A. 2007 Archaeological Investigations on Bora Station, Northwest

Queensland. Unpublished report to Australian Institute of Aboriginal and

Torres Strait Islander Studies, Canberra (Grant Number G2004/6898) and the

Woolgar Valley Aboriginal Corporation.

Wallis, L.A., H. Smith and D. Smith 2004 Investigations of Aboriginal hearth sites

along the Flinders River, inland northwest Queensland. The Artefact 27:59-76.

Whiting, B.M., D.P. McFarland and S. Hackenberger 2001 Three-dimensional

GPR study of a prehistoric site in Barbados, West Indies. Journal of Applied

Geophysics 47:217-216.


BOOK REVIEWS

RENEWING WOMEN’S BUSINESS: A DOCUMENTARY

Julie Drew & Wardaman Aboriginal CorporationJulie Drew & Burbank Production Services, Crow’s Nest, NSW,

2005, 57mins, DVD

Reviewed by Sally BabidgeSchool of Social Science, University of Queensland, Brisbane,

QLD 4072, Australia

Renewing Women’s Business is a documentary fi lm that follows

Lily Gin.gina Burdum, an elderly Wardaman woman from the

Victoria River District (referred to throughout the narrative

as Lily), and Julie Drew, an archaeologist from the University

of Sydney, on a camping trip. Lily is one of the few women of

her generation who was taken through initiation rites, and the

accompaniment and attention of her younger relatives on the

trip is central to the narrative. We see Lily, with the assistance

of Drew, teaching the younger women aspects of ‘women’s

business’ and ‘culture’ through viewing rock art, learning dances,

visiting waterholes and other places, and recounting ‘Dreamings’,

associated mythic beings, and the social regulations associated

with these. The DVD includes the fi lm itself as well as a ‘Picture

Gallery’ with many images of the group’s camping trip, rock art,

scenery and fi shing, as well as the fi lm crew and other ‘behind

the scenes’ images.

Spatial orientation is provided with a still of a good map of

the Northern Territory and Victoria River District, as well as the

location of the Wardaman community and Innesvale/Menngen

Station. The introduction to the fi lm also includes footage of

preparations for the bush trip: adolescent girls having a telephone

conversation with their teacher from Katherine School of the Air

explaining their coming absence in terms of going camping to

‘learn about culture’ (Chapter 2); loading a vehicle with swags and

supplies; and narrated introductions to the elder women central

to the fi lm as they warm themselves around a fi re next to their

house. This includes Lily’s sister (Queenie Nabijiji Morgan) who,

it is pointed out, was not initiated in her youth due to having

lived far away from the area from a young age with her ‘promise

husband’. Both older women worked on cattle stations, but Lily

grew up close to her family and country.

Drew’s personal depth of experience and knowledge of the

women, the region and the topic plays a prominent role in the

documentary, including as scriptwriter. The narration swaps

between Drew and May Rosas (who Drew introduces as the

narrator and ‘Lily’s niece’). They have apparently pitched the

fi lm at both a local and national audience. For example, we are

told that ‘elders must share their knowledge ... so that all young

Australians can value it’; and that ‘Lily is happy to share these

stories to help build a better understanding and respect for

Wardaman woman’s business’ (Chapter 12). Given the audience

they seek, I thought that the sparse use of subtitles was a tactic

that may annoy some who like every word uttered on screen

translated to text. However, I personally found satisfying the

resultant requirement that the viewer focus on the central fi gures’

movements and non-verbal communication in conjunction with

the spoken Aboriginal English.

The structure of the fi lm and steady camera work is easy to

follow and good to watch. As a subject, Lily seems comfortable

with the lens turned upon her, although Drew less so. In contrast,

the young girls interact directly with the lens, acutely aware of

its gaze. The girls – with their solemn expressions and sidelong

glances at the camera as they listen to Lily’s interpretations of

the fi gures in the rock art galleries; their raucousness as they are

taught to dance; and their hands cupped over their mouths as

they giggle (and are heard being told to ‘shush’) at the discussion

between Lily and Drew of fi rst menstruation rituals – lend a lively

presence to the fi lm that this reviewer appreciated. My favourite

scenes in the documentary were of the group of girls and women

sitting in a waterhole digging out the oily mud to smear on

themselves as ‘love magic’ (jirri) (Chapters 5-6); and later, the

conversation among a handful of adolescent and teenage girls

sitting on the riverbank telling the camera about the power of the

love magic, and about ‘skin’ laws in terms of who each of them can

and can’t ‘go with’. The girls’ preoccupation with sexual and social

relationships was telling of their gender and age. More than this,

these scenes demonstrate their lived social realities. On camera,

they engage in gender and generational-specifi c interpretations of

cultural knowledge in recounting ‘skin’ regulations. They convey

a sense of the contemporary-mythic when they tell a story about

a woman who successfully wielded love magic over her man, and

his later attempt at ‘breaking that spell’ (Chapter 12). Such scenes

demonstrate their lived social realities.

These scenes contrasted with the rather staged presentation and

representations of cultural knowledge in questioning and response

between Drew and Lily in the rock art galleries. The narrator

informs us that knowing connections between ‘Dreaming’, law,

and person means being a businesswoman, and respect for Lily’s

depth of knowledge is made plain. Nonetheless, I found Drew’s

constant interjection with questions and comment, a hindrance in

an appreciation of symbolic representation and myth that might

be seen in the rock art.

While in many ways an interesting fi lm, the production of

Renewing Women’s Business needs some fi nessing. The cover

provides only very basic information and the inside cover only

lists the images in the DVD’s ‘Picture Gallery’. Furthermore,

it would seem to make sense to indicate on the cover of the

DVD that there is a detailed website associated with the fi lm

(www.wardamanwomensbusiness.com), including teacher’s

notes, and other useful information. As with the DVD itself,

the information on the website would seem to me most useful

as a record of one woman’s knowledge for the local community,

but I feel it needs further refi nement if it is to be used by the

wider public or university educators. Perhaps the most obvious

lacunae in production detail was that nowhere could I fi nd

information around the issue of secrecy or gender sensitivity,

except for a brief warning to Aboriginal men at the very

beginning of the fi lm. Apparently it is suitable and appropriate

to be shown to non-Aboriginal men (given there is no warning


Book Reviews

to the contrary), but given the obvious question, I feel this

could have been addressed explicitly.

It is worth making some critique of the way in which Drew

(who is referred to as an ‘ethnographic archaeologist’) applies the

culture concept in the scripting and narration. To take the most

obvious example, the young girls are said to ‘live in two worlds’

(Chapter 2). Firstly, Lily herself (and her sister), as well as the

young girls’ mother’s generations, and the young girls themselves,

all live (currently and apparently have done so throughout their

lives) in a complex social and cultural context that cannot be

characterised as simply Aboriginal or non-Aboriginal. From an

ethnographic perspective, the whole of their social reality, from

which they draw meaning, and which they take for granted, is

their cultural world. In the way they engage with the cameras, it

is clear that the youngest take for granted the role of multimedia

and its power as a representational tool. At the same time, they

emphasise to their school teacher (over the phone) that they must

be absent for camping and ‘culture’, and they express their respect

for their grandmothers’ knowledge. All of these things and more

are part of their world, rather than necessarily belonging to one

of two separable ‘cultural spheres’.

A further comment on the use of the concept of ‘culture’

must be made. The narrator states: ‘Today the girls have lost

the knowledge of their ancestors’ traditions’ (Chapter 2). The

premise of the documentary, or perhaps simply what underlies

the narration of the fi lm, is the pretence that one documented

camping and education trip with a knowledgeable woman and an

inquisitive archaeologist will ‘renew women’s business’. Perhaps

the activities recorded on fi lm are a minor part in other processes

of cultural renewal and this fi lm documents only a fraction of

a larger picture. Perhaps it is the case that the documentary

represents only one excursion in a series; or perhaps much more

work is being done by Wardaman women to ‘renew women’s

business’. However, minimal information on the context of

cultural renewal is emphasised in the fi lm. I argue that in a fi lm

showing living people engaged in recording and representing

aspects of their culture, more must be done to develop a clear

language of the present in all of its complexity. By this, I mean

there needs to be an awareness of the politics of performance,

and greater sophistication in both the representation of the

process of salvaging an old person’s knowledge and young

peoples’ apparent interest in this knowledge renewal and what

they may do with it.

COASTAL THEMES: AN ARCHAEOLOGY OF THE SOUTHERN CURTIS COAST, QUEENSLAND

Sean UlmTerra Australis 24, ANU E Press, Canberra, 2006, xxviii+314pp,

ISBN 1-920942-93-9

Reviewed by Bryce BarkerSchool of Humanities and Communication, University of

Southern Queensland, Toowoomba, QLD 4350, Australia

Coastal Themes: An Archaeology of the Southern Curtis Coast,

Queensland by Sean Ulm is another in the long-standing Terra

Australis monograph series, which has had a new lease of life

in recent times. One of the strengths of Terra Australis is that

it is one of the few publications that allow the full presentation

of archaeological data in the form of site reports – something

that is increasingly diffi cult to access in published form. The

research outlined in this monograph is the culmination of

Ulm’s PhD research and continues a long line of regional coastal

research projects emanating from the University of Queensland’s

Department of Anthropology and Sociology (now School of

Social Science). This work is also the latest of a series of intensive

regional coastal studies on the east coast of Queensland spanning

some three decades, including Moreton Bay, the Cooloola Coast

and Fraser Island, the Keppel Islands, the Whitsunday Islands

and Princess Charlotte Bay.

The study aims to combine a broad regional archaeological

characterisation of the southern Curtis Coast, addressing issues

emerging out of archaeological studies in southeast Queensland,

with more specifi c methodological issues relating to chronology,

taphonomy and sampling.

In broad terms the research in this monograph attempts a

critique of the mid-late Holocene archaeological signature, by

arguing that more fi ne-grained regional analyses show that there

was no pan-continental uniformity of mid-late Holocene change,

and that rather there was considerable variation across the

continent. Ulm specifi cally questions Lourandos’ pan-continental

model of late Holocene change by making the case that his

use of supraregional trajectories as a primary locus of change,

amalgamating diverse sequences from widely separated regions to

defi ne overarching patterns, actually masks the regional diversity

of Holocene change. Ulm argues that intensive localised regional

studies, focusing on specifi c methodological issues such as fi ne-

grained chronologies will ultimately challenge the archaeological

signature of mid-late Holocene change. The author argues that

‘fundamental elements of our understanding of the mid-to-late

Holocene have been challenged in recent years’ (p.4) citing, for

example, that eel traps and swamp management ‘may date to the

early Holocene’ (p.4) and that backing technology is found to be

much earlier than a mid-late Holocene innovation (even though

it is clear that it was not a commonly adopted technology until

the mid-late Holocene). These challenges are all to be welcomed,

and intensive fi ner-grained regional studies will no doubt change

aspects of how we view the mid-late Holocene.

However, one issue that is not addressed in the author’s

critique of pan-continental models is that of scale. As Lourandos

(Lourandos et al. 2006:35) has stated

the question of scale is central to questions of intensifi cation

and complexity – the debate [keeps] slipping between scalar

levels – between big picture, to more nitty-gritty middle range.

It [is] a case of comparing apples to oranges. The two levels need

to be kept separate as they have different logic and data ... the

long-term archaeological trends [are] often criticised as not

including enough middle-range information; as if fi ner-grained

analysis would reveal further variation in the general trend itself.

But the two sets of data are quite different. Finer-grained data

would not necessarily alter the general trend, but just provide

more information; in this case, at fi ner temporal levels. It’s a bit

like saying, for example, that while population in Great Britain

generally has continued to increase over the last 150 years or

since the Industrial Revolution, when one looks at fi ner-grained


Book Reviews

regional British data the patterns are varied. Both may be correct,

and data from one level doesn’t necessarily alter the information

or pattern from the other.

While the fi ne-grained research outlined in this volume

highlights considerable regional and local temporal and spatial

variation, it still broadly supports a mid-late Holocene change;

such as evidence of increasing intensity of dynamic site and

landscape use over time, and an accumulative increase in the

establishment of new sites in the late Holocene on the central

Queensland coast.

As part of the more fi ne-grained regional study, Ulm

advocates a focus on refi ning regional chronologies, stating that

‘establishing secure regional chronologies remains a fundamental

key to building meaningful accounts of intra- and inter-regional

sequences in Australia’ (p.5). In particular, he critiques the

way in which some sites are dated, specifi cally the common

procedure of obtaining basal dates and subsequent assumptions

made as to continuity of use, with the surface being treated as

analogous to the contact period. Ulm’s meticulous collation of

radiocarbon dates in all Queensland archaeological sites has

shown that Holocene sites in Queensland have an average of

2.41 dates per site which he sees as being clearly inadequate to

properly address the complexity of regional variation across the

continent, especially in regard to assumptions about continuity

of site use. There can be no doubt that it is certainly preferable

to obtain as many dates as possible for a site. However, in many

cases the decision not to date anything other than the basal XU

and identifi able stratigraphic changes relate almost wholly to

cost. In regard to surface or upper-most XUs many sites have

post-contact artefactual material such as glass etc, and thus can

be relatively dated.

The other chronological issue raised in this volume relates

to environmental factors such as the uneven distribution of 14C

in the biosphere, which impact directly on samples selected for

dating. These factors lead to distinct regional differences which

can impact on the accuracy of radiocarbon determinations

measured on charcoal as well as marine samples. Ulm’s work on

obtaining a more fi ne-grained analysis in relation to chronology

is meticulous and breaks new ground in conclusively establishing

the degree of variation in marine reservoir effect; particularly

between samples taken from estuarine environments and the

routinely applied ∆R value for northeast Queensland. The

differences of up to 300 years can certainly be crucial when

dealing with sites of short occupation. However, for many

research models – depending on what questions are being asked

and the scale involved – a couple of hundred years difference may

not radically alter regional interpretations.

Other sampling issues Ulm addresses are the rockshelter

versus open site bias – where Ulm (p.7) states that ‘even in coastal

Australia, where the recent archaeological record is dominated by

shell middens, accounts remain based on rockshelter sequences’.

He states that ‘several studies have demonstrated that a high

degree of post-depositional movement of cultural material

between stratigraphic units can occur without damaging the

physical appearance of strata or strata boundaries ... calling into

question basic assumptions about the integrity of the rockshelter

deposits which form the basis of our understanding of the

archaeology of Australia’ (p.9). Taking a leaf from the author’s

localised regional approach, I would argue that this greatly

depends on local conditions relating to specifi c rockshelter sites

and that no assumption should be made that, because conjoin

analysis at Kenniff Cave showed downward movement, the

same conditions relate to other shelters in completely different

contexts with completely different geological sediments and

deposition histories. In any case, I am not sure too many

Australian archaeologists assume that rockshelter sites have total

stratigraphic integrity, but I am reasonably sure, based on a wide

body of research including my own, that relative to rockshelter

sites, open sites on the tropical coast are more problematic in

terms of post-depositional integrity. I am surprised that this is

still seen as an issue as a lot has been written about the reasons

for a preference for rockshelters – especially on the Queensland

tropical coast.

As an example of rockshelter sampling bias the author states

that ‘accounts of the regional archaeology of the Whitsunday

Islands on the central Queensland coast are based almost entirely

on evidence from small excavations conducted on rockshelters

despite open sites featuring in the ethnohistoric and archaeological

record’ (p.7). This statement is as baffl ing as it is misleading. There

are simply no known open sites of stratigraphic integrity in the

Whitsunday Islands. The reality is that the majority (not all) open

sites north of the tropic of Capricorn are disturbed, redeposited or

occur in problematic contexts.

Assessing site integrity through conjoin analysis of shell is one

of the methodological approaches taken by the author to provide

a sturdier framework for determining the degree of integrity of

open sites (Chapter 5). As Ulm points out, it is strange, given

‘explicit and implicit reference to this site type [i.e. open sites] as

stratigraphically problematic’ (p. 65) that further work has not

been done in regard to taphonomy of open midden sites. Ulm

states that ‘[i]n certain circumstances, bivalve conjoining may

be a useful adjunct to conventional approaches to shell midden

analyses involving very basic characterisation of assemblage

composition, with the potential to contribute an independent

form of evidence to our understanding of site integrity and

resolution, discard patterns and periodicity of occupation’

(p.77). If we are to have greater confi dence in the integrity of

open midden sites, it is clear that a much more sophisticated

and refi ned methodology such as that presented here, needs

to be applied. Methods such as conjoin analysis on bivalves,

coupled with recent developments in foraminiferal analysis in

middens and analysis of non-cultural shell deposits, can lead to

archaeologists being able to sample a broader range of site types

in future analyses.

Overall, the call made in this volume for a more fi ne-grained

regional approach to archaeological models is timely and to be

applauded. After some three decades of pioneering research

directly to the south and north of the Curtis Coast it is possible

to start focusing on a more refi ned and sophisticated regional

perspective. Inevitably the more work carried out in a region

relating to questions and problems already posed by prior

research will reveal a greater degree of detail and complexity than

the models that preceded it. This is certainly to be welcomed. In

this volume the author has demonstrated meticulous attention

to detail in what amounts to an almost fl awless presentation and

he has set the methodological standard for a more meticulous

and fi ne-grained analysis of sites in the future. Ulm concludes


Book Reviews

by stating that ‘[t]he major task ahead therefore remains a

basic one: to construct and compare detailed individual site

sequences from a range of site types, at the local and regional

level, to establish the existence of trends independent of site-

specifi c taphonomic and/or environmental factors’ (p.255). I

have no doubt that future research can and will focus on these

tasks; however, whether it will change how we view mid-to-late

Holocene change in Australia remains to be seen.

ReferencesLourandos, H., B. David, B. Barker and I.J. McNiven 2006 An interview with

Harry Lourandos. In B. David, B. Barker and I.J. McNiven (eds), The

Social Archaeology of Australian Indigenous Societies, pp. 21-39. Canberra:

Aboriginal Studies Press.

THE SOCIAL ARCHAEOLOGY OF AUSTRALIAN INDIGENOUS SOCIETIES

Bruno David, Bryce Barker & Ian J. McNiven (eds)Aboriginal Studies Press, Canberra, 2006, xiv+382pp,

ISBN 0-85575-499-0

Reviewed by Richard CosgroveArchaeology Program, School of Historical and European Studies,

La Trobe University, VIC 3086, Australia

This book is dedicated to Harry Lourandos and edited by

three of his ex-students from the University of Queensland. It

is divided into fi ve sections and each deals with an aspect of

‘social archaeology’, its emergence, development, application,

achievements and theoretical underpinning. Part 5 seems out

of place somewhat with a lone chapter by Barbara Bender

that may have been better placed in Part 1. It is generally well-

produced although there are a few typographical errors and one

clanger in Lourandos’ bibliographic references; his erroneously

titled 1968 paper.

Lourandos’ work had an archaeological and political impact

on his students and other followers as evidenced by this volume.

He generated vigorous debates about the changes recorded in the

Australian late Holocene archaeological record. He is generally

credited with introducing the concept of ‘intensifi cation’ into

Australian archaeology and, as argued by some throughout the

book, refocused thinking on the trajectories of past change.

He emphasised the need to move away from the ecological/

environmental deterministic explanations that he saw as

generally powerful in explanting change in the late Pleistocene

but not so in the late Holocene. The book illustrates how

dependent social archaeology is on the availability of good

ethnography and history as frameworks for explanation in late

Holocene change.

The fi rst part of the book investigates the tension that

Lourandos saw between the ecological approach and the

need to bring ‘people’ into archaeological explanations to

create a social archaeology. The fi rst chapter by McNiven et

al. is devoted to rehearsing the history of those changes and

referencing much of the work of Lourandos and others with

an interest in promoting the social explanatory agenda. There

is an interview with Harry Lourandos that repeats many of

the arguments made in his papers throughout his career and

claims that his most ardent critics simply misunderstood his

writings are peppered throughout the interview. The chapter

by Sandra Bowdler is a personal view on the achievements

of Lourandos and his background, putting into a particular

context the driving force behind his development of the social

archaeological agenda.

Part 2 of the book deals with the infl uence of earlier writings

on the perception of Aboriginal people, particularly the way

hunter-gatherers were characterised as separate entities from

agriculturalists and horticulturalists. This is the theme developed

by Bruno David and Tim Denham from the static/dynamic

dichotomy with clear links to Lourandos’ previous work strongly

articulated in his 1985 paper, ‘Intensifi cation and Australian

prehistory’. Ironically the latter term, ‘prehistory’, identifi ed by

David and Denham as ‘prejudicial’ (p.57) was used by Lourandos

throughout his 1985 paper. Bryce Barker’s chapter distinguishes

the early ethnographic observations of Roth and Tindale with

the archaeological results from his Whitsunday Islands work. It

is a cautionary tale of the over-reliance on historical writings,

where errors of fact become compounded by future writers

unable or unwilling to establish the archaeological veracity of

the observations.

Ian McNiven explores the exogenous and endogenous

infl uence on Aboriginal peoples using the appearance of

microlithic stone artefacts, canoes and the dingo as examples.

He challenges the view that change has only come from inside,

particularly through his work in the Torres Strait, and makes

the point that the dingo was probably introduced from New

Guinea. Recent genetic research by Hudjashov et al. (2007),

however, importantly suggests quite limited physical contact

during the Holocene. Material items and ideas can be quite

fl uid between separate populations, a point made by McNiven,

and recent microlithic technology dated to 15,000–13,000 BP

identifi ed in Queensland would suggest an earlier invention/

introduction not associated with the dingo. McNiven sees

outrigger canoe technology as a way that coastal Aborigines

increased their use of marine resources due to demands of

higher populations through social imperatives, although we are

not told what the specifi c catalysts for increasing populations

were. This is a general problem with Lourandos’ late Holocene

intensifi cation stance because of the inherent circularity of the

argument. Large populations generate social complexity; social

complexity brings about increased demand for productivity,

which leads to larger populations.

The intensity with which some of the writers defend

Lourandos’ contribution to Australian archaeology is evident

in the book. The ‘call to arms’ chapter by Deborah Brain

attempts to shore up his legacy by advocating a more tenacious

use of the undiluted ‘intensifi cation’ concept before it slips

away due to ‘a kind of creeping ambiguity’. She argues that

critics have not only misrepresented Lourandos’ work but also

his ideal by necessarily aligning it with so-called ‘traditionalists’,

a contamination no less. She states that Lourandos made

people (Australian archaeologists?) uncomfortable because he

saw Holocene Aboriginal societies deviating from traditional

notions of ‘hunter-gatherer cultures’ in the late Holocene. It

is also true that some were uncomfortable for other reasons,

particularly the lack of clearly stated archaeological correlates

of social complexity that delineated the late Holocene from


Book Reviews

other periods. Indeed some argued that the changes could be

explained in other ways and that the generalised continental

pattern did not stand up to scrutiny when regional signatures

were closely examined. The presence of rock art, body

adornment, composite hafted tools, raw material movement,

maritime technology, patterned land-use, resource management,

symbolic behaviour, increased site use, ritual burial of the dead

and the conquest of marginal environments were hall marks of

late Pleistocene and early Holocene regional archaeologies, not

just of the late Holocene.

Part 3, Anthropological Approaches and the chapters therein

more clearly demonstrate the links between ethnography and

understandings of past cultural complexity. We see through the

eyes of the seven anthropologists – Luke Godwin and James

Weiner; Marcia Langton; John Bradley; Amanda Kearney and

John Bradley; Franca Tamisari and James Wallace – the attempt

to integrate the social aspects of peoples’ lives with their material

remains as well as their landscapes. Views of people’s pasts are

couched in terms of the Dreaming, where stone artefacts, water

bodies, symbolic landscapes delineated by cycad patches and

processing stones, for example, take on different meanings from

those of the archaeologists. Armed with these insights it is not

surprising that a richer and more multilayered understanding of

the mundane material remains and resource zones can be had.

The ‘tyranny of the ethnographic record’ fi rst expressed by Wobst

always limits archaeologists delving into the deeper past because

the behavioural strands of ethnographic connection become ever

more tenuous, and the scale and resolution of the archaeological

record becomes coarser.

In Part 4 archaeological data and its relevance to explaining

the social aspects of past lives are discussed. Donald Pate’s

chapter is instructive in the use of scientifi c analyses to untangle

the threads of social connections. Using stable isotope analysis,

palaeopathology, non-metrical cranial measurements and

observed mortuary practices identifi ed at Roonka Flat, Pate

challenges the notion of egalitarian Aboriginal society. He makes

the point that these Murray River societies were chronologically

distinguished along lines of gender where males appear to be

common in the early Holocene burials whereas females are

increasingly interned, with grave goods differentially distributed

between older males and females. These he saw as providing

evidence of social stratifi cation through time, particularly male

authoritarianism.

Peter Veth provides a model for understanding the occupation

and settlement of the Western Desert using both archaeology and

historical documents. He suggests a six-phase occupation pattern

linked to the evidence for environmental change, associated

archaeological variability and possible language diffusion into

the region. He suggests that the tempo of occupation increased

with the establishment of territory based on the diversity of late

Holocene art in the arid zone.

Environmental change in southwestern Victoria at fi ne scales

are discussed by John Tibby, Peter Kershaw, Heather Bilth, Aline

Philibert and Christopher White. Their analysis of a sediment

core from Lake Surprise reveals that beginning about 3700 years

ago high levels of climatic variability are seen in the core. They

suggest that this instability may have been a catalyst for the

social changes identifi ed by a number of archaeologists working

in the region. This runs counter to the arguments for socially-

driven change within Aboriginal society in this area at this time,

although separating the competing arguments of the reasons for

change is diffi cult because there are no clear correlates to provide

primacy of one over the other.

Work by Cassandra Rowe and Melissa Carter discuss the

archaeology of the Torres Strait region where pre- and post-

sea-level change are identifi ed in relation to the changing

settlement patterns on Badu, Mer, Dauar and Waier Islands.

Both environmental data and ethnohistorical evidence are

used to compliment the archaeology of the region. Changes

in subsistence practices are identifi ed that depart signifi cantly

from the pre-2000 BP occupation period and challenge earlier

formulations of subsistence practise.

This book covers a lot of ground within the ‘intensifi cation’

gambit. However, those expecting a critical analysis from authors

who have taken different views on the ‘intensifi cation’ debate

will be disappointed and this remains one of the limitations of

the book. However, it will be of interest to students as well as to

a general archaeological audience interested in the origin and

development of ideas fi rst formulated by Harry Lourandos. It

can be said that he made a signifi cant and lively contribution

to an understanding of Indigenous Australian cultures. His

legacy will continue to endure within the archaeological

literature and broader debates on the nature of late Holocene

cultural changes.

ReferencesHudjashova, G., T. Kivisild, P.A. Underhill, P. Endicott, J.J. Sanchez, A.A. Lin, P. Shen,

P. Oefner, C. Renfrew, R. Villems and P. Forster 2007 Revealing the prehistoric

settlement of Australia by Y chromosome and mtDNA analysis. Proceedings of

the National Academy of Sciences 104(21):8726-8730.

NEOLITHIC

Susan Foster McCarterRoutledge, New York, 2007, xviii+221pp, ISBN 978-0-415-36414-0

Reviewed by Phillip C. EdwardsArchaeology Program, School of Historical and European Studies,


I imagine that Australian archaeologists fl ip through a new book

on early farming to see what it says about Melanesia. They will

be disappointed in the case of Susan McCarter’s Neolithic. The

region is ignored in the review of the world’s natural hearths

of farming and left off the map of them completely. Rather,

evidence is cherry-picked from the literature in support of

general points, mainly from the regions of Southwest Asia and

also Europe, with others used less often. The reader should not

expect an up-to-date summary of recent research. McCarter’s

book is intended as an introduction to Neolithic cultures and the

world of fi rst farming, and it does contain much useful material.

There are quite authoritative sections on biological evolution,

and the processes of plant and animal domestication, and there

is a nicely balanced treatment of past theoretical approaches to

early agriculture.

Perhaps because the pitch is intentionally general, the reader

can expect to encounter generalised appeals to authority, such

as ‘what most archaeologists think’ and ‘what they all believe’;


Book Reviews

for example, ‘Neolithic bowls held the same things as what

they do now’ (p.6), and ‘In all known societies people work

hard to make a living, love their families, worry about the

health ...’ (p.6). Therefore, the book is not positioned well as a

university text. It is pitched at a reasonable level for secondary

school students, but I think it will still struggle even to attract

support in that arena. Mimicking the web-screen, school texts

nowadays are packed with colourful, overlapping frames which

jostle for attention, whereas Neolithic is illustrated by black-

and-white ink sketches reminiscent of a 1950s production.

The illustrations, like the text, are determinedly generic. They

include images of unnamed mounds, ruined houses, mud-

brick houses, thatched houses, the wooden frames of houses,

mice, cats, dogs and men leading camels. Most arresting is the

tableau of a grey wolf standing protectively over a tiny Yorkshire

terrier, complete with a bow tied around its forehead (p.18).

Such a combination of elements lends the work a distinct air

of quaintness. Likewise, the glossary includes basic terms such

as ‘gender’, ‘grain’, ‘taming’, ‘tool’ and ‘war’, while ignoring some

fairly unusual features such as the food plants ‘black gram’ and

‘green gram’.

In some places the author lacks authority over the material

she employs. There is frequent reference made to the Jomon

culture of Japan with its precocious pottery and edge-ground

axes. Claiming that pottery in Japan occurred 8000 years before

anywhere else (p.103), the author ignores the even older late

Pleistocene ceramic traditions of Siberia and China, from which

the Jomon tradition may have emerged (neither is consideration

given to Greater Australia in terms of the prior antiquity of edge-

grinding). There is an appendix defi ning the geographical regions

of the world, which is also a little misleading. The Levant is not a

separate region to Southwest Asia, but a subregion of it; neither

should it be excluded from a defi nition of the Mediterranean

borderlands. Only a single archaeological site is placed on a

map in the entire book. This comes in the form of a large star

denoting Abu Hureyra in Syria, placed on an otherwise empty

map of Southwest Asia.

Given that the book foregrounds the term ‘Neolithic’ in its

single-word title (lacking the usual clarifi cations after the colon),

I would have expected it to pay a little more attention to the

development of the concept. Mention is made of Lubbock’s

original defi nition based on the advent of edge-grinding

technology, and there is a further brief description of how the

term became extended to denote farming people in general, but

no discussion of the contradiction inherent in this latter use. It is

unconvincing to see the Mississippian-period inhabitants of the

Southeast United States described as a ‘Neolithic’ people (p.8),

without further elaboration.

It is notable that Routledge released Neolithic in the same

year as its more authoritative text on early food production

(Denham and White 2007). Neolithic will not serve as an

advanced text, but it would prepare the general reader for

tackling the more complex papers to be found in its more

sophisticated 2007 stable-mate.

ReferencesDenham, T. and P. White (eds) 2007 The Emergence of Agriculture: A Global View.

London: Routledge.

ARCHAEOLOGY OF ANCIENT AUSTRALIA

Peter HiscockRoutledge, London, 2008, xviii+338pp, ISBN 978-0-415-33811-0

Reviewed by Brian FaganDepartment of Anthropology, University of California, Santa

Barbara, CA 93106, USA

The tyranny of the ethnographic record has dogged Australian

archaeology for generations. This is hardly surprising, given the

often exiguous archaeological signature, a long preoccupation

with chronology, culture history, and the Dreamtime, and

the availability of numerous, albeit often fractured, historical

accounts dating to the past two centuries. Other assumptions

have bedeviled research as well, notably a persistent notion that

ancient Aboriginal societies were conservative and changed little

over thousands of years. However, the past quarter century has

seen a dramatic fl owering of multidisciplinary research, notably

into climatic and environmental change, in which archaeology

has played an important part. Australian researchers have

both benefi ted from, and contributed importantly to, a new

generation of hunter-gatherer studies that have transformed

many of our perceptions of such societies, ancient and modern.

Fortunately, the days of preoccupation, nay obsession, with the

San of Southern Africa’s Kalahari Desert as a model hunter-

gatherer society are now history. Peter Hiscock’s introduction

to a now very complex subject refl ects the dramatically changed

face of Australian archaeology.

Some people still denigrate the writing of what Hiscock calls

‘popular texts’, but they are, in fact, among the hardest books of all

to write. The author has to navigate between different viewpoints,

write in an easily intelligible, jargon-free style, and make hard

decisions about what to include and what to omit. Hiscock has

faced these problems head-on, in a book that is aimed, he says, at

the next generation of researchers. He has written a synthesis that

focuses on key questions and examines them by using carefully

selected examples, or case studies as he calls them, of which Lake

Mungo is one instance. His concern is to balance science and

the humanities, teetering on the fi ne knife edge between oceans

of technological detail and the need both to entertain and to

explain the ambiguities of the archaeological record. The result

is a fascinating, state-of-the-art journey through Australia’s past,

which is certainly not aimed at freshmen or the general public,

but at students and readers with a serious interest in the subject,

and probably some previous knowledge of the subject.

Chapter 1 peers through what the author calls ‘the veil’ of

Antipodean history. Here we face the tyranny of the ethnographic

record, of analogy, which has coloured interpretations of

ancient Aboriginal society. Hiscock guides us through diverse

interpretations of Lake Mungo, using as a starting point a well-

known Giovanni Caselli painting, which shows the people using

artefacts that did not exist in their day. He discusses the effects of

smallpox on Aboriginal society, and advocates a ‘methodological

uniformitarianism’ that assumes that regularities in how the

world operated structured the processes of human behaviour.

These regularities also allow us to identify ancient physical

environments. This approach allows one to escape undue reliance

on analogy and ethnography.


Book Reviews

The remainder of the book focuses on key questions, starting

with fi ve chapters on the issue of fi rst settlement, which receives

welcome, critical treatment. Chapter 2 surveys some of the major

approaches – the environmental, demographic, genetic, and

cultural dimensions of the problem. Interestingly, this is the fi rst

synthesis I have read which considers the potential impact of

the epochal Mt Toba explosion between 75,000 and 71,000 years

ago. We are treated, also, to a probing analysis of dating methods,

which effectively debunks luminescence chronologies and places

fi rst settlement to somewhere before about 45,000 years ago

(there is an appendix on Radiocarbon Dating). Chapter 3 builds

on this assessment by examining the evidence for early settlement

across Australia. Did Aboriginal groups fi rst settle the coast or

both coast and interior? Inland settlement in fact unfolded very

early on indeed. Hiscock makes a strong case for the importance

of climatic shifts, and especially droughts during the Last Glacial

Maximum. Herein lies one of the central arguments of the book

– ancient societies throughout Australia changed constantly in

response to environmental and other factors and bore little

resemblance to historical groups.

From early settlement, the author moves on to other

fundamental questions. Here, as in the Americas, megafaunal

extinctions (Chapter 4) occurred after the Ice Age, triggered

in this case by climatic shifts not human intervention. Two

chapters discuss the first Australians themselves, and life in

Pleistocene Australia. A multidisciplinary perspective stresses

not successive migrations from outside but the diversity

of ancient human populations. These evolved in different

directions as a result of dynamic adjustments by a founder

population to varied social and material environments over

long periods of time. The hunter-gatherers who colonised

Australia were accustomed to diverse and harsh landscapes.

They adapted to them with elaborate technologies and

material expressions of ritual and symbolic practices. Such

societies were not simplified versions of later cultures, but an

intricate tapestry of constantly changing local and regional

groups, where social institutions were as important as material

culture in shaping human existence.

Hiscock devotes Chapter 7 to Tasmania and its isolation,

long the subject of theorising about the effects of rising sea-

levels about 14,000 years ago. Again, he argues that the complex

interplay of economy, environment, social institutions and

technology, produced ever-changing societies. The Tasmanians

moved inland and adapted to drier, more variable climates some

4000 to 5000 years ago, at about the time when El Niño events

became more commonplace in distant Peru.

From fi rst settlement and isolation, Hiscock moves on to

technology, not with a dreary catalogue of artefacts and culture

history, but with an adaptive perspective. Were there major

changes some fi ve millennia ago that resulted from a new package

of ideological, social, technological, and economic behaviours?

Chapter 8 argues that technological changes resulted from

shifting adaptations in different areas that stemmed from highly

varied responses to the abundance of food resources that tried to

minimise risk. In other words, what happened in general terms

in Tasmania also occurred on the mainland.

Hiscock then devotes three chapters to coastal, inland, and

arid economies respectively. Again, he eschews linear culture

history and looks at the constant economic, technological, and

subsistence changes over the millennia. On the coast, he stresses

both the essential continuity of basic food getting practices and

the major changes in emphasis that kicked in over long periods

of time. Arid environments saw ever-shifting cycles of hunter-

gatherer mobility and of culture as groups adjusted to periods

of drought and higher rainfall. This approach is invaluable for

beginning students, for it turns the archaeological record from

a mind-numbing recital of artefacts into a story of dynamic,

ingenious change and opportunism that was the mark of

Aboriginal societies from the beginning. The same adaptive

approach informs Chapter 12, which discusses population

growth and mobility. Hiscock argues convincingly that there was

never a unidirectional growth trend, but a constant fl uctuation

that refl ected the harsh realities of local environments.

Chapter 13 extends these arguments into the social realm, and

covers issues of interaction through time. Here, the emphasis is

on the close entanglement of environment, economy, ideology

and social life, well illustrated by the example of Rainbow

Serpent images and other paintings. The physical and social

realms engaged with one another in complex ways over long

periods of time.

Finally, Hiscock confronts that he calls ‘the ethnographic

challenge’, studying societies of the past millennium by using

carefully selected examples. He discusses remembered landscapes

like Ngarrabullgan on Cape York in the far north, unraveling a

mosaic of archaeology, oral tradition, and remembered history.

There are issues of contact, between northern groups and

Macassan trepang collectors, which began in about 1720, and the

effects of such interactions on local societies. He argues that the

diversity and rapidity of culture change over the past thousand

years supports a portrait of ancient Australian societies as part

of a changing, varied cultural system. Archaeology reveals a

dynamic, vibrant past when ancient Aboriginal people constantly

reorganised not only their economies, but their social lives and

worldviews as well. This dynamic reality of constant change and

opportunistic adaptation is a cause for celebration and is the

death knell of earlier theories that thought of the Australian past

as conservative and unchanging.

Ancient Australia is one of those books that many of us will

return to again and again, not only for its insights into local

archaeology, but as a fi ne example of how new generations

of archaeologists are transforming our discipline into a truly

multidisciplinary enterprise. Hiscock has written a nice

primer on the perils of ethnographic analogy and brought a

formidable critical intelligence to bear on such issues as fi rst

settlement. I would not necessarily describe this clearly written

book as an easy read and it is certainly not a popular book

and the entertainment quotient is fairly low. Instead, and

more importantly, this is a serious and defi nitive synthesis

of Australian archaeology, with an excellent and up-to-

date bibliography that will appeal to a broad readership of

archaeologists, both local and international, and to generations

of serious students. I suspect that ‘read Hiscock’ will become

a regular part of the litany of many university curricula. And

so it should. Hiscock has written a potential classic, which is

perceptive, provocative, and right on the cutting edge. And,

at this stage in research, that’s as important as any number

of entertaining books, especially for anyone interested in

world prehistory.


Book Reviews

ARTIFACT CLASSIFICATION: A CONCEPTUAL AND METHODOLOGICAL APPROACH

Dwight W. ReadLeft Coast Press, Walnut Creek, CA, 2007, 363pp,

ISBN 9781598741025

Reviewed by David FrankelArchaeology Program, School of Historical and European Studies,


This important contribution summarises a lifetime involvement

with teaching, developing and debating approaches to artefact

classifi cation. Dwight Read brings to this task his own varied

contributions and highly-developed logical and mathematical

skills. Classifi cation can take several forms and have different

aims. While pragmatic classifi cations help organise masses of

material, aiding information retrieval, Read is more concerned

with the deeper issues underlying the ways in which we

structure artefacts and analyse our data and how these affect

the questions we ask. His work is situated fi rmly within the

Americanist tradition of archaeology as anthropology, implicitly

favouring a processual agenda and with an underlying principle

that our classifi cations should in some sense mirror the emic

classifi cations and concepts of their makers and users. Running

through the book is a concern for the compatibility of the

ideational and the material or phenomenological – that is how

to relate ‘ideal’ forms with the varied and diverse examples

found in the real world.

The initial literature review deals with the classic systems

of typological analysis which have characterised American

archaeology over several generations, from Rouse, Brew and

Krieger in the 1930s and 1940s to Ford in the 1950s and on

to Adams and Adams’ 1990s arguments for multiple, special-

purpose, rather than universal typologies and Dennell’s

‘systematics’. The fundamental concepts, logic and structures of

these varied approaches are summarised and critically evaluated,

providing a basis for establishing key principles and defi nitions.

Read appears most critical of the Adams and Adams utilitarian

approach, although both he and they are concerned with

problem-oriented approaches. While it is diffi cult in a study

of principles such as this to situate general ideas in specifi cs,

nevertheless some greater consideration could have been given

to the ways in which aims and problems affect – or are affected

by – approaches to classifi cation. A longer discussion of pottery

typologies, built on an analysis of the infl uential type-variety

system as developed in the 1950s, follows. As with much

American archaeology, methods and ideas developed elsewhere

in the world receive relatively little attention. It is a pity that

other formal approaches to defi nition and analysis were not

also addressed in a similar fashion. The work of David Clarke

in Analytical Archaeology (1968), for example, deals explicitly

with many similar issues, situating defi nitions of attributes

and of types within a set of structured spatial, temporal and

cultural frameworks, and developing systems leading directly

to explanation.

Having attacked and developed some key principles in

discussing these approaches, Read moves on to consider ‘objective’

approaches to classifi cation. A strong critique of Spaulding’s early

statistical techniques provides a starting point for drawing out

and clarifying additional concepts, such the nature of attributes

and how their association or combination should be used and

understood – in part a consideration of patterning on individual

objects and patterning of groups of objects. This sets the scene

for a further critique of other techniques, such as numerical

taxonomy. Once again, summary and critique allow a discussion

of related issues: here problems and sometimes confusions in the

use of different types of variables and dimensions of variation

are dealt with.

This leads directly on to a consideration of measurement.

Here, as elsewhere, Read is concerned that the techniques used

and attributes selected should have some cultural saliency, and

be useful for or relevant to understanding past behaviour. The

discussion then moves to a more technical consideration of

particular ways of measuring and defi ning artefact shape. Two

issues come to the fore: whether it is possible to defi ne artefact

classes that are internally homogenous and externally isolated;

and a belief that objects are artefacts by virtue of the way they

are conceived by their makers and/or users. Aspects of such

‘conceptualisations’ are then explored in a brief discussion of

production sequences and chaîne-opératoire.

In developing methodologies of quantitative classification

Read reconsiders what he has previously identified as ‘the

double bind problem’, where the identification of culturally

salient artefact types depends on comparisons of individual

artefacts with others of the same type, although obviously we

cannot know what that type is in advance. He resolves this

problem by a comparison of measurements of individual items

with a group of other items, in order to identify aggregate

patterning as the basis for class definition. Appropriate

statistical procedures are described in detail with case

studies of stone tools and pottery vessels. Further qualitative

methods are then presented which look at patterning based

on type frequency counts, leading to productive insights into

behavioural patterns.

While earlier discussions and examples tend to concentrate

on functional or ‘usage’ types, the fi nal section deals with the

vexed question of style and function. Here an approach is set

up which separates traits as functional, isochestic, truncated

and neutral. This forms the basis for considering aspects of

cultural transmission, development and selection, set within

a discussion of evolutionary archaeology. Here more reference

could have been made to the literature on learning and cultural

reproduction: considering those aspects which go to defi ne

groups in the present and, more problematically, in the past.

These issues, as with others, involve the nature and degree

of variability in attributes, artefacts and assemblages. However,

there could have been some greater recognition that variability

is not so much a problem to be overcome, as a problem to

be addressed. This necessitates clearer defi nitions of context,

assemblage defi nition and scale. While relationships and

traditions may be seen as shifting modes of individual or sets of

attributes through time or across space, the degree of variability

within cultural systems can give different insights, into, for

example, how inclusive or exclusive a community might be,

or the degree to which individuality may be expressed. The

measurement of variability, of course, presupposes equivalent


Book Reviews

units of observation, for variation is likely to increase with

duration and distance – often themselves dependent on

understanding degrees of depositional and taphonomic

integrity. These issues of context, scale and integrity are not

clearly addressed by Read, although his views on the nature

and impact of natural or artifi cial assemblage boundaries

would be of interest, and would expose additional structures

in the archaeological record which constrain or defi ne research

problems and the identifi cation of patterns within and between

different datasets.

How relevant is this book to Australian readers? On the

one hand, it is a valuable study of some key issues in and

approaches to archaeological practice. On the other, its

relatively limited scope, its focus on material which falls

more readily into different classes and its failure to address

the inherent complexities in the archaeological record limit

direct application to the generally more amorphous local

assemblages of stone tools, all too often from contexts of low

temporal definition.

WHALERS AND FREE MEN: LIFE ON TASMANIA’S COLONIAL WHALING STATIONS

Susan LawrenceAustralian Scholarly Publishing, North Melbourne, 2006,

viii+211pp, ISBN 1 74097 087 X

Reviewed by Martin GibbsDepartment of Archaeology, School of Philosophical and

Historical Inquiry, University of Sydney, NSW 2006, Australia

One of the great tragedies of Australian historical archaeology

has been our failure to translate the results of our research into

stories which are accessible to the public. Remote industrial

frontiers, urban slums, convicts, graveyards – most of us have

undertaken or are aware of any number of projects which would

make for riveting books that would get across our contributions

to history and knowledge. However, the volumes produced over

the last 10 years which have been aimed at the public could

probably be numbered on two hands.

With Whalers and Free Men, Susan Lawrence has produced her

second book which translates her academic research into a form

which should satisfy both the public wanting a good story and

archaeologists wanting information on a major archaeological

project. The nineteenth century shore-based whaling industry

has been a popular theme for historical and maritime

archaeologists since the 1980s, with regional and site-specifi c

studies undertaken throughout all southern Australian states

and New Zealand. Lawrence’s work focuses on the Tasmanian

shore whalers of the 1820–1840s period, examining several sites

associated with Hobart entrepreneur James Kelly.

Chapter 1 ‘Kelly’s People’ provides the historical context for the

whaling industry, James Kelly’s involvement and the operations of

his Adventure Bay and Lagoon Bay stations. One of the pleasing

elements of this book is that Lawrence manages to humanise

the history of the Tasmanian whaling industry by introducing a

range of individuals associated with Kelly’s operation, including

the Aboriginal men and women who established working and

domestic relationships with the whalers during what would

prove to be a period of tragic change and loss.

Chapter 2 ‘Building a Station’ details the industrial and

domestic operations associated with shore stations and shifts

into a description of the archaeological investigation of the

two whaling sites. The chapter takes on a narrative form which

includes not only a description of the results of the surveys

and excavations, but an explanation of the investigative

processes and insights into how Lawrence and others reached

their interpretations. What might otherwise have been tedious

technical detail becomes surprisingly readable and would give

both general readers and students a better idea of how projects

of this type proceed.

Chapter 3 ‘Whalers Rubbish’ is a discussion of the artefacts

recovered, divided into major functional categories. This section

sticks to identifi cation and broad historical context, with some

comparison to other archaeological sites. For those requiring a

more detailed dataset, there are extensive appendices.

Chapters 4 and 5, ‘Life at the Stations’ and ‘After the Whalers

Left’ draw together the archaeological and historical stories. Not

surprisingly, the subtheme of these chapters is the relationship

between what the documentary record tells us of the life of the

whalers, versus what is provided by the archaeological record.

Life on the maritime industrial frontier was hard and often

violent, but at the same time regulated by a well-established set

of contracts and traditions. Whaling especially was linked into

a wider international set of structures and hierarchies based in

part upon an individual’s performance with oar or harpoon,

often over-riding the divisions of race and class seen in ‘normal’

settlements. Once again Lawrence includes the questions which

drove the analysis, while being honest about the limits inherent

in the structural and artefact evidence (and about where her

questions simply couldn’t be answered), but still manages to

weave together a fair idea of what life may have been like. The

fi nal section on what is likely to have happened to the sites after

the whaling era fi nished, as indicated by the artefacts of later

timber-getters, bushwalkers, fi shermen and tourists, brings the

reader back to the present.

Whalers and Free Men is an easy read, with suffi cient

illustrations of site plans and artefacts that the lay reader should

come away with a good sense of the contribution that the

archaeological research has made. Production of this volume is

mostly good, with the exception that some of the photographs

have reproduced a little dark and are hard to make out.

Presumably the absence of plates on glossy paper (which might

have made these easier to see), is also one of the reasons why this

book is a very reasonable price.

INAUTHENTIC ARCHAEOLOGIES: PUBLIC USES AND ABUSES OF THE PAST

Troy LovataLeft Coast Press, Walnut Creek, CA, 2007, 168pp,

ISBN 978-1-59874-011-0

Reviewed by Denis GojakBanksia Heritage + Archaeology, PO Box 457, Newtown, NSW

2042, Australia


Book Reviews

Most archaeologists have their own favourite inauthentic

archaeology, whether it is Indiana Jones’s or Lara Croft’s exploits,

sewer vents shaped like ancient Egyptian obelisks, Stonehenges

made from Cadillacs or a secret fascination with claims for

Egyptian pyramids in Queensland. Part of our assimilation into

a professional culture involves becoming intuitively aware of

what is ‘IN’ and what is ‘OUT’ as far as archaeology goes. Most

of us are clear-cut on what constitutes legitimate archaeology

and therefore we should not have much of a problem deciding

what is not legitimate or authentic. In this useful book Troy

Lovata challenges the idea that such a demarcation between

authentic and inauthentic archaeology is straightforward and

a simple matter of crossing a threshold. He further argues that

the professional archaeology community needs to become more

involved in understanding what happens with the many uses of

archaeology that are not directly controlled by archaeologists.

The book is written for undergraduates as an introduction

to the idea that archaeology also exists outside the boundary of

the professional domain. Each chapter is self-contained, with a

bibliography and a page of critical questions and class exercises.

The case studies are broad-ranging – the Piltdown fake (which

is told in cartoon form), a modern recreation of an Anasazi

cliff-dwelling, torreons (cylindrical Spanish towers) in the

American southwest, the work of three artists who incorporate

archaeology into their work and a fi nal chapter on fake henges

and Easter Island moai. Each case study looks at one or more

aspects of the boundary issue, and the case studies present

increasingly complicated situations. The deliberate hoax of

Piltdown presents an initial clear-cut demarcation between

real and fake archaeology, but as the case studies proceed

Lovata introduces more complex situations. The Manitou

Springs site in his next case study is an Anasazi cliff-dwelling

of the same type as Mesa Verde, but built about a century ago,

outside the prehistoric Anasazi area, but using remains from

an authentic ruin transported to the site. Still clearly fake

but, like excursions to Old Sydney Town and Sovereign Hill,

affording an opportunity to access and interact with the past

that is not possible with whatever is left of the ‘real thing’. The

interpretation of the site stresses the support of pioneering

archaeologists in creating an accessible and generally accurate

facsimile, but also underplays the fact that it is not a real ruin.

Lovata explores the consequences of this approach and how the

public and professional communities respond.

The Manitou Springs case study introduces the idea that

archaeology is not just what archaeologists regard as authentic.

There are other audiences, including the public, Indigenous

and ethnic groups who use the past in ways that are different

to those of the professional heritage community. The torreon

chapter brings home the idea of ethnic identifi cation being

reinforced by the use of buildings and motifs from the past. In

this case torreons, cylindrical defensive towers built in Spain

and then a feature of Spanish settlement in the US southwest,

mark a specifi c ethnic affi nity. Modern forms of the torreon

are now included in advertising, fast food joints and public

art. Are these fake or just the most recent phase of the long

history of that building type? The chapter also makes the point

that use of motifs of the past is not passive but also ideological

and have different readings through time and to different

audiences. This is reinforced in the fi nal chapter, a fun look at

the construction of fake Stonehenges and Easter Island moai in

the United States.

Inauthentic Archaeology covers two related but distinct

issues – appropriation and use of the past and appropriation of

archaeology. Appropriating the past is nothing new – it is one

of the ways that power is legitimised or authority undermined.

With Piltdown, Manitou Springs and torreons Lovata examined

how images of the past were used in modern contexts. In his

next chapter, focusing on three artists, he looks at how these

artists use archaeological methodology. Artist Mark Dion, for

example, stages art performances that have involved the creation

of archaeology-style digs where ‘sites’ such as the banks of the

Thames are gridded-out, collected, the artefacts analysed and

classifi ed and ultimately displayed. In effect, he is parodying the

sort of archaeology that we do, and subverting it by showing

that it is possible to go through the motions of survey, recovery

and classifi cation without being an archaeologist at all. This

has sometimes provoked a surprising degree of criticism from

archaeologists, based on Dion’s own comments in his interview

with Lovata and online discussion group comments. While

some of this is possibly philistinism, at least some of it must

be from the discomfi ture of archaeologists being shown up as

‘performing’ a role.

In the study notes Lovata asks ‘Is archaeology a set of

artifacts? Is it a series of stories about the past? Or is it a certain

way of acting, dressing, and looking at the world?’ (p.134).

These questions come to the core of Lovata’s argument. While

most archaeology is unarguably authentic, there is a grey area

that brings us into the realm of different uses for the past (e.g.

torreons and carhenges), and in using archaeology as a means

of reaffi rming an object’s authenticity or worth (e.g. Manitou

Springs or Mark Dion’s work). The case studies bring out the

issues involved in these situations, and the commentary and study

notes set them into a broader context, and provide guidance for

discussion or further enquiry.

Why is any of this important? Lovata’s argument is that as the

creators of the knowledge of the past we have a responsibility to at

least know to what uses that information is put by others. He also

argues strongly that archaeologists have public stewardship roles

in not only discovering stuff, but also telling the stories of the past

as an ethical obligation. To do this successfully sometimes means

using other tools – hence the Piltdown cartoon, understanding

how artists tell archaeological stories, the use of fake ruins to

make the past come alive and so on. Underpinning all of these

is a concern that unless archaeologists understand what is going

on and take some ownership of the process by which their data

is used, then they will lose the ability to create enduring benefi ts

for public knowledge and heritage conservation. Archaeologists

have lost their position as the main voice of authority about

the past, if they ever had it, and now contend with many other

groups to be heard.

I think the book certainly succeeds as a primer on the subject,

although the exclusive focus on the US, apart from Piltdown,

may serve to reinforce a prejudice that it is something that

only foreigners do. That would be disappointing, as the lessons

are directly applicable to an Australian context. Australian

archaeologists have had decades of trying to negotiate and

maintain a professional culture that is constantly being challenged

by Indigenous people, and has learned the hard way that it cannot


Book Reviews

maintain the sole authoritative voice about the past. Inauthentic

archaeologies abound here as much as anywhere, from claims

for secret visitors to Australia (the Gympie pyramid, Stradbroke

Island shipwreck, the Mahogany Ship), use of historical imagery

for ideological purposes (Eureka fl ags waved by everyone from

neo-Nazis to the far Left), pretend heritage from the educational

to the surreal (Kryal Castle) and a continuing debate about

whether there is such a thing as a single national history and a

question of what place material evidence has in it.

The book is a good introduction to identifying and defi ning

inauthentic archaeology. As an introductory text, Lovata

portions his argument and evidence out in small self-contained

chunks that reinforce one point at a time. I would recommend

it to anyone interested in the broader role of archaeology and

the use of archaeological information in society. While it is a

primer, it provides one of the few book-length treatments of

the subject and raises questions that are relevant to current

professional archaeological practice in Australia. What is

valuable about this book is that it does not shy away from

engaging with inauthentic archaeologies. They do not have to

be our secret pleasures or shames.

LITHICS IN THE LAND OF THE LIGHTNING BROTHERS: THE ARCHAEOLOGY OF WARDAMAN COUNTRY, NORTHERN TERRITORY

Chris ClarksonTerra Australis 25, ANU E Press, Canberra, 2007, xvii+221pp,

ISBN 9781921313288

Reviewed by Richard A. GouldDepartment of Anthropology, Brown University, Providence, RI

02912, USA

In northern Australia much of what we know about prehistory

is derived from stone tools and tool-making debris, mainly

because of the durability of lithic materials in the archaeological

record. How much of the totality of past human behaviour can

archaeologists reclaim through this body of evidence? The author

has made a rigorous and commendable effort to probe the limits

of inference based on lithic technology. This study centers on the

lithic reduction process as an analytical platform for two goals:

(1) characterising changes and variability in stone technology

in the interior of northern Australia about 120km southwest of

Katherine, Northern Territory, from the late Pleistocene until

around 1500 BP, and (2) drawing inferences about changes in

adaptive behaviour linked to climatic changes that affected the

resources required for Aborigines living directly off the land.

The author does a fi ne job of analysing lithic technology

in relation to strato-chronological evidence. His analysis

conforms to what I have referred to elsewhere as ‘Crabtree’s

First Law of Lithic Reduction’ which highlights how the last

step in producing a stone artefact can obliterate all or most

preceding steps. This observation provided a rationale for

studying quarries and workshops in order to fi nd evidence of

steps in the production process (including unfi nished, broken,

and rejected pieces) and did much to redirect attention by

archaeologists in North America away from an over concern

for fi nished bifaces, like fl uted points. The concept of the

lithic reduction process followed from this assumption and

continues to be applied effectively and refi ned to account for

local and regional variations in prehistoric stone tool-making

– as demonstrated in this volume.

Clarkson’s analysis identifi es some important points about the

culture-historical sequence of lithic technology in the study area.

The lithic reduction sequences there reveal a dendritic model of

changes that is non-linear, and they also reveal variable rates of

change in different components of the lithic assemblages. While

lithic reduction is linear in the sense that it is always a subtractive

process – involving removal of material from a piece without

any way of building the piece back up – it is non-linear in the

way some variables like discard rates peak, dip, and peak again

through time. These fi ndings should not surprise archaeologists,

but it is helpful to see them worked out in detail with reference

to the chronology of site assemblages at four stratigraphically-

excavated rockshelters; Nimji (a.k.a. Ingaladdi), Garnawala 2,

Gordolya, and Jagoliya. For example, the ‘event tree’ in Figure

6.3 encapsulates changes in core forms resulting from different

modes of reduction and the relative frequencies of each stage in

these excavated assemblages. This is followed by a discussion of

shifts in tool-making strategies and the technical and behavioural

factors that may account for these shifts.

Given the careful attention to lithic reduction sequences and

stratigraphic relationships throughout this volume, it seems

odd that more detailed attention was not paid to the matter

of performance as defi ned by Schiffer and Skibo (1987, 1997)

and cited by Clarkson. The author presents good evidence for

the relative use-lives of different kind of tools, but the physical

basis of this performance relative to the types of lithic raw

materials used is hardly discussed. Similarly, the author’s ‘start-

up costs’ in producing certain tools like tulas and points take

in obvious factors like manufacture, hafting, and the objects

to which they were hafted, but only partially explore the effort

(cost) of obtaining the lithic materials, especially if the source

was not local to the site area. The question of local versus non-

local sources is never fully defi ned or analysed, and the lithic

landscape is not treated in detail. In short, the direction of the

lithic analysis is clear and convincing, but the conclusions are

still somewhat incomplete.

Another anomaly worth mentioning here is the relative

scarcity of ground-stone pieces. Only three were reported

from Nimji, one from Garnawala 2, two at Jagoliya, and none

from Gordolya. So for all practical purposes, ‘stone artefacts’

in this monograph means only fl aked stone tools and debris.

This raises the question of direct versus indirect indicators of

economic activities, especially in relation to ecological stressors

like prolonged drought. In Chapter 2 the author launches into

a review of optimal foraging theory as it pertains to human

hunter-gatherers by examining variables such as dietary

breadth, central place versus fi eld-processing of resources,

encounter rates and patch choice, mobility and settlement

patterning, and other factors that have fi gured prominently in

a growing body of literature on the subject that was adapted

originally from the biological sciences. Taken by itself, this is an

excellent critical and concise review of this subject, and its best

feature is the way it addresses risk. This factor was neglected at

fi rst in some biological models that emphasised optimality over


Book Reviews

risk mitigation. Today’s optimal foraging theorists, however,

have become much better at integrating risk into their models.

Clarkson’s treatment refl ects this modifi ed view of optimality,

which archaeologists and anthropologists working in Australia’s

arid and semi-arid zones have come to appreciate.

Taking this amended optimal foraging model along with

different archaeologists’ commentaries, the author offers

predictions (test implications) that should apply to changes in the

lithic assemblages within the study area from the late Pleistocene

onward. This kind of hypothesis-testing approach has its place

in studies of this kind and can serve as a useful framework for

considering alternative kinds of behaviour by mobile hunter-

gatherers under ever-changing ecological conditions. But this

kind of reasoning sometimes requires piling assumptions on

top of each other, and care is needed to avoid exceeding the

reasonable limits imposed by limited physical evidence, in this

case fl aked stone tool technology. The author explores these

hypotheses and their test implications in a rigorous and detailed

manner, but the question remains: Are there more convincing

ways to infer changes in past economic and adaptive behavior

more directly, with fewer assumptions?

When Clarkson states, ‘the study of stone artefact assemblages

has the potential to contribute a vital and unique perspective

on the past because they provide a tangible record of human

behaviour intimately linked to the means by which people

extracted a living from their environment’ (pp.162-163), he

takes the position that, however indirect and inferential they

may be, there are predictable linkages between lithic technology

and adaptive behaviour among marginal hunter-gatherers. What

about more direct sources of information about how the physical

evidence of archaeology connects with ancient adaptive behaviour,

such as faunal studies, residue analysis, bone and soil chemistry,

and phytoliths? Direct approaches like these are becoming

increasingly important in archaeology and are providing

alternative pathways for hypothesis-testing that require simpler

hypotheses and fewer assumptions. To his credit, Clarkson is

always careful to note how lithic reduction and archaeological

sequences only suggest, but do not demonstrate or prove, the

validity of his hypotheses. There is, in fact, empirical evidence

from ethnographic studies of Aboriginal Australians to suggest

that some of these hypotheses may not be borne out as predicted,

although this in no way negates their value as a framework for

evaluating adaptive behaviour. So let the testing proceed ...

From a regional perspective, Australian archaeologists will

fi nd Clarkson’s volume a useful guide to comparative analysis

of lithic assemblages in other areas. For now, however, they

will probably fi nd the author’s conclusions on technological

changes more compelling than his further inferences about

palaeoecological and economic adaptations. Clarkson’s

monograph defi nitely raises the bar for the controlled analysis

of lithic technologies in Australian prehistory and represents

a positive contribution that other Australian archaeologists

should consider seriously in relation to their own studies.

ReferencesSchiffer, M.B. and J.M. Skibo 1987 Theory and experiment in the study of

technological change. Current Anthropology 28:595-622.

Schiffer, M.B. and J.M. Skibo 1997 The explanation of artifact variability.

American Antiquity 62:27-50.

BOX OFFICE ARCHAEOLOGY: REFINING HOLLYWOOD’S PORTRAYALS OF THE PAST

Julie M. Schablitsky (ed.)Left Coast Press, Walnut Creek, CA, 2007, 256pp,

ISBN 978-1-59874-056-1

Reviewed by Peter HiscockSchool of Archaeology and Anthropology, Australian National

University, Canberra, ACT 0200, Australia

This curious book will appeal to archaeologists who are annoyed

when Hollywood constructs fi lm plots that do not conform to

their understanding of the evidences and, far more importantly,

to teachers who intend to teach history through a commentary

on fi lmic representations. Containing 13 chapters, this edited

volume brings together a number of authors who comment on

the historical reality of fi lm depictions and the archaeological

evidence for events displayed in movies. The collection emerged

from papers given at a conference of the Society for Historical

and Underwater Archaeology held in New York, and the fi lms

discussed refl ect the maritime, historical, and very American

focus of the participants. Hence, with the exception of chapters

on the way Hollywood represents mummies and Vikings, the

chapters focus entirely on historical events in America or on the

way there. Hence the book offers considerations of Pocahontas,

Amerindians, African Americans, the Wild West, the gangs

of New York, historic Chinatowns, Confederate submarine

battles, and American pirates as well as the Titanic. These

are undoubtedly worthy and interesting topics, but this focus

seems parochial, and may not be central to the needs of history

teachers elsewhere in the world. Subjects well-represented

in American cinema are not explored: the crusades, building

pyramids, Alexander the Great, Roman politics and battles,

Palaeolithic life, land battles of the American revolution and

civil war, the bombing of Pearl Harbour and so on. Of course

this observation does not diminish the value of the book,

but it may well limit its value for non-American readers and

teachers. For those searching for wider coverage other books

may be worth considering (e.g. Jon Solomon’s The Ancient

World in the Cinema, Tony Barta’s Screening the Past, or Hughes-

Warrington’s History Goes to the Movies). However, what Box

Offi ce Archaeology provides is a focus on the archaeological

material, written by archaeologists, and perhaps this is just

what some readers are looking for.

More critically I note the existence of a tension in

almost every chapter, between recognising that Hollywood

productions have a very different objective to archaeological

research and the temptation to complain about historical

inaccuracies. The goal of many contributors, either implicitly

or explicitly expressed, is to correct misunderstandings of

the past purveyed by film-makers, an agenda reflected in

the subtitle of the book. The editor, Julie Schablitsky, frames

this in terms of the capacity of archaeologists to contradict

popular myths expressed on celluloid (p.12). The list of

corrections offered by the archaeologists in this volume is

very lengthy, ranging from the pertinent observation that

magical incantations found with mummies were meant to

revive them in the afterlife (not to physically reanimate them


Book Reviews

in this life), to the quirky complaint that films about Vikings

do not accurately depict sex with slaves about to be sacrificed.

Of course even observations of the most trivial ‘errors’ add

to the realisation that films do not offer us the kinds of

representations of the past that professional archaeologists

and historians would present in texts. Even academics

such as Robert Rosenstone (in History on Film/Film on

History), who encourage the expectation that history can be

communicated through cinematic representations, argue that

film is a radically different medium for delivering history and

cannot be bound by the previous rules of academic historical

analysis. Yet who could expect otherwise? Hollywood makes

films for entertainment and profit, not only to convey some

sense of historical events. Those films may reveal much about

contemporary stereotypes and mythologies, but historical

accuracy is not the central (or even the secondary) concern

of most directors. Whining about historical faults, minor or

major, in non-documentary movies is not only unlikely to

alter the way Hollywood operates but is an approach which

obscures the fundamental difference between movie-making

and archaeology. In an exceptional final chapter, Vergil Noble

embraces the reality that films about the past reflect mythic

imperatives rather than archaeological evidence, that this has

always been so, and that this is appropriate and laudable. He

sagely advises that archaeologists would be well-advised to

focus on how they can improve their own discipline rather

than attempt to remake Hollywood in their own image. Of

course we can all be fascinated by film representations of our

discipline, but ultimately technical critiques such as those

found throughout Box Office Archaeology must been seen as

statements about archaeology rather than about cinema.

SALVAGE EXCAVATION OF SIX SITES ALONG CADDIES, SECONDS PONDS, SMALLS AND CATTAI CREEKS IN THE ROUSE HILL DEVELOPMENT AREA, NSW

Jo McDonald Cultural Heritage Management Pty LtdAustralian Archaeological Consultancy Monograph Series 1,

Australian Association of Consulting Archaeologists Inc.,

St Lucia, 2005, xxx+488pp, ISBN 0 9590310 1 4

Reviewed by Fiona HookArchae-aus Pty Ltd, PO Box 177, South Fremantle, WA 6162,

Australia

The publishing of grey literature has long been a talking point in

Australian cultural resource management (CRM) archaeology

and this book is the fi rst in the Australian Association

of Consulting Archaeologists Inc.’s (AACAI) Australian

Archaeological Consultancy Monograph Series with the aim of

‘providing examples of best practice consultancy reports in

archaeology and cultural heritage management’ (p.ii). The

volume is a small but heftily edited version of Jo McDonald

and Beth White’s (2001) consulting report prepared for the

Rouse Hill Infrastructure Consortium. Set out as 13 chapters,

the book details the results of the salvage excavation of six

sites with the main research aim of investigating landscape

and material culture relationships. Additional research aims

included relating results to other studies in the Rouse Hill

area as well as the wider Cumberland Plain and investigating

technological organisation of specifi c identifi ed activities such

as knapping fl oors. The excavation programme was the result

of a series of surveys and test-pitting excavations.

The first three chapters place the project and the Rouse

Hill Development Area into context, including a discussion

of the Indigenous involvement in the project. Chapters 4

and 5 outline the field, laboratory and analytical methods, as

well detailing the rationale behind the flaked stone artefact

investigation. While one of the aims of publishing grey

literature is to provide the wider archaeological public with

the results of CRM projects, Chapters 1 to 5 provide both

students and CRM practitioners with examples of why and

how such analyses are conducted in the first place. The lack

of discussion concerning the design of the testing phase prior

to detailed open area excavation is disappointing. A section

on the design of the testing phase was included in the original

report as part of Appendix 5, but no appendices are included

in the published form.

The results of the excavation programme for each site are

presented in Chapters 7 to 11. These chapters detail the soils

and stratigraphy within each excavation. In all sites the only

cultural material recovered were fl aked stone artefacts, hence

the heavy emphasis on technological analysis. The recovered

artefact assemblage was analysed in terms of raw material and

reduction technology. I was surprised at the lack of detailed,

scaled technical plans for the excavations. Site plans do not show

topography, nor is it shown how the stratigraphy of the more

complex sites relates to topography, nor how topography relates

to artefact distribution.

The excavated artefact assemblages are then discussed

collectively in terms of: artefact densities and landscape, raw

material uses and inter- and intra-site spatial patterning,

technological strategies, and settlement organisation. The fi nal

chapter discusses the results of the excavations in relation to the

wider eastern Sydney region. The project results fi t within the

temporal, spatial and technological aspects of the eastern Sydney

regional sequence and allow the reader to fully comprehend the

depth of research conducted at Rouse Hill.

While I understand the cost imperatives involved in

reducing a large A4 technical report into a B5 book, the size

of the text and illustrations is sometimes frustratingly small.

Particularly difficult are the black-and-white graphs. Either

size reduction or the conversion from colour or greyscale

to black-and-white makes many graphs impossible to read

(see for example Figure 0.6). However, many of these issues

are mitigated by the availability of the monograph as a free

downloadable file on the AACAI website (www.aacai.com.

au). It is also problematic that the appendices have not been

included, presumably for reasons of space, since they are

referred to in the text. Notwithstanding these small matters,

this book provides a long-awaited example of detailed

research focused CRM project, and should become standard

reading for CRM practitioners and students.


Book Reviews

ARCHAEOLOGICAL THEORY AND THE POLITICS OF CULTURAL HERITAGE

Laurajane SmithRoutledge, Oxford, 2004, xi+260pp, ISBN 978-0-415-31833-4

Reviewed by Thomas F. KingSWCA Environmental Consultants, PO Box 14515, Silver Spring,

MD 20911, USA

This book purports to be about archaeology and cultural resource

management (CRM) in Australia and the United States, but it is

really only about archaeology. The introduction frames Smith’s

argument – which seems to be that through CRM-as-expression-

of-processual-positivism, archaeologists have provided

government with a tool for the oppression of Indigenous people.

Chapter 2, ‘The Cultural Politics of Identity’, elaborates on

this theme based on the history of archaeological/Indigenous

relations over the last 40 years. Chapter 3, titled ‘Archaeological

Theory and the ‘Politics’ of the Past’, asks whether ‘contemporary

archaeological theory (can) make sense of what archaeology ‘does’

within the context of CRM?’ (p.33). Smith doesn’t tell us why

this question is worth asking and I for one haven’t a clue. The

chapter is largely given over to a critique of processualism, and

a milder one of post-processualism, offering the unremarkable

observation that both ‘explain the construction of archaeological

knowledge and discourse with little or no reference to infl uences

external to the academy’ (p.56).

Ironically, Chapter 4, ‘Archaeology and the Context of

Governance’, offers ‘state theory’ as a remedy to this defi ciency,

but goes on to discuss abstract notions of governance with

practically no reference itself to anything ‘external to the

academy’. As an archaeologist who has spent the last 40+ years in

‘the context of governance’, I would have liked to have seen some

attempt to connect with the real world.

Chapter 5, ‘Archaeological Stewardship’, casts a jaundiced eye

on the claims of archaeologists to be ‘stewards’ of archaeological

resources, as these claims have informed the development of

CRM since the 1970s. Chapter 6, on ‘Signifi cance Concepts and

the Embedding of Processual Discourse in Cultural Resource

Management’, explains how archaeologists have tended to value

sites on the basis of their research potential, and suggests that

by doing so they have seized a position of dominance in CRM.

Chapter 7, ‘The Role of Legislation in the Governance of Material

Culture in America [sic: the United States] and Australia’,

discusses the development of the archaeological parts of United

States and Australian heritage legislation. The last two chapters

before a short conclusion are case studies – Kennewick Man in

the United States, Franklin Dam and related cases in Australia.

Throughout the book, Smith castigates archaeologists for

failing to respect and include interests in the past other than those

derived from the positivist scientifi c tradition. Unfortunately,

Smith’s treatise exemplifi es the narrow-mindedness she deplores.

Smith criticises archaeology for being ‘overly self-referential’,

but her book is a perfect example of self-reference. Consider,

for example, how Smith defi nes CRM. On page 1 she states

it means ‘the process and procedures, often underpinned by

public policy and legislation, used to protect, preserve and/or

conserve cultural heritage items, sites, places and monuments’.

Then, she defi nes it as ‘the processes, informed by public policy

and heritage legislation, that manage and protect Indigenous

cultural heritage, and in doing so, construct and defi ne relations

between archaeologists, Indigenous interests and government’

(p.9). CRM is again later defi ned as ‘the process concerned

with the management of material or tangible cultural heritage’

(p.195). These are pretty fair expressions of what archaeologists

in the southwestern United States, who more or less coined the

term in the 1970s, took (and generally still take) CRM to mean.

However, would any non-archaeologist defi ne it this way? Are

‘cultural resources’ really only tangible things? Are they really ever

tangible things? Is a dance form not a cultural resource? A song,

a story, a traditional belief? What about a community’s feeling of

identity, and an individual’s feeling of membership in a culture?

What about religion?

But, you protest, Smith is an archaeologist writing for other

archaeologists. Why should she not conceptualise CRM as other

archaeologists do? I see that there are three reasons. First, Smith’s

distaste for the infl uence of scientifi c positivism on CRM would

make no sense if she didn’t believe that other points of view, other

realities, should have access to and roles to play in the enterprise

of CRM. By failing to even consider broader-than-archaeological

defi nitions of the term, she displays as narrow an understanding

of the fi eld as do those she criticises. Second, by viewing CRM

through an archaeological lens, Smith fails to acknowledge the

roles played by innumerable other interests – for instance those of

taxpayers, property owners, practitioners of non-archaeological

heritage professions, government agencies (except as neocolonial

stereotypes) and local (non-Indigenous) communities. This

results in a portrait of CRM that is very incomplete and hence

misleading. Third, the appropriation of CRM by archaeologists

in the United States has caused government operations, notably

environmental impact assessment (EIA), to effectively ignore

those portions of the cultural universe that don’t happen to be

archaeological sites. This is rather a problem for people who

practice and value non-tangible aspects of culture – presumably

the kinds of people whose perspectives Smith would like to see

CRM accommodate. But Smith, like the archaeologists who

invented CRM, doesn’t seem to relate well either to living cultures

or to government operations. It is indicative that she manages

only two brief references to EIA despite its absolute centrality to

CRM and the opportunities it offers to involve and respect the

non-archaeological interests Smith says CRM ought to engage.

To Smith, CRM is all about archaeology – Indigenous

archaeology, that is. Smith never acknowledges the existence

of the historical archaeology of non-Indigenous populations,

any more than she does such companion CRM disciplines as

architectural history, historical architecture, and history. She

also gives strangely short shrift to the government agencies that

actually do CRM or cause it to be done. They are alluded to, but

only as shadowy abstractions somewhere in the background,

manipulating archaeologists and being manipulated by the

deus ex machina of ‘archaeological theory’. Nor do the regulated

industries and private development interests that mostly fi nance

CRM play much of a role in Smith’s worldview, and she seems

uninterested in the effects of the fi eld’s increasing domination

by profi t-making consulting fi rms. To Smith, the character of

the CRM enterprise, and its ills, result from the dominance

of processual archaeology in the 1980s when the fi eld was


Book Reviews

developing. Processualists are portrayed as allowing, or making,

archaeology become a ‘technology of government’ at the expense

of Indigenous interests. The world of politics and bureaucracy

within which CRM actually exists is of as little relevance to

Smith as it is to the unengaged archaeologists she criticises. And

the possibility that government might have some reason for its

use of ‘technologies’ like archaeology other than the colonialist

oppression of Indigenous populations appears foreign to her.

This book is very much a study by, for, and about academic

archaeology, and it describes a CRM that I at least don’t recognise.

Perhaps it’s an accurate portrayal of how things are in Australia,

but its image of the United States is a fantasy. Indicative of this

fantastical quality is that in Smith’s discussion of CRM in the

United States, Smith devotes only about two pages of badly

outdated description to the National Historic Preservation Act

1966 (NHPA) – the primary driver of the whole enterprise.

She gives much more attention to the Native American Graves

Protection and Repatriation Act 1990 (NAGPRA), including a

chapter-long discussion of the Kennewick Man case. This is an

interesting and fairly accurate discussion, but rather marginal to

the overall practice and challenges of CRM in the United States.

And yet, one of the most maddening things about this

book is that on some very basic points I think Smith is

right, and she expresses some useful notions, albeit buried

in convoluted prose. The implicit dominance of positivism

among archaeologists in the 1970s and thereafter (whether

processualists or culture historians) did help put archaeology

on a collision course with the rising tide of Indigenous

awareness and demand for respect. Archaeological ‘science’ is

widely assumed in government to represent truth, in contrast

with Indigenous ‘myth’. Archaeologists dominate the CRM

enterprise, and do impose their worldviews on its practice,

to the detriment of all others. Notions of the pre-eminence

of ‘scientifi c’ thinking are embedded in government policies

and procedures, biasing decision-making against Indigenous

values. Indigenous interests and values are now gaining

broadened political understanding and acceptance, however,

and many archaeologists are struggling to deal with their

relatively diminished authority. These are real issues, which

– among many others – could benefi t from serious exploration.

Unfortunately, Smith doesn’t provide it in this book.

Smith’s failure to connect with the world beyond the academy

also deprives her of some analytical opportunities. For example,

United States National Park Service’s insistence that all Tribal

Historic Preservation Offi cers who receive its grant support

maintain archaeologists on staff is an expression of archaeology’s

dominance of governmental thinking in CRM, but it has

created a context in which archaeologists are learning to respect

tribal views and values, and tribes are coming to understand

and appropriate archaeology. This sort of thing is apparently

unknown to Smith; aside from a few approving (and entirely

merited) references to the work of Larry Zimmerman, she seems

unaware that archaeologists in the United States are even working

with Indigenous communities, much less for them.

I could go on and on, but let me end with a more technical

quibble. As I worked my way through Smith’s book, I kept

thinking, with apologies to Lerner, Lowe and Henry Higgins,

‘Why didn’t the processualists teach their children how to write?’ A

minor irritant is Smith’s consistent use of the word ‘subsequently’

when she means ‘consequently’, but more generally, her style is

turgid, pedantic, and impenetrable – which seems strange in

someone who argues for inclusiveness.

THE ARCHAEOLOGIST’S FIELDWORK COMPANION

Barbara Ann KipferBlackwell Publishing, Malden, 2007, xvi+467pp, ISBN 1-4051-

1886-5

Reviewed by Ian J. McNivenProgramme for Australian Indigenous Archaeology, School

of Geography and Environmental Science, Monash University,

Clayton, VIC 3800, Australia

Students sometimes ask me: What is the golden rule of taking

field notes? Usually I answer saying your notes should stand

on their own such that another person could read them and

understand exactly what you did and be able to take all of

the materials you surface collected or excavated and relocate

these back in the landscape with 3D precision. As such, your

notes must relate clearly with your labelled materials. In the

case of excavation, obviously other issues are important such

as making sure your excavation units or spits are not too big

and that they do not cut across stratigraphic units. Where

do archaeologists learn such information? As students, much

can be learnt in university classes augmented by jumping at

every opportunity for practical experience helping out on

site surveys and excavations. In addition to social and ethical

issues, two key things you gain from practical experience are

physical familiarity with the realities of the archaeological

record and how to do fieldwork by doing it and watching

and quizzing the fieldwork director and other experienced

fieldworkers. However, all professionals know that a broad

range of skill levels exist amongst practitioners – some

archaeologists have extraordinary skills in deciphering

sedimentary changes and identifying stratigraphic units,

some don’t. How many archaeologists have the confidence

to publish backplots of spits onto stratigraphic section

drawings? For archaeology students, the learning experience

on fieldwork is increased greatly by being aware of the broad

range of issues and recording techniques associated with best-

practice fieldwork. This is where field method guidebooks can

be of enormous value.

Barbara Kipfer’s The Archaeologist’s Fieldwork Companion

is fi lled with technical and practical information on how

archaeologist’s record, excavate and to some extent analyse

sites (mostly terrestrial). It draws heavily on The Crow Canyon

Archaeological Centre Field Manual (2001) and as such has a strong

American slant. The volume is structured such that it functions

best as a reference guide to archaeological fi eld techniques. The

seven chapters are arranged alphabetically to emphasise the guide

function of the volume; 1. Classifi cation and Typology, 2. Forms

and Records, 3. Lists and Checklists, 4. Mapping, Drawing and

Photographing, 5. Measurement and Conservation, 6. Planning

Help, and 7. Resources. In other words, the volume is not a step-by-

step manual on how to undertake fi eldwork. Individual chapters

have hundreds of entries arranged in alphabetical order to assist


Book Reviews

easy and quick access to technical information. This encyclopedia-

like format is a good idea, but it does take some time to become

familiar with the categorisation of information.

As with any archaeology fi eld guide, deciding what

information to include and at what level of detail is a balancing

act. Some readers of Kipfer’s volume will wish more was said on

some topics and less on others. For example, while I appreciate

the volume is not a guide to archaeological excavation, more

examples of best-practice in terms of how to excavate a broader

range of different types of sites would have been useful. On

the other hand, the detail in some conversion tables can only

be described as overkill. As the volume has been written to

appeal also to ‘enthusiasts’, professionals will fi nd some of the

lists of animal species and stone tool types a little rudimentary.

Yet nearly 100 mock recording forms provide a wealth of ideas

for designing your own project-specifi c recording forms. And

in a sense, this is the strength of the volume – it provides a

broad range of technical detail on the types of information that

could be recorded in the fi eld and standardised techniques for

recording such information. There is something for everyone in

this volume. The wire spiral binding gives the added advantage

of being easy to use in the fi eld.

Recommendations on best-practice field techniques

generally translate well to the Australian context. I liked the

recommendation that ‘you can never have too many photographs’

(p.275) when it comes to recording excavations. With digital

photography, it is easy to run off 5–10 images for each spit.

The recommendation to sieve onto plastic (e.g. plastic tarps) is

excellent, as is the suggestion to locate the sieve area well away

from potential areas of excavation. How many photos have we

seen in publications of the sieve spoil heap located adjacent to

the excavation pit! In terms of sieve size, Kipfer points out that

the ‘standard screen size for archaeology is 1⁄4-inch [6mm] mesh’

(p.219). For most Indigenous archaeological sites in Australia,

6mm is too large as considerable proportions of stone artefact

and faunal assemblages would be missed. In Australia, most

skilled practitioners would recommend that 2 to 3mm should

be considered the standard screen size, supplemented by 1mm

mesh sieves where required and samples of sediment that pass

through the fi nal sieve for each spit.

For Australian archaeologists, the American slant of The

Archaeologist’s Fieldwork Companion should not dampen

your enthusiasm to purchase the volume. Indeed, such a slant

provides an opportunity for Australian practitioners to consider

the distinctiveness of Australian archaeology. In some cases, the

slant provides a few amusing asides – I now know how to treat

severe frostbite! And fi rst on the list of ‘fi eld etiquette’ is ‘no

obscene language or behavior’ (p.191) – if such was the case, then

90% of my fi eld crews would have been dismissed in the fi rst fi ve

minutes! For Australian archaeologists, I recommend this volume

as a useful companion to the core volume The Archaeologist’s

Field Handbook by Heather Burke and Claire Smith (2004).

ReferencesCrow Canyon Archaeological Center 2001 The Crow Canyon Archaeological

Center Field Manual. Retrieved 10 March 2008 from http://www.crowcanyon.

org/fi eldmanual.

Burke, H. and C. Smith 2004 The Archaeologist’s Field Handbook. Crow’s Nest,

NSW: Allen & Unwin.

LANDSCAPES, ROCK-ART AND THE DREAMING: AN ARCHAEOLOGY OF PREUNDERSTANDING

Bruno David Leicester University Press, London, 2002, xiii+235pp, ISBN

07185-0243-4

Reviewed by Tim MurrayArchaeology Program, School of Historical and European Studies,


Landscapes, Rock-Art and the Dreaming: An Archaeology of

Preunderstanding (LRD) seeks to encapsulate David’s views

about the social archaeology of Holocene pre-European

Australia, while situating itself within a stream of discourse about

history, anthropology and the Other. In seeking to undertake

an archaeology of the Dreaming, David gives us his take on

a range of issues that were core to post-processualism during

the late 1990s when that approach began to morph into more

generalised post-colonialist (as distinct from generally post-

modern) discourse. Along the way LRD imparts information

about rock art (especially arguments about its regionalisation

in Cape York), the archaeology of Ngarrabullgan and adjacent

sites, the archaeology of Dreaming rituals among the Arrente,

the antiquity of seed-grinding in Australia, and a rehearsal of

now quite conventional arguments for the development of more

complex social relations in late Holocene Australia that span the

scalar gamut from the pan-continental to the micro-regional.

All of this adds up to an ambitious agenda, made all the more

so by David’s attempts to move discussion of core aspects of

how Aboriginal people and their lives are understood away from

timeless essences towards conceptual frameworks that stress

dynamism, contingency, heterogeneity, and agency. Calls for

‘history’ and attacks on essentialism in Australian archaeology are

not novel, but the linking of such perspectives with arguments

about trends towards social complexity in late Holocene Australia

is a new twist. The bulk of the conceptual apparatus is derived

from standard post-processual positions of the 1990s with a

few idiosyncratic elaborations, especially Gadamer’s reading of

‘preunderstanding’.

LRD is not an easy read. It is densely written: repetitive,

discursive, and much concerned with self-conscious ‘internal’

dialogue. Its philosophical and methodological apparatus

(especially its focus on landscapes) is expounded with more

passion than clarity, and there is a strong tendency to discuss

complex conceptual matters in terms of big monolithic blocks

of discourse as if they are philosophically unambiguous – when

precisely the opposite is the case. When linked with David’s

method of working from present to past, and seeking in the past

states that explain the present (by virtue of their being antecedent

to it), this leads to a strong logical circularity, especially the fallacy

of affi rming the consequent. Nowhere is this more apparent than

in David’s discussion of the Dreaming.

For David, LRD is about using the archaeology of the

Dreaming as a vehicle for exploring Western constructions of

Aboriginality, and of continuing arguments against what he

(and others) has identifi ed as a damaging focus in Australian

archaeology on matters environmental and ecological.


Book Reviews

Broadening focus to a concern with the lives of people in all

their complexity has the potential to consider archaeological

phenomena not just as evidence of subsistence or of

environmental change, but as key documents for writing social

history. Another element of the story is an attack on the ‘timeless

Aborigine’, of the ethnographic present being retrodicted

millennia into the human history of Australia, so that social

archaeologists might populate its past with ‘real’ people. The

pitfalls of this type of analogical reasoning have been as widely

canvassed as its potential to enhance understanding.

David is at pains to stress his desire to tell a different story, one

of heterogeneity and dynamism. Of course, ever since Mulvaney,

even archaeologists of a positivist or environmental bent have

accepted the reality of change during the human history of

Australia, whether it be in technological or environmental

contexts, or geographical distribution. By the same token they

have also had to grapple with the power of the present as setting the

exemplar of how the Aboriginal past is to be understood, thereby

establishing the inferential power of the ‘timeless Aborigine’.

Signifi cantly, creating an image of what pre-European Aboriginal

societies were like that does not privilege those images created

by ethnography in the nineteenth and twentieth centuries, poses

very diffi cult problems for Australian archaeologists of whatever

persuasion. The reality of dynamism and transformation, of

heterogeneity as distinct from homogeneity, and of change

resulting from a wide range of variables (whether or not they

are classed as social or environmental), requires us to develop

plausible frameworks for demonstrating the social history of pre-

European Australia that embrace the instability of the concepts

and categories we use to understand the present.

LRD, despite its constant hype about undermining of ways of

looking at the world, does not succeed in doing this. Of course

making an argument that everything (even the Dreamtime) has

a history requires more than simple assertion. While we might

be sympathetic to the idea that worldviews change and that those

in any part of Australia were not homogenous at contact, nor

were they unchanging since the fi rst settlement of Sahul tens of

millennia ago, developing this perspective into something we can

get our theoretical teeth into requires more than frankly risible

notions of preunderstanding. This entirely begs the question of the

archaeological content of a concept like the Dreaming (or indeed

of ‘preunderstanding’), which is most assuredly not answered by

comments such as: ‘Aboriginal knowledge was always founded

on a prior truth steeped in the Dreaming’ (p.4).

In David’s hands, the Dreaming begins to look like it has all

the archaeological virtues and vices of a concept like culture,

perhaps the grand-daddy of all usefully ambiguous concepts.

David has this to say: ‘Because things are meaningful, people

engage with their world in culturally patterned ways. Objects are

engaged through historically and culturally ordered behaviour,

resulting in an ordered archaeological signature that is the mark

of preunderstanding’ (p.7).

But if all is in fl ux, what can be identifi ed, what can be analysed?

David appreciates that there is a signifi cant problem here:

The way in which interpretation and understanding are mediated

by preunderstanding is thus an expression of specifi c relations

between the engaging object in an already meaningful world,

and the interpreting subject. These relations are never fi xed; they

are continuously in a state of becoming. As human identity is

constructed through engagement, so too is human identity in a

continual state of change. As a result, both interpretation and

preunderstanding are in constant state of fl ux, being redefi ned

by the co-engagement of experiencing and communicating

agents and objects through time. It is in the nature of such

relations between people and their engaged material world that

an archaeology of preunderstanding, ‘mind’ and human identity

can take place (p.8).

Though this is tricked up as habitus in David’s exposition,

it is still culture, and it still requires us to freeze a dynamic set

of interactions between people, things and the world in which

they live into something that can be observed and analysed in

cultural terms.

Thus the goal of capturing dynamism in this way requires

David to make things static. And we haven’t even got to the point

of discussing the ways in which this kind of approach can be

plausibly operationalised archaeologically, where signifi cant issues

related to the structural properties of archaeological records as

records of human action exist. If archaeological sites (especially

those of the pre-European past in Australia) are frequently

palimpsests of aggregations of samples of human action, then it

requires us to be properly sceptical of approaches that read back

from an ethnographic present as if this warranted the assumption

of there being pretty straightforward correspondences between

archaeological and ethnographic observables.

Thus establishing the antiquity of the Dreaming is a far more

problematic business than asserting that all cultural data of the

last 1500 years or so were the product of a cultural context glossed

as ‘the Dreaming’, and that cultural data from periods antecedent

to that were the products of ‘preunderstanding’. I don’t want to

labour this point for too much longer, but the circularity (and

inherent instability) of this kind of reasoning should at least be

recognised. But this does not seem to concern David:

Sometimes the Dreaming as we know it today will be implicated

in the material products of past human behaviour in Australia; at

other times it will not be. Whether or not the Dreaming we know

from ethnography will be implicated in more ancient times will

depend on whether or not past behavioural conventions and their

operative contexts are recognized to have been akin to those of

recent times. My objective is not to retrieve the experience of past

meaningful landscapes, but to track back in time the antiquity of

the recent Dreaming’s ordered material expressions to identify its

historical emergence (p.8-9, original emphasis).

I have no doubt that that there will be archaeologists,

anthropologists and historians who will be attracted by LRD

and its attempts to historicise a conceptual will-o-the-wisp like

the Dreaming. Certainly the notion that the Aboriginal world of

the ethnographies cannot be assumed to have been the product

of timeless, unchanging, essences is worth rehearsing again. But

dealing with the consequences of this invocation of dynamism

and change (both for archaeology and anthropology) requires

a great deal more theoretical development and philosophical

clarity than we see here. The proposal that we track back from

present to past is as old as the earliest prehistories of Lubbock and

John Evans. The fear that by doing this we run the risk of making


Book Reviews

the past an eternal ethnographic present is of similar antiquity.

In this sense it does not advance us by simply asserting that the

Dreaming was a conceptual milieu that ordered the material

things recovered by archaeologists. While this might have some

limited validity in the analysis of rock art, it is highly dubious

when applied to other archaeological manifestations.

A COMPANION TO SOCIAL ARCHAEOLOGY

Lynn Meskell & Robert W. Preucel (eds)Blackwell Publishing, Malden, 2004, xi+430pp,

ISBN 978-14051-5678-3

Reviewed by Alistair PatersonArchaeology, M405, University of Western Australia, 35 Stirling

Highway, Crawley, WA 6009, Australia

The idea of a social archaeology has a long pedigree, and is

reinvented for each generation of archaeologists. It is a term that

has gained currency in recent years, with a journal badged in that

name in press since 2001. A Companion to Social Archaeology is

a solid compendium of both past and cutting-edge research into

social archaeology. What is meant by ‘social archaeology’ today?

The editors tackle this from a historical perspective in their

introduction to Part I, settling fi rst upon Hall’s (2001) defi nition

that ‘Social archaeology refers to the ways in which we express

ourselves through the things we make and use, collect and

discard, value or take for granted, and seek to be remembered by’

(p.3), before highlighting that ‘every form of political economy

requires its own history and past narrative’ (p.3), and arguing

for a greater infl uence of archaeology in modern societies.

The introduction is worth mulling over, for it provides some

historical depth to the issue. Meskell and Preucel see the origins

of a social archaeology in the 1930s with V. Gordon Childe’s

attempt to direct the attention of archaeologists to the study of

past societies, and in Grahame Clarke’s (1939) Archaeology and

Society. They highlight how Childe, Clarke and others, such as

K.C. Chang, were aware that the practice of archaeology was

linked in part to modern social and political interests, a self-

awareness central to today’s social archaeology. The arrival of

processual approaches saw social aspects of past people relegated

to a subset of a cultural system, and the authors refl ect on the

work of Lewis Binford, and Colin Renfrew’s characterisation

of social archaeology as interested in the social unit and its

relationship to the material record available to archaeologists.

The postprocessual era, deriving from Ian Hodder’s critique of

processual archaeology and later the work of Michael Shanks

and Christopher Tilley, saw a general rejection of the focus on

economic aspects of past societies in favour of the social and

political aspects as well as a renewed self-awareness regarding

archaeology as a practice conducted in the present and thus

socially and politically ‘embedded’. With a potted history behind

them the editors and their contributors move on to consider

the last quarter century. Some contributors in this volume meet

the challenge of bridging the concerns of processual and post-

processual archaeology in contemporary practice. Others raise

issues that have grown in importance in recent times, such as the

importance of non-Anglo-American approaches to archaeology,

Indigenous approaches to the past, and the relationship between

the practice of archaeology and modern communities.

I have read edited volumes recently which while well-

intentioned did not have the quality and scope of content to

tackle the intended topic – that is not the case in this volume. It is

thorough, with 17 chapters by experts in their respective areas. The

quantity of chapters makes a thorough description unattractive,

and instead I will briefl y detail the four sections of the book:

Knowledges, Identities, Places and Politics.

The topic of ‘Knowledges’ requires reviewing differences in

archaeological thought, with the most signifi cant distinction being

between processual and post-processual approaches. Ian Hodder’s

and Bruce Trigger’s chapters take this issue to task, with Hodder

expanding on the editors’ introduction to explore ‘the social in

archaeology’, and Trigger reviewing cross-cultural comparisons in

studies of the origin of state societies. Thomas Patterson’s review

of the infl uence of Marxist thinking on archaeologists provides

a comprehensive literature survey well beyond the boundary of

archaeology. Issues of embodiment, gender, sexuality and age

are considered by Rosemary Joyce who provides a useful review

of the last decade of work on these topics in archaeology. Clive

Gamble and Erica Gittins argue there is no social archaeology

without an understanding of the Palaeolithic, and go on to explore

approaches to a social archaeology of the Palaeolithic, arguing that

such approaches ‘would signifi cantly alter not only the Paleolithic’s

relationship with the rest of archaeology, but also the relationship

the West has with its own identity and past’ (p.113).

The section on ‘Identities’ brings together contributions

concerned with the historicity of the archaeological study of

identity, and recent trends in the area, including within gender

and feminist archaeologies, and in studies of race, class and

embodiment. Roberta Gilchrist reviews the recent increased focus

on age and aging, from studies of childhood to the culturally and

historically contingent understandings of the life experiences of

women and men. Chris Gosden reviews how archaeological and

anthropological constructions of self and other, and the present

and past, have been shaped by their colonial and post-colonial

context. He considers the relevance and potential of recent

anthropological and other theory informed by post-colonial

thought regarding personal and group relationships with the

material world. Victor Buchli reports on the construction of the

capital of Kazakhstan and how architectural styles refl ect the

complexity of the construction of identity, even in the recent past

let alone in prehistory. Paul R. Mullins reviews an area of core

relevance to historical archaeology, namely the archaeological

study of consumption over time.

In ‘Places’, Emma Blake provides a useful overview of recent

theories of space in the social sciences, predominantly geography,

as well as archaeology. Additionally, she considers archaeological

studies of urban contexts and of border/contact zones before

providing a case study of spatiality of the ancient Phoenician colony

of Motya (Sicily). Wendy Ashmore provides a very comprehensive

review of recent archaeological studies of landscape with a focus

on social archaeological perspectives. Julia A. Hendon argues for

the study of household production and social relations with a focus

on individual agency, daily practice, and economic production at a

household level. Ian Lilley’s chapter is about diaspora theories; he

explores the archaeological potential of these against the evidence

of Lapita-using people.


Book Reviews

The recent politicisation of archaeology and heritage by events

such as the looting of the Baghdad Museum and the destruction of

heritage in the Bamiyan Valley serves to introduce the fi nal section

on ‘Politics’. Reinhard Bernbeck and Susan Pollock examine

archaeology in the Middle East: who is involved, how they are

involved, and archaeology’s links to contemporary trends such

as globalisation and the breakdown of nation states. Gustavo C.

Politis and José Antonio Pérez Gollán consider the situation in

Latin America, and the sociopolitical contexts in which archaeology

has developed in this diverse region. Randall H. McGuire provides

a historical review of two North American competing histories

– (1) Native American histories sustained despite colonial and

nation state contexts; and (2) archaeologically-informed and, until

recently, largely non-Indigenous explanations of the past. Finally,

Koji Mizoguchi considers the history of archaeology in Japan and

the role of archaeology during the formation of the nation.

The strengths of the volume are clear to me. The volume works:

the four sections at fi rst seemed like an unnecessary contrivance,

yet they have presented a clear – yet not singular – argument for

what social archaeology is today. On the whole each contributor

has made a serious contribution in their chapters, and has provided

a guide to further reading. Both professionals and students will

fi nd the volume a very useful reference.

In this archaeological pilgrimage A Companion to Social

Archaeology has been a well-met companion on the road.

ReferencesClarke, G. 1939 Archaeology and Society. London: Methuen.

Hall, M. 2001 Social archaeology and the theatres of memory. Journal of Social

Archaeology 1:50-61.


RE-EVALUATING THE AUSTRALIAN SMALL-TOOL TRADITION: A PERSPECTIVE FROM HAZELWOOD, VICTORIA

Joseph Alexander BrookeB. Archaeology (Hons), Archaeology Program, School of

Historical and European Studies, La Trobe University, October

2006

This thesis presents the analysis and interpretation of mid-

Holocene stone artefact assemblages from the Driffi eld/

Hazelwood area, eastern Victoria. Investigations were designed

to generate information about the relationships between stone

technology, subsistence activities and mobility patterns. It is

argued that during the mid-Holocene, tool types, technologies,

raw material use and reduction strategies indicate that people

practised relatively mobile foraging strategies and that there was

some form of change more recently, during the late Holocene,

when people were less mobile.

Results from Driffi eld/Hazelwood were compared to studies

of temporally comparable stone artefact assemblages from other

parts of southeastern Australia, including Gariwerd/Grampians,

Wilson’s Promontory, Capertee Valley and the southern coast of

New South Wales. Similarities and differences in the composition

and characteristics of mid-Holocene stone artefact assemblages

in southeastern Australia refl ect a relatively unifi ed technology.

Some variations in assemblage composition north and south

of the Murray River are evident and some possible infl uencing

factors for these discussed.

These discussions provide the basis for assessing the different

classifi catory schemes used to describe mid-Holocene changes

in stone technology. It is suggested that the characteristics and

signifi cance of the concept of the Australian small-tool tradition

need to be re-evaluated. Toward this goal, consideration is given

to the explanations that have been offered for the introduction

of microlithic industries in other parts of the world. It is argued

that rejection of the Australian small-tool tradition, as some

have suggested, is premature and more consideration should

be given to the causes of regional and temporal variation in the

characteristics of these assemblages, as Gould fi rst advocated

when he introduced the concept of the small-tool tradition to

Australian archaeologists in 1969.

AN ANALYSIS OF FLAKED STONE ARTEFACTS FROM KUTIKINA CAVE, SOUTHWEST TASMANIA, AUSTRALIA

Jennifer BurchB. Archaeology (Hons), Archaeology Program, School of Historical

and European Studies, La Trobe University, October 2007

It has long been argued that the late Pleistocene archaeological

records of southwest Tasmanian sites are highly patterned and

refl ect a long-lasting and regionally restricted behavioural

‘system’. In a recent analysis of fl aked stone artefacts at Bone

Cave, Holdaway (2004) concluded that there was little evidence

for technological or typological change over a period of some

20,000 years. Although changes in patterns of raw material use

are linked by Holdaway (2004) to shifts in mobility patterns, the

results of the Bone Cave analysis encouraged retention of the

concept of a distinctive southwest Tasmanian ‘archaeological

signature’. However, the impact of scale, resolution, and

sampling on the behavioural information that can be generated

from these assemblages has received little attention. It is

possible that rather than refl ecting behaviour, the uniformity

seen at Bone Cave is simply the result of shallow, confl ated

deposits, the use of large analytical units representing long

time spans, and the analysis of small assemblages excavated

from spatially restricted deposits. To examine these issues, this

project investigated late Pleistocene fl aked stone artefacts from

Kutikina Cave (previously Fraser Cave) and compared these

with the stone assemblages from Bone Cave.

The thesis presents a systematic analysis of a portion of

the Kutikina Cave stone assemblage, investigating patterns

of change and continuity through time in raw material use,

technology and the types of artefacts produced. Results indicate

that there were changes over time in the way that raw materials

were used, the way cores were reduced and in the types of tools

produced. Stone assemblages dating to between c.17,000 BP and

c.15,000 BP at Kutikina Cave are compared to those excavated

from chronologically commensurate deposits at Bone Cave to

examine regional variability within the southwest system. The

thesis concludes that the differences between the Kutikina Cave

and Bone Cave stone assemblages are likely to be a function of

the ‘telephone box’ style excavations necessitated by the extreme

richness of archaeological deposits in southwest Tasmanian

cave sites. The analysis of the Kutikina Cave stone assemblage

indicates that the archaeological record of southwest Tasmania

is highly variable.

ReferencesHoldaway, S. 2004 Report of the Southern Forests Archaeological Project: Continuity

and Change: An Investigation of the Flaked Stone Artefacts from the Pleistocene

Deposits at Bone Cave, Southwest Tasmania, Australia. Volume 2. Bundoora:

Archaeology Program, School of Historical and European Studies, La Trobe

University.

WALKING BETWEEN TWO PARADIGMS

Bettyann DoyleBA (Hons), School of Social Science, University of Queensland,

June 2007

In this thesis I examine Barada people’s dilemma of ‘walking

between two paradigms’ when recording and documenting their

cultural landscapes. Owing to legislative requirements, Barada

are forced to identify with and embrace Western processes of

recording cultural landscapes (rigid lines marking individual

ownership) as opposed to their Indigenous concepts of

boundaries (fl uid and identifi ed by Dreaming tracks). Barada

Barna Kabalbara Yetimarla (BBKY) is a Geographical Information

System tool used by Barada in Central Queensland. I examine

THESIS ABSTRACTS


whether the BBKY GIS software differs from generic Western-

designed and operated GIS software and whether BBKY meets

the needs of its Indigenous users and, if so, who holds control and

power over these mapping processes. Semi-structured interviews

with Barada are used to explore tensions in recording Barada

cultural landscapes and are analysed in the context of recent

debates about Indigenous knowledge and Indigenous cultural

heritage management. A pattern emerges demonstrating that

although archaeologists and researchers from other disciplines

use GIS for recording landscape boundaries, there are problems

in this approach for Barada owing to the fact that it is diffi cult to

digitise narrative and the fact that Indigenous landscapes do not

exhibit rigid boundaries. I discuss the continuum of control that

Barada have in mapping their cultural landscape and argue that

BBKY sits at the intersection between two knowledge systems.

Finally, I argue that BBKY as a tool enhances Barada’s control

and power over their cultural heritage.

ARCHAEOLOGICAL RESIDUE AND STARCH ANALYSIS: INTERPRETATION AND TAPHONOMY

Michael HaslamPhD, School of Social Science, University of Queensland,

November 2006

Microscopic analyses of artefact residues and sediment

microfossils follow a sequence of sample selection, observation,

description, identifi cation and interpretation. Each of these stages

requires separate, but inter-related, methodologies contributing

to the eventual published or disseminated results. Archaeologists

are increasingly using residue and microfossil identifi cations and

interpretations to discuss artefact function and site development,

making inferences about past subsistence and other activities

that in turn impact on broader debates about past economic and

social practices. It is the contention of this thesis that the stages

of archaeological microscopic analysis themselves require closer

examination, as a means of assessing and progressing the viability

of reconstructions drawn by residue and microfossil analysts.

In particular, three components are examined: identifi cation,

taphonomy, and the underlying theoretical framework of

residue and microfossil interpretation. These components are

investigated using light microscopic analysis of ancient starch

and stone tool residues, with a focus on research conducted, and

by researchers based, in the Australasian region.

A selection of published peer-reviewed papers forms

the central chapters of the thesis. These studies investigate

archaeological starch identifi cation and misidentifi cation,

ancient starch taphonomy (including decomposition,

preservation and movement through sediments) and the value

of social archaeological theory in archaeological stone tool

residue analysis. The roles of analytical scale and narrative

presentation are explored as one way of coming to terms

with, and communicating the fi ndings of, typically small-scale

analyses that record the results of very specifi c past actions.

Original introductory and concluding chapters contextualise the

research within current and past trends in microscopic residue

and microfossil interpretation. This discussion includes the

infl uence of sample sizes over the validity of stone tool residue

analyses and the place of quantitative and qualitative approaches

to archaeological sediment starch studies. Outcomes of the

project include an emphasis on understanding taphonomic

transformation of recovered residue and microfossil assemblages,

recognition of the role played by identifi cation in subsequent

interpretation, and the contribution of an alternate theoretical

and methodological framework for interpreting microscopic

residues on stone artefacts.

GOLD FEVER: DISEASE AND ITS CULTURAL RELATIONSHIP: A CASE STUDY ON THE DEVELOPMENT OF THE COLONY OF VICTORIA, 1850 TO 1900

Phil RobertsMA, School of Archaeology and Anthropology, Australian

National University, June 2006

This thesis is a study of the use of disease and disease terminology

as a tool for the reconstruction of cultural environments in

historical archaeological studies. This is achieved through

exploring disease behaviour in the colony of Victoria, experienced

by European, Asian and Aboriginal groups, from 1850 to 1900,

during and in the aftermath of the socio-economic phenomenon

of the Victorian Gold Rush. The data for this thesis derive, for the

most part, from government registers of mortality in the colony.

The aims of this study were explored in the context of several

key questions: (1) how does disease change with the socio-

economic developments of Victoria 1850-1900?; (2) given a

common geography and historical events, how do disease rates

differ between the Asian, European and Aboriginal groups and

what are the potential causes for this?; (3) using this data, is it

possible to differentially diagnose particular diseases through the

examination of terminological usage?; and (4) in answering these

questions, how useful are indicators of disease in archaeological,

anthropological and epidemiological investigations when

attempting to determine the environments in which people lived

and are living?

The key fi ndings were that when the terms ‘quinsy’ and

‘laryngitis’ were used to describe death they most commonly

refer to diphtheria infection. The term ‘puerperal fever’ refers to

group A streptococci bacterial infection of the vagina and uterus

following childbirth. The terms describing liver disease most

commonly refer to alcoholic diseases of the liver.

In terms of disease change with the socio-economic

developments of Victoria from 1850 to 1900, alcohol abuse

behaviour changed from binge-drinking in the 1850s to long-

term alcoholism in the 1870s. Diet changed throughout the

study period and patterns of fertility were refl ected in disease

patterns. Between the minority groups the use of the mortality

term ‘unknown’ differs markedly, potentially showing changes

in racial prejudice with economic success. The different ethnic

groups also had very different rates of accidents, potentially

indicating employment levels and risk of suicide.

It was demonstrated that mortality in Victoria throughout the

period of study changed markedly. Variation was found between

the rates of listed causes of mortality in different socio-economic

environments and in different ethnic groups, demonstrating the

utility of disease in reconstructing past cultural environments.

Thesis Abstracts


Thesis Abstracts

Further, it is shown that different diseases have varying levels of

value in reconstructing cultural change. Finally, it is clear that

different causes of mortality need to be considered on a disease-

by-disease basis in ascertaining how sensitive the behaviour

of the disease is in communicating information about human

behaviour in the past.

THE RISE AND DEMISE OF PATENT MEDICINE ABORTIFACIENTS AND THEIR INFLUENCE ON THE AGENCY OF VICTORIAN WOMEN

Noel SprengerBA (Hons), School of Social Science, University of Queensland,

October 2007

Details of abortifacient patent medicines and a study of the

social history of their use by women to terminate unwanted

pregnancies during the Victorian era are presented. Many

commentators have theorised as to causes for the decrease in

births throughout the Western world from the mid-nineteenth

century and consideration must be given to factors such as the

economy, growing secularism, the rise in feminism and the use

of contraceptive methods. However, little attention has been

given to patented chemical abortifacients as one of these causal

factors and it is contended that the rise of patent medicines in the

‘golden age’ (1865–1907) matched the fall in the birth rate and

was, in part, responsible for this decline. The story that unfolds

is complex. Its primary focus lies in ascertaining the effects of

abortifacients on the agency of Victorian women. Pregnancy

and birth, the realm of female healers and midwives throughout

history had become a subject of public accountability, and the

body of a woman (a permissible topic in polite conversation

only in terms of nurturing) was subjected to medical and state

intervention. While juridical authorities commenced the task

of the management of pregnancy by the introduction of laws

making abortion illegal, irrespective of the considerations of

women, an orchestrated campaign by the male-dominated

medical profession insured that doctors successfully replaced

midwives in pre- and post-natal matters. This disruption to

female networks previously used to disseminate knowledge of

tried and tested chemical abortifacients to terminate unwanted

pregnancies should have resulted in the loss of reproductive

control for Victorian women, but apparently this did not occur.

The manufacturers of patent medicines intervened to provide

a range of abortifacient products of varying toxicity and result.

The demise of patent medicines, including abortifacients, early in

the twentieth century, through the demonstrated power of white,

Anglo-Saxon males in the law-making and medical professions,

evidently resulted in a reduction in women’s reproductive options.

For archaeology, the presence of abortifacient containers, as a

component of assemblages in Victorian-era sites, can provide

signifi cant insights into life (or death) choices of individuals

from that period, as well as a good indication of the dating of

historical archaeological sites.

UNFINISHED BUSINESS: THE LOWER MURRAY LAKES ARCHAEOLOGICAL STUDY WITHIN AN HISTORICAL AND POLITICAL CONTEXT

Kelly Denise WiltshireB. Archaeology (Hons), Department of Archaeology, Flinders

University, October 2006

Once the site of Australia’s fi rst systematic excavation at Devon

Downs, South Australia has seen a dynamic shift in the conduct

of archaeological work in the past 20 years, from a concentration

on archaeological research to a preoccupation with heritage

management. This thesis explores the relationship between

archaeological research and Indigenous heritage legislation by

considering their political contexts and how this has infl uenced

the practice of archaeology in South Australia. The premise of

this thesis is set within the context of a case study, the Lower

Murray Lakes Archaeological Study (LMLAS), conducted by

archaeologist Roger Luebbers during the mid-1980s along the

shores of Lake Alexandrina. This case study is considered within

an historical context to examine how issues and debates within

Australian archaeology infl uenced Luebbers’ research agenda.

Additionally, this case study is used as a specifi c example to

illustrate the outcomes of its political context. Finally, this study

explores contemporary issues faced by the Ngarrindjeri Nation

as a result of the context in which the LMLAS was conducted.

By retrospectively examining Aboriginal heritage protection

in South Australia, with a focus on the Aboriginal and Historic

Relics Preservation Act 1965, the Aboriginal Heritage Act 1979

and the Aboriginal Heritage Act 1988, this research contributes

an in-depth written account exploring the history of Indigenous

heritage legislation in South Australia, while demonstrating how

a lack of government will and poor administration policies have

lead to the insuffi cient management of Indigenous heritage.


OBITUARIESLESTER RICHARD HIATT (1931–2008)

In the fi nal years of his retirement, Les Hiatt returned to Stroud,

Gloucestershire, the county from which his grandfather had

migrated to Australia. It was there that Les Hiatt’s remarkable life

came to an end. Les started out in Gilgandra, western New South

Wales. He attended boarding school in Sydney and graduated

in dentistry at the University of Sydney. Between graduating

and setting up his dentistry practice in Bourke in 1957, Les sat

additional courses in anthropology. He was a keen golfer, an

accomplished boxer and he moved competently around the dance

fl oor of the Bourke Golf Club. It was there he met and married

Betty Meehan. Betty and Les left Bourke for Sydney, where Les

returned to his studies in anthropology. He graduated in 1958

and received a research scholarship at the Australian National

University. It was there he worked with prominent anthropologists

John Barnes and Bill Stanner. For his PhD research, he and Betty

set off to Maningrida, central Arnhem Land, to study Anbarra/

Gidjingarli society. Les gained a teaching position at the University

of Sydney in 1965, and remained there as Reader and Professor

until his retirement. During his years in Sydney, Les and Betty were

members of the Sydney Push, an anarchist, libertarian grouping at

the centre of intellectual life during the late 1950s and 1960s.

Until the early 1990s, archaeology was integrated within

the teaching of anthropology at the University of Sydney.

Archaeology during the 1960s was enlivened by classical

archaeologists, such as Vincent Megaw, Judy Birmingham, and

Rhys Jones, Richard Wright and John Clegg, all actively involved

in Australian research. Jim Allen, Ian Glover and Betty Meehan

were graduate students. In terms of intellectual infl uences, there

were many areas of common interest between the newly-arrived

ecological anthropologist, Nic Peterson, and the Cambridge-

trained archaeologists, infl uenced by the Higgs-Jarman approach

to palaeoecology. This led to what has been termed the ‘Sydney

School’ of ethnographic archaeology. The infl uence of Les Hiatt,

and the other social anthropologists at Sydney, was less direct,

but pervasive, both in terms of critical intellectual debates and

the creation of a libertarian ethos, socially and academically.

Although he himself did not return to Arnhem Land for

any extensive periods, Les Hiatt was instrumental in assisting

other fi eldworkers, such as Annette Hamilton, to work there.

The relationships Les and Betty had formed with individual

Aboriginal people, together with Nic Peterson’s demonstration

that Aboriginal communities in Arnhem Land still followed

many aspects of the hunting and gathering life, led directly to

Rhys Jones’ and Betty Meehan’s decision to spend two years on

the Blyth River at an Anbarra outstation, where they conducted

ethnoarchaeological fi eldwork. These connections continued

through Les Hiatt’s Chairmanship of the Australian Institute

of Aboriginal Studies, working closely with Peter Ucko and

other Principals, a time when the Institute shifted its focus from

research about, to research for and with, Aboriginal communities.

This came to the fore in 1982, when Frank Gurrmanamana

and other community members from central Arnhem Land

brought a Marrajirri ceremony to Canberra, where the Rom

Poles which stand at the Institute remain a symbol of continuing

relationships between Anbarra and the wider community. Frank

and Les were also central characters in Kim McKenzie’s (1980)

fi lm Waiting for Harry (note: not the author). Les Hiatt’s tribute

to Frank Gurrmanamana found its greatest expression in the

translation of interviews recorded in 1960 (Gurrmanamana et

al. 2002) and used by the National Museum of Australia in their

First Australians Gallery.

Les Hiatt was at his best in philosophical and moral

explorations of Australian anthropology, exemplifi ed in

Arguments about Aborigines: Australia and the Evolution of Social

Anthropology (Hiatt 1996). These explorations continued one

of Les’ central interests, the fl uidity and fl exibility of Aboriginal

social and territorial arrangements, a topic which is of direct

relevance to archaeologists. Les could be a diffi cult man, he

was equally impatient with the imperatives of the moral left

and the moral right and he refused the kinds of departmental

administrative roles most of us complain about. He was a star

performer at the annual cricket match between Sydney and

Canberra anthropologists and archaeologists, and also at the

after match functions, where he earned a degree of notoriety.

Consistent with his libertarian principles and in true

Anbarra fashion, Les Hiatt was a polygynist. Also, in line with

Anbarra concepts of generosity (Hiatt 2004), he was generous

in distributing his wives amongst his classifi catory brothers well

before his death. Les’ passing reminds us that our mourning for

Rhys Jones and Peter Ucko continues. Wherever they are, I hope

the wine is up to scratch, I know the conversations will be.

Harry Allen

ReferencesGurrmanamana, F., L. Hiatt, K. McKenzie, B. Ngurrabangurraba, B. Meehan, and R.

Jones 2002 People of the Rivermouth: The Joborr Texts of Frank Gurrmanamana.

Canberra: National Museum of Australia and Aboriginal Studies Press.

Hiatt, L.R. 1996 Arguments about Aborigines: Australia and the Evolution of Social

Anthropology. Cambridge: Cambridge University Press.

Hiatt, L.R. 2004 Edward Westermarck and the origin of moral ideas. In A. Barnard

(ed.), Hunter-Gatherers in History, Archaeology and Anthropology, pp.45-56.

Oxford: Berg.

McKenzie, K. (Director) and L. Hiatt (Narrator) 1980 Waiting for Harry [motion

picture]. Canberra: Australian Institute of Aboriginal Studies.

Figure 1 Frank Gurrmanamana and Les Hiatt in Maningrida in 1975 (Photograph: Margaret Clunies Ross).


BACKFILLMinutes of the 2007 Annual General Meeting of the Australian Archaeological Association Inc.Sydney University, New South Wales

22 September 2007

1. WelcomeThe 2007 AGM commenced at 5:45pm. The President welcomed

everyone to the conference and the AGM with a special welcome

to Professor Mike Parker-Pearson, Professor Stephen Shennan

and Associate Professor Diana di Zerega Wall.

The President commenced proceedings by acknowledging

the Eora (Cadigal) nation as Traditional Owners of the land.

The President thanked the organisers of the New Ground

Conference.

2. ApologiesApologies were received from Tim Denham, Judith Field, Pat

Gaynor, Paul Greenfeld, Ian Lilley, Angela McGowan, John

Mulvaney, John Sherwood, Michael Strong and Sylvia Yates.

3. Confi rmation of the Minutes of the 2006 AAA AGMThe minutes of the 2006 AAA AGM held at Beechworth Victoria

on 9 December 2006 were published in Australian Archaeology

(64:64-73). Motion: ‘That the minutes of the 2006 Annual

General Meeting of the Australian Archaeological Association

Inc. as circulated, be taken as read and confi rmed’. Moved: Stuart

Rapley. Seconded: Val Attenbrow. Motion carried nem. con.

4. Business Arising from Previous AGM

4.1 Northern Territory ArchaeologyNorthern Territory representative Daryl Guse raised concerns at

the 2006 AGM about the treatment of cultural resources in the

Northern Territory and the fact that archaeology was no longer

taught at Charles Darwin University. The Executive contacted

a number of agencies and the university and now has some

support for a future forum on the state of archaeology in the

Northern Territory. The Executive proposed this be initiated

by AAA, although other agencies will no doubt be interested

in being involved. The incoming Executive will be briefed on

this matter.

5. ReportsThe President noted that hard copies of all Executive reports had

been pre-circulated during the conference with only summaries

to be presented at the AGM (full reports reproduced below).

5.1 President’s Report (Alistair Paterson)Welcome to the 2007 Annual General Meeting of the Australian

Archaeological Association Incorporated. I would like to

commence proceedings by acknowledging the Eora (Cadigal)

nation as Traditional Owners of this land. I thank the organisers

of the New Ground Conference and particularly Sydney

University for organising this conference.

I acknowledge the Executive and the respective subcommittee

chairs:

• Samantha Bolton, Chair, Information Technology

Subcommittee

• Richard Fullagar, Chair, Code of Ethics Review

Subcommittee

• Sean Ulm, Chair, Prizes and Awards Subcommittee

• Sean Ulm, Chair, Editorial Subcommittee

• Stephen Free, Chair, Indigenous Subsidies Subcommittee

• Jane Balme, Chair, Australian National Committee for

Archaeology Teaching and Learning (ANCATL)

• Michael Westaway, Chair, National Archaeology Week

Subcommittee

This is my fi nal report to AAA as I am standing down with the

remainder of the Western Australian Executive at this meeting,

having served two terms. I would like to thank my team during

this time, the Executive, the editors of Australian Archaeology,

the subcommittees, the state representatives, and the numerous

members of AAA who assist in the running of the organisation.

I have learned that there are some very dedicated people in the

organisation who are willing to give freely of their valuable time

and expertise. This is a vitally important Association, and as

archaeological resources face many challenges in the years ahead

I urge members to become involved in whatever way they can.

We are run only by the actions of those who volunteer.

When we were elected we had specifi c aims to improve

membership numbers and the fi nancial position, and to improve

some of the internal functions of AAA. Tonight your Membership

Secretary and Treasurer will report that our membership levels

and fi nances are healthy. I have been particularly appreciative

of the signifi cant efforts made to relocate, redesign, rebuild and

improve the AAA website which was relaunched in January. I

urge members to help make the members-only section a resource

for Australian archaeologists. I will briefl y report on some of our

activities since the last AGM.

Northern Territory representative Daryl Guse raised concerns

at the 2006 AGM about the treatment of cultural resources in the

Northern Territory and the fact that archaeology was no longer

taught at Charles Darwin University. We contacted a number

of agencies and the university and now have some support for

a future forum on the state of archaeology in the Northern

Territory. We propose this be initiated by AAA, although other

agencies will no doubt be interested in being involved. We will

brief the incoming Executive on this and other matters in our

handover.

We have committed fi nancial support to the Australian

National Committee for Archaeology Teaching and Learning

(ANCATL) for meetings and promotional material given their


Backfi ll

important initiatives, particularly since securing funding by the

Carrick Institute.

AAA has maintained close contact with the ASHA and AIMA

Presidents during the year, normally in the form of regular

informal phone calls and email messages. We have been able to

work together on several submissions. Joint submissions, where

appropriate, are a direct benefi t to members of the archaeological

community. A joint meeting of all three Executives will be held

during this conference.

In September 2007, working collaboratively with the

World Archaeological Congress, AAA tendered a submission ‘A

past for all Australians: Archaeology and Australia’s National

History Curriculum’ to The Hon. Julie Bishop MP, Minister

for Education, Science and Training. This proposal was written

to encourage the Australian government to develop a new

national curriculum that includes archaeological knowledge

and methods for understanding the past. Claire Smith, Iain

Davidson, Heather Burke, Sean Ulm, Steve Nichols, and others

are to be commended for their efforts in preparing this lengthy

and detailed submission.

In January 2007, a joint submission was made with ASHA and

AIMA regarding ‘The inclusion of archaeology in the Cossack

(WA) Management Plan’ to the Project Steering Committee,

Cossack Management Plan.

In March 2007, a joint submission was made with ASHA and

AIMA regarding ‘Greater protection for the Burrup Peninsula

and Dampier Archipelago Rock Art Sites’ to the Hon. Malcolm

Turnbull MP, Minister for Environment and Water.

In March 2007 a joint submission was made with ASHA and

AIMA regarding ‘Support for consideration of archaeology in the

WA Heritage Review’ to the Hon. Michelle Roberts, Minister for

Heritage, Western Australian Government.

I congratulate past and present editors of Australian

Archaeology as the journal has been awarded the highest quality

rating possible in the initial listing of the European Reference

Index for the Humanities. Other journals that achieved an ‘A’

ranking include World Archaeology, Antiquity and the Journal of

Archaeological Science.

Finally, I will outline plans for future AAA conferences.

The University of Queensland has indicated they will host the

2008 conference; it will be organised by Chris Clarkson, Pat

Faulkner and Andy Fairbairn. The venue will most likely be

in or close to Brisbane. Flinders University has indicated that

they can host the 2009 conference. ANU has indicated they

will host the 2010 conference.

5.2 Secretary’s Report (Fiona Hook)Further to the matters described in the President’s report, the

Executive held fi ve meetings over the year where procedural

matters and the issues already outlined by the President were

discussed. In between meetings members of the Executive

generally communicated by email with 561 emails received by

the Secretary.

CorrespondenceThe Secretary received 75 emails and 16 letters during 2007 from

members and the public. The main themes to emerge from the

correspondence are the same as 2006, including:

• Requests for information from prospective archaeology

students.

• Amateurs/students looking for volunteer excavation

experience.

• International and Australian archaeologists looking for

work.

• Membership and Australian Archaeology queries.

• Scientifi c and alternative archaeology enquiries.

The vast majority of the correspondence relates to the fi rst

three themes; however, with the introduction of the members-

only webpage requests for employment information have

reduced in the last three months.

Telephone CallsThe Secretary received 12 telephone calls from amateur

archaeologists and prospective students.

5.3 Treasurer’s Report (Adam Dias)

OverviewThe AAA fi nancial year ending 31 August 2007 saw a substantial

profi t, largely due to the infl ux of new members and a reduction

in conference costs. The fi nancial year also saw the successful

distribution of a $20,000 grant from the Department of

Environment and Heritage (DEH) to increase Indigenous

participation at the Beechworth conference.

Unfortunately, the AGM fell too close to the end of the

fi nancial year to allow for the accounts to be audited. As such, the

fi gures presented in this report should be considered provisional

pending the fi nal audit (Editors’ Note: The 2006–2007 audit

was completed on 14 November 2007 and the fi gures presented

below found to be accurate).

The year saw a substantial increase in both expenses and

income. Certain costs were carried over from the previous

fi nancial year, which artifi cially infl ated 2006–2007 costs,

including printing and insurance. Additional costs were also

incurred with the formation of the University of Western

Australia account to handle online transactions and to transfer

and update the website. It should also be noted that both the

income and expenses fi gures include the $20,000 DEH grant.

Despite the rise in costs, 2006–2007 saw a profi t largely due to

a substantial rise in real income (Table 1).

IncomeThe breakdown of income for the year is detailed in Table 2. As

with previous years, subscriptions remain the primary source

of income for the Association. A sizable proportion (30.8%)

of the 2006–2007 income was in the form of a DEH grant to

aid in Indigenous participation in the Beechworth conference.

This fi gure infl ates the total, and should not be considered true

income as it was reserved, and exhausted, fulfi lling the terms of

the grant.

The combined income from the savings accounts and two

investment accounts amounted to $2430.60, a rise of $1039. As

with the previous year, the Treasurer was unable to investigate

the potential for higher returns on the investment accounts due

to the demands of fi eldwork, and the investments renewed at the

default rate.


Backfi ll

A small amount of profi ts ($382.60) is a carryover from GST

erroneously removed by the University of Western Australia

account on subscriptions taken in 2005–2006. These funds were

transferred in September 2006, and should be considered part of

the 2005–2006 income.

ExpenditureAs outlined in Table 3, 2006–2007 saw a substantial increase

in expenditure. It should be noted that this initial fi gure is

misleading, with $20,000 coming from DEH funding received

expressly for Indigenous conference participation, and a further

$12,000 in costs carried over from the previous year. Removal

of these amounts reduced expenditure to $27,277. This still saw

a rise in real expenses of approximately $6000 from 2005–2006

largely due to increases in printing costs, bank charges, prizes

and charges related to the website.

Expected Expenditure 2007–2008In the coming 2007–2008 fi nancial year, no new major costs

are anticipated. A small amount of costs ($700) associated

with the 2006–2007 audit were delayed and will appear in the

2007–2008 accounts.

Summary of AssetsAs of 31 August 2007 the AAA assets totalled $77,457.76 (Table

4). This does not include $554.50 in unpresented cheques, or the

contents of the Bruce Veitch Award fund. The memorial fund

grew during 2006–2007 due to interest accrued and a generous

anonymous donation of $500.

The details of the two long-term accounts held by the

Commonwealth Bank are detailed in Table 5. As mentioned above,

due to fi eldwork constraints it was not possible to reinvest the term

deposit at higher rates of return. Hopefully, this situation will be

remedied by the incoming Treasurer. The contents of the long-

term accounts continue to represent approximately two years of

journal printing, which was the intended purpose of the funds.

5.4 Membership Secretary’s Report (Annie Carson)The 2007 membership year began with a fantastic infl ux of

subscriptions to AAA, refl ecting a positive response to the early-

bird renewal notices sent to members in December 2006. This

procedure was implemented for the fi rst time last year and allows

for other targeted mail outs to be arranged in March/April of the

New Year to maximise renewals.

As of 6 September 2007 AAA had a total of 547 members, 86

of whom joined for the fi rst time this year (Tables 1-2). This is a

2005–2006$

2006–2007$

Assets

Current Assets

36,878.87 CBA Cheque Account 35,199.97

23,436.61 CBA Term Deposit 24,303.89

18,118.29 CBA Cash Management Trust 19,109.41

3826.89 UWA EFT Account 10,613.92

82,260.66 Total Current Assets 89,227.19

Liabilities

Current Liabilities

10,483.83 Bruce Veitch Award 11,214.93

0 Unpresented Cheques 554.50

10,483.83 Total Current Liabilities 11,769.43

71,776.83 Net Assets 77,457.76

Equity

46,822.23 Retained Earnings 71,776.83

24,954.60 Operating Profi t 5,680.93

71,776.83 Total Equity 77,457.76

Table 4 Balance Sheet, 31 August 2007.

Statement Date

Particulars Amount Invested

Maturity Paid

30/06/2007 CBA Term Deposit 24,303.89 867.28

30/06/2007 CBA Cash Management Trust

19,109.41 997.12

Table 5 Investment Register.

2005-2006$

2006-2007$

34,625.22 Total Income 64,958.20

9670.62 Total Expenses 59,277.27

24,954.60 Operating Profi t 5680.93

46,822.23 Retained Earnings 71,776.83

71,776.83 Total Equity 77,457.76

Table 1 Profi t and Loss – Summary.

2005–2006$

Income 2005–2006$

630.70 Copyright fees 0.00

8541.13 Conference profi ts 1600.00

1910.00 DVD/Back issue sales 2785.00

1391.50 Interest received 2430.60

19,296.89 Subscriptions 37,632.58

0.00 Other Income 0.00

2855.00 Donations 500.00

0.00 DEH funding 20,000.00

0.00 Bank alterations 10.02

34,625.22 Total Income 64,958.20

Table 2 Profi t and Loss – Income.

2005–2006$

Expenditure 2006–2007$

750.00 Auditor’s fees 627.00

454.49 Bank charges 923.71

1179.81 DVD production 0

3300.00 Journal printing 25,075.34

400.00 Sundries 684.50

0 Insurance 1732.50

0 Internet fees 1018.40

1092.95 Printing, stationaryand postage

6928.56

513.00 Prizes 2250.00

1980.37 Conference 0

0 DEH funding 20,007.26

9670.62 Total Expenditure 59277.27

Table 3 Profi t and Loss – Expenditure.


Backfi ll

fantastic growth in member numbers from last year’s total of 491,

with Student members growing more than any other category. A

breakdown of memberships per state shows continued growth

in memberships from Victoria and Queensland, with steady

numbers in the Northern Territory and Tasmania, despite the

fact that archaeology is no longer taught at universities in those

states (Table 3).

The increase in membership numbers for 2007 can be

attributed to a combined effort from the Executive, Editors and

State Representatives during the past two years. AAA now offers

members more than ever before, much of which has been made

possible through the amazing efforts of Webmaster Samantha

Bolton. The creation of the online subscription and back issue

order facility accessed through the AAA website has signifi cantly

improved the effi ciency of the subscription process for members,

the Membership Secretary and the Treasurer. In addition, the

newly-launched members-only section of the website provides

members with access to PhD, MA and honours thesis abstracts

from all Australian universities, plus job advertisements and

curriculum vitae postings.

In April 2007, the Membership Secretary began a promotional

campaign with newly-designed brochures and posters to

advertise the recent developments in AAA and the added benefi ts

of membership. Over 60 posters and 900 brochures were posted

to 61 universities and government heritage institutions Australia-

wide. Improved membership numbers in 2007 indicate that the

promotion was somewhat successful in raising AAA’s profi le and

encouraging people to join. Thank-you State Representatives for

helping compile mailing lists for each State and Territory. One

suggestion for next year’s Membership Secretary would be to use

the same promotional material to target more institutions within

Australia and particularly overseas.

The other main project for the Membership Secretary in

2007 has been the development of a ‘Procedures Manual’ for

future Membership Secretaries. The manual is designed to

improve the handover process by outlining the requirements of

the position and providing clear instructions on how to process

and manage subscriptions.

I’d like to thank Alistair, Fiona, Adam, Sam and Kelly for

their continued support and help throughout the past two years.

It’s been a great pleasure to work with you all on the Executive.

Thanks must also go to Sean Ulm and Geraldine Mate for

friendly guidance and hard work in getting journals to members.

Lastly, thanks must go to the members of AAA who have shown

much patience and understanding during prolonged periods

of delay with their subscriptions or journals during my time as

Membership Secretary.

5.5 Editors’ Report (Sean Ulm and Annie Ross)During 2007 Numbers 64 and 65 of Australian Archaeology will

be published. AA64 was posted in late May and the anticipated

distribution date for AA65 is mid-November.

This year we continued to focus attention on improving

the quality and standing of the journal. The most signifi cant

development in this area was the appointment of an Editorial

Advisory Board of international standing. Board members are

expected to review up to three manuscripts per 12 month period;

advise on Editorial Policy and Procedures; and encourage

submission of high quality papers to the journal. Up to fi ve Board

members will be changed biennially. Members are invited to

make an expression of interest if they would like to be considered

for future appointment to the Editorial Advisory Board.

The introduction of the Editorial Advisory Board means

that articles and short reports submitted to AA are now

reviewed by at least four people – a senior and junior referee,

a member of the Editorial Advisory Board and at least one of

the Editors. This further formalisation of AA’s review process

ensures that a range of voices are heard in the review process

and that contributors are provided with suffi cient feedback to

improve manuscripts.

Year Members New Members2007 547 86

2006 491 78

2005 482 93

2004 560 110

2003 568 126

2002 497 83

2001 367 67

2000 363 38

Table 1 AAA members and new members, 2000–2007.

Type 2006 2007

Ordinary 309 325

Student 68 96

Overseas 19 19

Retiree 29 30

Institutional 59 68

Life Member 7 8

Non-Member Journal 0 1

Total 491 547

Table 2 AAA membership types, 2006–2007.

State 2005 2006 2007

NSW 155 132 128

QLD 93 92 110

WA 64 57 64

VIC 60 75 92

ACT 50 43 47

SA 38 34 36

NT 14 12 13

TAS 7 4 7

Table 3 AAA membership by state, 2005-2007.


Backfi ll

The quality of AA was recently recognised in its ranking in

the top band of journals in the initial listing of the European

Reference Index for the Humanities. The journal was awarded

an ‘A’ category listing by a panel of experts chaired by Lin Foxhall

(University of Leicester) and jointly sponsored by the European

Science Foundation and the European Commission ERA-NET

‘Humanities in the European Research Area’ (HERA) project.

Other journals that achieved an ‘A’ ranking include World

Archaeology, Antiquity and the Journal of Archaeological Science.

This is a signifi cant achievement for the journal.

The June 2007 issue of AA was the fi rst of three consecutive

issues to be submitted for evaluation to Thomson Scientifi c

(formerly ISI) for possible inclusion in the Web of Knowledge.

If successful, inclusion of AA in the Web of Knowledge will

signifi cantly increase the visibility of the journal and allow

tracking of citation rates and impact factors.

During the year we also continued to investigate a range of

online options for AA. The key issues of concern are affordability

and visibility, combined with a commitment to making at least

some of AA’s content open access. We hope to have online

arrangements in place in the near future.

During the year we promoted AA at various conferences and

workshops and entered into reciprocal promotion arrangements

with Antiquity and The Artefact. These initiatives, along with

the more substantial efforts of the Membership Secretary and

AAA Executive to increase membership levels, will require us

to increase our print run in future. Such increases will require

a modest increase in the budget allocated to the production

of AA.

We take this opportunity to acknowledge the hard work of

the other members of the 2007 Editorial Committee consisting

of Geraldine Mate (Editorial Assistant), Chris Clarkson, Lara

Lamb and Catherine Westcott (Short Reports Editors), Ian Lilley

and Jill Reid (Book Review Editors) and Stephen Nichols (Thesis

Abstract Editor).

There will be several changes in the composition of the

Editorial Committee for 2008. Ian Lilley is standing down as

Book Review Editor. He will be replaced by Chris Clarkson who

will move sideways from his current position as Short Reports

Editor. Editorial Assistant Geraldine Mate, who has been the

glue keeping the AA roadshow together for the past two years,

has reluctantly had to redirect her energies elsewhere. We thank

Geraldine for her hard work, energy and enthusiasm in helping

get the journal settled in its new home. Geraldine will be replaced

as Editorial Assistant in the near future.

In closing, we would like to thank contributors, referees,

the Editorial Advisory Committee and the AAA Executive for

their support.

5.6 Media Liaison Offi cer’s Report (Kelly Fleming)There was limited movement on the media front for 2007.

The following is a list of media releases, and other activities,

for 2007.

Media Releases 200726 April – National Archaeology WeekUWA Media Offi ce released a statement of activities taking place

and the details of the national website which was picked up by a

few WA news sources.

27 August – Archaeology Education in the SpotlightUWA Media Offi ce released a statement about the two-year

project set up by the Australian National Committee for

Archaeology Teaching and Learning (funded by a grant from the

Carrick Institute for Learning and Teaching in Higher Education)

to develop national benchmarks for archaeology degrees. Also

posted on AUSARCH-L, ASHA and AACAI listservers.

Other Activities 200727 April – Call for Media ItemsDue to the completion and launch of the new website and media

statement page a message was posted on AUSARCH-L, ASHA

and AACAI listservers calling for media items to be sent to the

AAA media offi cer.

22 June – Media Links on AAA WebsiteThe potential for creating a media links page on the AAA website

was discussed with the AAA Webmaster. This would allow links

to media items, and other stories of interest to the organisation,

to be added to the site on a regular basis. Work is ongoing.

27 June – AAA Award NominationsDetails and calls for nominations sent to UWA Media Offi ce but

unfortunately it was not picked up. Also posted on AUSARCH-L

and ASHA listservers.

5.7 Webmaster’s Report (Samantha Bolton)The new website, hosted at the University of Western Australia

(UWA) was launched on 6 February 2007. It has the same basic

structure as the previous website, with an updated look and a

few new features.

The new features include the Register of Archaeology Work

Experience Partners, an initiative of the Australian National

Committee for Archaeology Teaching and Learning (ANCATL),

a members-only area, and access for the AAA Executive to view

the membership database and add media releases.

The members-only area was launched on 29 June 2007. The

features include:

• latest news and announcements for members;

• jobs currently available in archaeology;

• list of current cvs of people looking for work in archaeology;

• list of honours and postgraduate theses in archaeology

submitted to Australian universities; and

• volunteer opportunities currently available.

There is also a facility to allow members to update their

membership details, submit a job advertisement, post their cv,

and advertise volunteer positions in archaeology. There are plans

to add a discussion board at a later date.

Usage of this area, in particular the jobs page, has been a little

disappointing. Members are strongly encouraged to advertise

jobs here, and help build this page and make it a major resource

for archaeology in Australia. There will be more active promotion

of this section of the website over the next 12 months.

Other plans for the website are to improve the layout of the

front page, in particular the news section, and continue adding

features, so that the website is current and relevant, and visitors,

particularly members, have reason to keep visiting the site.


Backfi ll

The other major addition to the website is an area allowing

Executive access. This is mainly used by the Membership

Secretary who is able to manage the membership database

through the website and the Media Offi cer who can add media

releases as required.

Viewing statistics apply from 24 May 2007, when Google

Analytics was applied to the website, until 30 August 2007.

During this 3 month period, which included the launch of the

members-only area, there have been 6900 visits, at an average of

33 visits/day. 48% of these viewed the front page only. 68.60% are

fi rst time visitors, and the average time spent on the site is seven

seconds, at an average of fi ve pages per visit. After the front page,

the next four most popular pages are Australian Archaeology,

member pages, study options and the work experience register.

The total hits on the National Archaeology Week website

since December 2006 was 527,000, peaking in May, just before

and during National Archaeology Week, with an average of 4000

hits a day. The new NAW website, also to be hosted at UWA, will

be launched in time for NAW 2008.

The transfer of the website to UWA caused problems for

the email addresses [email protected], as the

university does not have the capability to successfully manage

and forward a number of email aliases. As a result, the email

addresses are now being managed through our domain name

hosts, MelbourneIT. We have increased the number of email

aliases, with all State Representatives having their own addresses,

as well as the Executive. Thus as the people in the positions

change, the email addresses will remain the same.

The launch of the new-look AAA website has been very

successful, with the addition of new features and facilities for

members. I am constantly aiming to make this resource useful

and relevant to members and non-members alike, and am open

to any suggestions as to how to improve the site.

5.8 Australian National Committee for Archaeology Teaching and Learning Report (Jane Balme)As it is clear from recent postings on email groups that some

members are not aware of the origins or previous work of this

subcommittee, I thought that I would start this year’s report by

briefl y outlining its history and previous activities. I will then

turn to our activities over the last year.

ANCATL stands for the Australian National Committee for

Archaeology Teaching and Learning and it is a subcommittee

of the Australian Archaeological Association Inc. Originally

called JISCATL, it formed in 2002 at the Townsville joint

conference with ASHA, AIMA and AACAI. The role of the

committee is to provide a forum for discussion of archaeology

teaching and learning and related issues, and to provide a

conduit for involvement in these issues for other archaeological

associations and interest groups. The main direction for the

activities of this subcommittee was provided by a workshop

held in Redfern, Sydney, in 2003 that was attended by a variety

of archaeologists from all major associations, universities and

industry. At this workshop a charter (known as the Redfern

Charter) of purpose for the committee was created consisting

of five items:

i. Better coordination of opportunities for students and early

career graduates to gain vocational experience.

ii. Accept the principles of the UK benchmarks for teaching and

learning as a step towards the development of a model for

Australian undergraduate archaeology degree structures.

iii. Support for greater interaction and exchange in teaching and

learning between universities on the one hand and regulatory

authorities and industry on the other.

iv. A commitment to gathering reliable data for benchmarking

of a variety of archaeology activities similar to UK survey

instrument.

v. Develop mechanisms to tell the archaeological story to the

publics that own and relate to that archaeology.

Various members of the committee separately and together

have undertaken a variety of activities to fulfi l our charter. One

of the fi rst of these was the study ‘Australian Archaeology in

Profi le: A Survey of Working Archaeologists 2005’. This was

based on 301 responses from Australian archaeologists in a

variety of professions to document the kinds of work done by

archaeologists in Australia as well as the skill areas that these

archaeologists identifi ed were important in their profession

and that they believed were important in training students. The

results of this survey were published in a special teaching and

learning volume (61) of Australian Archaeology and we have

drawn on this survey to identify skill gaps in university teaching

and think about how students might obtain these skills.

In 2006, Wendy Beck, with the help of Catherine Clarke and

ANCATL, applied for and received a Carrick Institute grant to

investigate benchmarking of honours degrees in archaeology. The

purpose of this project is to produce a document that will make

explicit learning criteria, levels of achievement and assessment

practices in archaeology honours programmes. We began work

on the project this year by creating an advisory committee and

surveying university teachers about current practice in Australian

universities. A workshop attended by a group representing all the

relevant universities was held in Melbourne to discuss the results

of the survey and to devise a set of draft benchmarks based on

these results, the United Kingdom benchmark document, the

results of an updated version of the 2005 working archaeologists

survey and the results of discussions at the Redfern workshop.

The benchmark project is not due for completion until the end

of 2008 but will continue to draw on a variety of sources of

information in the production of the fi nal document, including

broad consultation across the discipline.

The other major project this year directed at the fi rst of

the Redfern Charter objectives was the creation of the Register

of Archaeology Work Experience Partners. This is a list of

organisations willing to consider placement of students and

graduates and is available through the AAA website. The list

has been available since March this year and so far we have

been receiving very positive feedback from both employers

and students. We have just begun to advertise this register with

posters and bookmarks. We will shortly be sending posters to

all university archaeology departments and encourage members

to draw students’ attention to the list as an important source of

practical training.

There have been other smaller projects (such as providing

comments on the new K-6 NSW Science and Technology

syllabus) as well as continuing projects (such as Sean Ulm’s

constant update on the profession’s profi le in Australia).


Backfi ll

Most of these activities of ANCATL are organised by a very

few people and anyone that would like to have input in the work

of this subcommittee is most welcome.

5.9 Prizes and Awards Subcommittee Report (Sean Ulm, Val Attenbrow, Fiona Hook and Ken Mulvaney)The Prizes and Awards Subcommittee was formed in 2005 to

manage the awards offered by the Association. In total, there are

10 awards in seven categories, including the Rhys Jones Medal

for Outstanding Contribution to Australian Archaeology, John

Mulvaney Book Award, Bruce Veitch Award for Excellence in

Indigenous Engagement, Life Membership for Outstanding

Contribution to the Australian Archaeological Association Inc.,

and conference paper and poster prizes. The subcommittee also

manages the Laila Haglund Prize for Excellence in Consultancy on

behalf of the Australian Association of Consulting Archaeologists

Inc. In 2007, nominations were received for all award categories.

During 2007 the subcommittee reviewed descriptions and

selection criteria for the assessment of each award and added

to entries of previous awardees on the website. We also liaised

with the New Ground Conference organising committee to

establish paper and poster judging procedures for the joint

conference format.

The subcommittee would like to thank President Alistair

Paterson for active participation in the committee. In 2008, Sean

Ulm will stand down as the chair of the subcommittee and be

replaced by Val Attenbrow.

5.10 Code of Ethics Review Subcommittee (Richard Fullagar)The review process for the Code of Ethics is complete, however,

the role of the subcommittee has changed to encompass the

complaints process of the Association. Thanks to Sean Ulm for

his advice. The President suggests the rewording of subcommittee

title to take into account role of the ethics subcommittee in

assisting the President with Code of Ethics complaints.

Motion: ‘That AAA change name of the Code of Ethics

Review Subcommittee to the Code of Ethics Subcommittee.

Moved: Chair. Motion carried nem. con.

5.11 State Representatives’ Reports

5.11.1 Queensland Report (Lara Lamb)During the past year, there were no issues brought to me as State

Representative by any of our members.

As State Representative I did act in a minor capacity to assist

in the bolstering of our Association numbers, by providing our

Membership Secretary with the contact details of several state and

educational institutions for mail-out purposes. At an individual

level, I distributed several dozen membership pamphlets to my

fi rst year introductory archaeology course, and I believe there

were several resulting applications for membership as a result.

I would like to suggest to the incoming Executive that the role

of State Representative be more clearly defi ned and communicated

to both incoming and incumbent State Representatives.

5.11.2 New South Wales Report (Val Attenbrow)My report is the same as Lara’s for New South Wales.

5.11.3 Northern Territory Report (Daryl Guse)I would like to thank the Executive for their assistance through

the year and, as outlined by President Paterson, the proposed

meeting in the Northern Territory to discuss archaeological issues

with the University, Government and Indigenous organisations.

5.11.4 Western Australia Report (Stuart Rapley)My report is the same as Lara’s for Western Australia.

Motion: ‘That the reports as presented are accepted as read’.

Moved: President. Motion carried nem. con.

6. Appointment of AuditorAddition Bookkeeping Services was appointed as the auditor

for the 2006–2007 fi nancial year. However, given the change of

Executive, the incoming Executive will appoint an auditor for

2007–2008 as per Item 9.4 of the Constitution.

7. Other Business

7.1 Review of NSW Heritage ActVal Attenbrow raised the issue of the proposed review of this Act

and requested that the Executive prepare a submission on behalf

of AAA with assistance from the NSW State Representative. The

President suggested that it should be a joint submission with

ASHA. Marilyn Truscott advises that she is willing to assist in

this submission.

7.2 Council for Australian ArchaeologySean Ulm discussed the proposed establishment of a peak

disciplinary body comprising the Presidents of AAA, ASHA,

AIMA and AACAI. He noted that AAA is always supportive of

joint submissions given their larger impact. Sean proposed two

motions:

1. ‘That AAA support the establishment of a Council for

Australian Archaeology’; and

2. ‘That AAA provide annual contribution to assist in the

running of the Council’.

Peter Veth strongly supported the fi rst motion, arguing that

such a body will increase the lobbying power of the individual

associations and formalise the process of joint submissions that

already exists.

Annie Ross suggested that the Council have the power to coopt

members to assist the Council to allow for specifi c expertise. Ken

Mulvaney suggested that the membership of the Council needs

to be broadened to include the Presidents of AURA, AIATSIS

and the archaeology section of the Australian Academy of the

Humanities to increase the lobby base. Josephine Flood reiterated

the need for the involvement of AURA.

Luke Godwin asked what the basis for the establishment

membership of the Council was, such as the percentage of

membership base, and what mechanisms would be used for each

group’s participation? Sean Ulm indicated that the organisation

and operation of the Council was yet to be discussed. Luke asked

what were ASHA’s and AIMA’s thoughts on the establishment of

the Council? Phil Bowman indicated that AIMA had given its

in-principle support. Luke suggested that we don’t provide our


Backfi ll

support until more details are available. Sean Ulm reminded the

membership that the Council isn’t a separate organisation and is

just a collective body.

The meeting then discussed the processes for the running

of the proposed organisation and whether the President could,

without going back to the members, participate in the Council.

Annie Ross stated that the Executive could call a general meeting

at any time to discuss any issues. However, Ian Johnson pointed

out that the Executive is elected by the membership and trusted

to work in the interest of the Association, therefore, issues do not

need to be brought back to the membership for decision unless

they involve constitutional change.

The President reminded the membership that the executives

of the different groups will meet to formalise and discuss the

process for improving a process that is already in place. The

formation of the Council is to try and set it up to do the best

possible job. Val Attenbrow pointed out the commitments

discussed in the plenary session and hoped that they would also

be raised in the closing plenary session.

Motion: ‘That AAA support the establishment of the Council

for Australian Archaeology’. Moved: Chair. Motion carried nem.

con.

Motion: ‘That AAA allocate an initial amount of $2500 for

the Council of Australian Archaeology, if established’. Moved:

Chair. Motion carried nem. con.

A general discussion ensured as to what the money would be

used for. Suggestions included correspondence, teleconferences

and possible travel for the President to attend meetings.

8. Election of Offi cers of the CommitteeThe following AAA Offi cers were elected for 2008:

ExecutivePresident – Ian McNiven

Secretary – Tim Denham

Treasurer – Bruno David

Public Offi cer – Sally Brockwell

Editors – Sean Ulm & Annie Ross

CommitteeMembership Secretary – Liam Brady

Media Offi cer – Peter Veth

State and Territory RepresentativesACT – Helen Cooke

NSW – Jodie Benton

NT – Daryl Guse

QLD – Lara Lamb

SA – Lynley Wallis

WA – Stuart Rapley

VIC – Nikki Stern

TAS – Denise Gaughwin

9. Close of MeetingThe President thanked members for attending the AGM. The

meeting was closed at 6:32pm.

RHYS JONES MEDAL FOR OUTSTANDING CONTRIBUTION TO AUSTRALIAN ARCHAEOLOGY: JEREMY GREEN

Jeremy Green has been awarded

the Rhys Jones Medal for 2007 in

recognition of his pioneering work

in the development of maritime

archaeology in Australia.

Jeremy was instrumental

in the initiation of the fi eld

of maritime archaeology

in Australia. Following his

pioneering excavations on

the Dutch wreck Batavia in

the early 1970s, he established

the Department of Maritime Archaeology at the Western

Australian Museum and has helped lead the department since

that time. He has been involved in research in over 15 countries,

and has developed training programmes, including UNESCO

Regional Workshops to advance underwater cultural heritage

in countries, such as China and Sri Lanka. Since 1994, Jeremy

has been head of the Australian National Centre for Excellence

in Maritime Archaeology.

Jeremy’s career includes a long span of fi eld and academic

experience around the world. From 1967, he worked at the

Research Laboratory for Archaeology at Oxford University on

the development of an underwater metal detector, a proton

magnetometer, and underwater photogrammetry. These

techniques were investigated in fi eld conditions in England and

the Mediterranean. After joining the Western Australian Museum

in 1971, Jeremy developed programmes involving pre-settlement,

post-settlement and wreck inspection. In 1973, he co-founded

the Maritime Archaeological Association of Western Australia

(MAAWA). He also helped to found the Australasian Institute for

Maritime Archaeology (AIMA), was made foundation President,

and has been Editor of its bulletin since 1977.

Jeremy holds numerous positions and awards, including

membership of the Council for Nautical Archaeology, Honorary

Fellowship of the Centre for Prehistory, University of Western

Australia, Fellowship of the Institute of Archaeology, and Fellowship

of the Australian Academy of the Humanities and Research

Associate of the Institute of Nautical Archaeology. He is Adjunct

Associate Professor at Curtin University of Technology and James

Cook University and Advisory Editor of the International Journal of

Nautical Archaeology. Jeremy is a recipient of the Keith Muckelroy

Prize for Achievement in the Field of Maritime Archaeology,

The Rhys Jones Medal is the highest award offered by the

Australian Archaeological Association Inc. It was established

in honour of Rhys Jones (1941-2001) to mark his enormous

contribution to the development and promotion of archaeology in

Australia. The Medal is presented annually to an individual who

has made an outstanding and sustained contribution to the fi eld.

2007 AAA Conference Awards


Backfi ll

LIFE MEMBERSHIP FOR OUTSTANDING CONTRIBUTION TO THE AUSTRALIAN ARCHAEOLOGICAL ASSOCIATION: COLIN PARDOE

Colin Pardoe has been awarded

Life Membership of the Australian

Archaeological Association Inc. in

recognition of his long-standing

support and contributions to

the success of the Association,

including as President, organising

conferences, presenting papers

at numerous conferences, and

convening the legendary annual

‘Big Man and Small Boy Awards’.

During his time as President

in 2000–2001, along with Louis Warren and Richard Fullagar, he

contributed greatly to raising the level of professionalism and the

profi le of the Association. This was a critical time for the Association,

and for Australian heritage and archaeology more broadly, and

Colin’s leadership played an important role in negotiating valuable

outcomes for the Association and the discipline.

The ‘Creation’ conference he co-convened with the South

Australian Museum, Flinders University and University of South

Australia in 1996 on the Fleureiu Peninsula, South Australia, was

memorable not only for the unexpectedly cold weather but also

because of the large attendance and great atmosphere.

The papers Colin has presented at numerous conferences are

always stimulating and always presented in a manner that makes

human biology understandable and of relevance to an audience

of archaeologists whose aim is to fi nd out how people lived and

behaved in the distant and recent past.

He has also made a great contribution to Australian

archaeology outside the Association. His numerous publications

on the human biology of Indigenous Australians place ancestral

human remains in a broader Australian and world context

that informs on the life of past populations and complements

the fi eldwork and writing of mainstream archaeologists. His

writings have not only been directed to enlightening biological

anthropologists and archaeologists, but importantly to the

Aboriginal communities whose ancestors he studies. His plain

English reports written for Aboriginal communities not only

provide insightful details about the lives of the people whose

remains he has studied, but are also of a high standard and

model for others to follow.

Colin’s many and diverse contributions to the Australian

Archaeological Association Inc. and the broader discipline

of Australian archaeology make him a worthy recipient of

Life Membership.

The Life Membership for Outstanding Contribution to the

Australian Archaeological Association Inc. award was established

to recognise signifi cant and sustained contribution to the objects

and purposes of the Australian Archaeological Association Inc.

BRUCE VEITCH AWARD FOR EXCELLENCE IN INDIGENOUS ENGAGEMENT: BRUNO DAVIDBruno David has been awarded the Bruce Veitch Award for 2007

in recognition of his commitment to working professionally and

innovatively with Indigenous

people in an atmosphere of

friendship, learning and respect.

Bruno has spent an

extraordinary amount of his

working life collaborating with

local Indigenous communities

in Australia, Torres Strait,

Papua New Guinea and the

Pacifi c. These projects have been

intimately based on the interests

of local communities, with Bruno

frequently responding to requests for personal visits and further

research emerging from previous collaborations. Bruno involves

local Indigenous communities in all stages from preliminary

visits prior to fi eldwork, proper employment of fi eld crews

nominated by community leaders, and, importantly, in the

analysis, interpretation, writing and presentation of results.

Desi Grainer (Kuku Djungan Aboriginal Corporation) writes

about the Ngarrabullgan research in Cape York Peninsula, that

‘through the years in which Bruno has worked with our people

he has shown respect and leadership in how he went about his

business amongst the Djungan people and at the same time

helping to support and empower Djungan people to understand,

acquire and record knowledge ... to protect and conserve country

for future generations.’

Bruno’s recent work on a major gas pipeline in Gulf Province,

Papua New Guinea, involved many communities and detailed

consideration of highly signifi cant and spiritual places, requiring

mitigation and archaeological investigation. Douglas Simala

(Community Affairs-PNG Gas Esso Highlands Ltd) speaks highly

of Bruno’s work among the Rumu people of Kikori: ‘Dr Bruno

won the hearts of the clan elders and was given the exclusive right

and freedom of movement to enter Rumu tribal land ... and is

a household name for his input to the cause of preserving the

cultural identity of the Rumu people.’

Bruno has always sought funding for community

representatives to participate in laboratory analyses of their

excavated materials. His logic is that he goes to community lands

in good faith; and it is equally appropriate that he invites people

back to where he works and lives, so that people can experience

and understand that side of the research process. He insists that

funding be made available for this if excavations are done during

a consultancy.

Bruno’s work consistently endeavours to relate the

archaeology of places and objects, of whatever antiquity, to

a living social landscape that is demonstrably meaningful to

the people with whom he works. His ability to work with a

variety of archaeological datasets from stone artefacts, rock

art, and burials to ecology, social relations and spiritual worlds

challenges conventional interpretations of the past and brings

to the forefront Indigenous knowledge and recent local histories.

In much of his published work, local Indigenous people feature

prominently, often as co-authors, in acknowledgement of the

collaborative nature of his research.

The Bruce Veitch Award for Excellence in Indigenous

Engagement was created to celebrate Bruce’s important

contribution to the practice and ethics of archaeology in

Australia. It is awarded annually to an individual or group


Backfi ll

who has undertaken an archaeological or cultural heritage

project which has produced a significant outcome for

Indigenous interests.

JOHN MULVANEY BOOK AWARDThe 2007 John Mulvaney

Book Award is presented to

Mike Morwood and Penny van

Oosterzee’s The Discovery of the

Hobbit: The Scientifi c Breakthrough

that Changed the Face of Human

History (Random House, 2007).

The book provides a valuable

insight into the thoughts and

experiences which lead to the

discovery of a new species of

human, Homo fl oresiensis, at the

site of Liang Bua in Flores, Indonesia. The book exposes Mike’s

contribution and commitment to archaeology and his passion

for discovering ‘the fi rst Australians’. It is written in a style that

is approachable and easy to read yet is still able to convey the

science behind the discovery. The book is also appealing to

academics from many different disciplines and has been able

to transcend academia’s often rigid disciplinary boundaries. It

is rare that a published work in archaeology generates such a

diverse range of interest.

The popularity of The Discovery of the Hobbit has provided a

great service to Australian archaeology by increasing the public’s

awareness of, and interest in, archaeology.

The John Mulvaney Book Award was established in honour of

John Mulvaney and his contribution and commitment to Australian

archaeology over a lifetime of professional service. It was created

to acknowledge the signifi cant contribution of individual or co-

authored publications to Australian archaeology, either as general

knowledge or as specialist publications.

CONFERENCE PAPER PRIZES(Judging Committee: Ballot of registered conference delegates)

Best Overall Paper Prize(Sponsored by AAA, ASHA, AIMA)

SahulTime: A Web-Deliverable Temporal GIS for Archaeological VisualisationsMatthew CollerThis collaborative project between Monash Information

Technology, Archaeology and Geography offers a prototype

for archaeologists to explore the emergence of Pleistocene and

Holocene sites within the wider landscape. SahulTime is an

interactive model of the Australia-Papua New Guinea continent

over the past 100ka years, bringing together bathymetric surveys

with Chappell’s sea-level curve. The imagery and functionality

are similar to Google Earth, with the addition of a slider for

time. Temporal GIS is not new to archaeology, but SahulTime

will be fully web-deliverable and engineered for ease-of-use

and collaboration. Potential applications range from plotting

archaeological datasets in space-time, to creating an educational

tool where reconstructive visualisations are ‘geotagged’, just like

photos in Google Earth.

Best Student Paper Prize(Sponsored by AAA, ASHA, AIMA)

Breaking New Ground: Innovative Approaches to Neanderthal Behavioural ComplexityMichelle C. LangleySince the discovery of Neanderthals 150 years ago, researchers

have considered them to have been largely incapable of

behavioural change and innovation owing to a long-standing

perception that their 220,000 year long archaeological record is,

for all intensive purposes, unchanging. Traditional synchronic

approaches to the study of Neanderthal behaviour have

perpetuated this perception, and the absence of new approaches

has resulted in a stagnation of debates concerning Neanderthal

behavioural complexity. The development and application of

an innovative diachronic approach to this issue provides a new

line of enquiry into Neanderthal behaviour that challenges

traditional models of Neanderthal behavioural complexity.

Diachronic approaches have great potential to extract new

information from the archaeological data currently available

and initial results suggest change in Neanderthal behavioural

complexity never previously demonstrated.

Runner-Up Student Paper Prize(Sponsored by AAA, ASHA, AIMA)

Performing Gentility: Gender, Class and Power on the Colonial Goldfi eldsKate QuirkIn the archaeology of the nineteenth century, gentility is

frequently characterised as an oppressive and coercive social

force demanding of conformity. In this paper I advocate a

different approach, in which gentility is seen not as a means

of social control, but rather as an arena for strategic human

action. Drawing on research in the social archaeology

of Paradise, a colonial Queensland mining settlement, I

demonstrate the strategic value of genteel practice with

reference to case studies of three family residences. Though

of diverse class origins, each case illustrates the ways in which

genteel values and behaviours could be employed in the

pursuit of social and economic advancement. In taking an

inherently social approach to gentility at Paradise, it becomes

possible to move away from traditional techno-environmental

understandings of goldfields life, and instead repopulate the

historical landscape with men, women and children who were

agents of their own destiny.

The Laila Haglund Prize for Excellence in Consultancy(Sponsored by the Australian Association of Consulting

Archaeologists Inc.)

The Concept of Harm and the Application of the Duty of Care Guidelines in Queensland Aboriginal Cultural Heritage LegislationLuke GodwinThis paper will examine the concept of harm as it is applied

in Queensland in the context of applying the Duty of Care

Guidelines. It questions whether the Guidelines meaningfully


Backfi ll

address the conservation and management of cultural heritage

or merely provide a self-assessed and unregulated mechanism

to assist development proponents. The apparent disjunction

between the purpose of the Aboriginal Cultural Heritage Act 2003,

the Queensland Government’s claim of ownership of Aboriginal

cultural heritage (except in limited circumstances) and their

inability to audit the application of their own Guidelines is

explored. An example from central Queensland is presented

to demonstrate how these Guidelines can be applied with little

regard to the archaeological consequences.

CONFERENCE POSTER PRIZES(Judging Committee: Jessica Berry, Sean Ulm, Andrew Wilson)

Best Overall Poster Prize(Sponsored by AAA, ASHA, AIMA)

Injalak Hill Rock Art Recording ProjectSally K. May and Melissa JohnsonThis poster focuses on the Injalak Hill Rock Art Recording

Project in western Arnhem Land and the associated rock art

recording fi eld school. This recording project aims to develop

a baseline record of the rock art from Injalak Hill, to monitor

changes in the rock art for the purposes of conservation, to

develop a detailed chronology for rock art from Injalak Hill,

and to assist in the interpretation of this rock art for a wide

audience. Since 2004 international and Australian students have

visited Kunbarlanja (Oenpelli) as part of a fi eld school to learn

practical archaeological skills and to develop other practical and

personal skills necessary to conduct archaeological research with

Aboriginal communities. In particular, the aim was to provide

a space for students to learn about rock art in its present-day

cultural context. This poster focuses on the successes, the

surprises and the academic outcomes of the rock art recording

project and the fi eld school.

Best Student Poster Prize(Sponsored by AAA, ASHA, AIMA)

The Historic Shipwrecks Amnesty Collections: A Resource for the Study of Human Impact on Shipwreck SitesJennifer RodriguesIn 1993, a nation-wide amnesty was declared in Australia

calling for members of the public, who were in possession of

historic shipwreck relics, to declare their collections without

fear of prosecution. The amnesty was declared in response

to an amendment in the Historic Shipwrecks Act 1976, which

provided automatic blanket protection for all shipwrecks in

Commonwealth waters that were at least 75 years old, whether

or not located. In response to the amnesty, thousands of

shipwreck artefacts were declared by divers, private collectors,

coin dealers, fi shermen, schools, and small regional museums,

which had accepted donated objects from divers over the

years. The amnesty enabled authorities to inventory these

artefacts and enhance information already recorded. The poster

presents the background to the amnesty; the results of a recent

preliminary assessment of the collections, which refl ect the

level of early looting on Australia’s shipwreck sites; and outlines

the benefi ts and implications of such collections. It provides

an illustration of some of the objects found in the collections,

both privately held as well as those held by the State cultural

heritage management agencies and museums, and presents

some basic statistics on the degree of human impact on those

shipwreck sites affected.

Runner-Up Student Poster Prize(Sponsored by AAA, ASHA, AIMA)

Blinded by Bungwall: A Review of Bevel-Edged Artefact Research from Southeast QueenslandAmy Tabrett and Sean UlmBevel-edged artefacts found in southeast Queensland are

conventionally linked with processing of the root of the

bungwall fern (Blechnum indicum). This association is based on

ethnohistoric accounts by early European observers in southeast

Queensland. The validity of the link between bevel-edged

artefacts and B. indicum is central to models of increasing socio-

economic productivity in southeast Queensland. B. indicum

has been argued to be a dietary staple and the appearance of

bevel-edged artefacts in the archaeological record has been

equated with the emergence of B. indicum processing. However,

review of the few use-wear and residue studies conducted on

bevel-edged tools, in conjunction with an examination of the

ethnohistoric literature, reveals that bevel-edged tools were

probably used to process a number of plant foods. This fi nding

suggests that, in the past, archaeologists have given excessive

authority to written sources of evidence about these tools,

resulting in narrowly targeted laboratory studies and leading

to inaccurate models of Aboriginal subsistence behaviours.

Now that use-wear and residue analyses are becoming more

feasible for larger sample sizes, it is recommended that a

thorough investigation of bevel-edged tool use be conducted

in order to create an improved understanding of their use and

signifi cance.

BIG MAN AND SMALL BOY AWARDS(Judging Committee: Cos Coroneos, Steve Free, Jo McDonald,

David Nutley, Cassandra Phillippou, Kathryn Przywolnik, Annie

Ross, Katrina Stankowski, Peter Veth)

When you Wish Upon a Star AwardAnnie Clarke: For saying at the plenary ‘Mike Parker-Pearson

made three good points last night – and two of them were Rachel

Welsh ... I lost interest after that ...’

The Penal Servitude AwardElle Casella: For saying in her paper on proto-convicts that ‘idle

hands lead to unnatural sex crimes’.

The Bluebottle Sword of Honour AwardChris Clarkson: For talking about the Goons Index of Artefact

Reduction (only the lithic jocks noticed this slip of the tongue).

The Jamie Drury Data Makeover AwardMichael Westway: For spicing up his presentation by showing a

picture of a dead tree – and saying ‘this tree is rooted’.


Backfi ll

The Noel Coward Rough Guide to Bali AwardDenis Byrne: For worrying about the de-eroticisation of Gay

Heritage – and saying – ‘what would be the point of that?!’

The Surgeon of Crowthorne AwardCos Coroneos: For wrongly suggesting that the Wingicarribee

fl ows through the Berrima River.

The Back Crack and Sack AwardAndrew Sneddon: For saying in his paper ‘testosterone-driven

moments shouldn’t shape government policy’.

The Elle Macpherson AwardDavid Rose: For answering a question by Peter Veth – by saying

that ‘you should read that article by Peter Veth ...’

The Keira Knightly OscarPeta Knott: For saying (while dressed as a pirate) that ‘Tasmanian

children were very imaginative in their methods of dying’.

The Professor Trelawny Divination AwardJeremy Ash: For showing slides in his presentation which were

date stamped October and November 2007.

The Raymond Dart Drop-Bear AwardMichelle Langley: For stating in her abstract that she had studied

Neanderthal behaviour in the UQ Department in 2006.

The Lonely New Zealander AwardRoss Anderson: For taking a blow-up sheep on fi eldwork with

him – as evidenced in his slides ...

The Malcolm Turnbull Pulp Fiction AwardAngie McGowan: For wrestling with the issue of calculating

heritage potential in Tasmania – and describing this as the

slippery slide ...

The Mr Incredible AwardEoghan Kieran: For having ‘over 100% coverage of mussels on

the river bed’ and for missing his paper by half an hour ...

Shrek Donkey AwardJustin Shiner: For asking (in the paper presented on behalf of

him and Simon Holdaway) ‘Are we there yet?’

Two Holes in One AwardMike McCarthy: For talking about Shipwreck Graveyards and

saying that, ‘only organised folk can turn their own knobs [Ed:

presumably on their tanks] – they must all play golf!’

BIG MAN AWARDMary Casey: Who said at the closure of the Parramatta session

‘Now everyone, before you go, I want you all to come down the

front and look at my lovely jugs’.

PS: The ad hoc committee would also like to extend their

congratulations to Tim Owen who turned 30 this year – and

promptly retired!!!

The Big Man and Small Boy Awards are designed to provide

maximum embarrassment and humiliation at the annual

Australian Archaeological Association conference. All submissions

must have been made in public, preferably during a presentation.

The ad hoc committee accepts nominations from attendees. The

committee does not appear to be bound by any rules, guidelines or

ethics statements – we do what we want.

On Receiving Life MembershipDear Editors,

During the presentation of the Big Man Awards a few years

ago, I let slip that the AAA conference was the biggest part of my

social life. This appeared to shock many people, who dutifully

advised me during the course of the evening that I should go

out more. The eventful life of the Australian Archaeological

Association Inc. has been witness to major changes in the way

we do archaeology, both in the practical sense and in the social. I

am proud to be associated with AAA.

With the general merriment of so many people attending

the annual conference dinner last year (and with bloody Ken

Mulvaney screaming in my ear during the announcement of

the award), I missed what President Alistair said. I was unsure

whether this was an award for a lifetime spent, or for the

unused portion. In either case, it comes at the cusp of a personal

celebration, when I will have spent half my life in Australia.

I want to thank each and every one of you for this award. I

hope that the unused portion will prove as enjoyable and fun as

the portion that I have spent with you all so far.

Unfortunately, I was unable to attend the 2007 conference,

except for the dinner, on account of the drought and the need

for water supply pipelines. Nevertheless, you may rest assured

the ad hoc committee of the Big Man Awards is relentless.

You needn’t think, however, that sucking up to me with this

award will protect any of you from potential humiliation and

embarrassment at future AAA conferences.

This is darned near the greatest honour I’ve had, certainly in

archaeology and biological anthropology.

As always,

Colin PardoeColin Pardoe Bio-Anthropology & Archaeology

Turner, Australian Capital Territory

Letters to the Editors


Backfi ll

AustArch1: A Database of 14C and Luminescence Ages from Archaeological Sites in the Australian Arid ZoneA.N. Williams1, M.A. Smith,1,2 C.S.M Turney3 and M.L Cupper4

IntroductionAustArch1 is a Microsoft® Excel® database listing radiocarbon,

luminescence and uranium series ages from archaeological sites

in the Australian arid zone. It was originally compiled to support

analysis of time-series trends by the authors (Smith et al. 2008;

Williams et al. in press) and is now available online at http://

palaeoworks.anu.edu.au/databases.html. The database is intended

as a resource for archaeologists working in the arid zone. It

provides a ready checklist of dated sites as well as a comprehensive

listing of radiocarbon, luminescence and uranium series age

determinations, and in conjunction with calibration programs

such as OxCal or Calib, can be used to generate radiocarbon

density plots for analysis of trends in occupation. Research in the

arid zone has grown over the past decade, making this central

listing of chronometric data particularly relevant, and we expect

that AustArch1 will become a useful tool for both consultant and

academic archaeologists.

FormatAustArch1 is a Microsoft® Excel® fi le, listing ages (rows) by site,

location and biogeographic region. The latter are based on the

Interim Biogeographic Regionalisation of Australia (IBRA) 6.1

divisions (Thackway and Cresswell 1995). Longitude and latitude

are compiled from published sources, or estimated from locality

maps. All 14C ages are given as conventional (uncalibrated) ages.

The fi le includes a bibliography of published and unpublished

sources for 14C, TL/OSL and U/Th data.

Geographic ScopeAustArch1 covers the entire arid zone, including the following

IBRA regions: BHC, BRT, CAR, CHC, COO, CR, DMR, FIN, FLB

(northern section only), GAS, GAW, GD, GS (Shark Bay area

only), GSD, GVD, HAM, LSD, MAC, MGD, ML, MUR, NUL,

PIL, SSD, STP, TAN, YAL (http://en.wikipedia.org/wiki/Interim_

Biogeographic_Regionalisation_for_Australia). The Murray-

Darling Basin (MDB) is included for comparison.

CoverageAustArch1 is currently the most comprehensive listing of

radiocarbon ages for the Australian arid zone. We estimate that

the database includes >97% of available 14C ages – totalling 971

ages from 286 sites – derived from published and unpublished

research over the past 40 years. In addition to radiocarbon data,

there are 115 TL/OSL or U/Th ages available for arid zone sites (97

dating occupation sediments or hearths). Major gaps in coverage

are the Great Victoria Desert, Nullarbor Plain, Tanami Desert and

the northern (subtropical) margins of the arid zone, refl ecting

the paucity of archaeological work in these areas. We have not

attempted to audit the database for technical and archaeological

validity of the ages, preferring to leave this to users.

Updates and ErrorsWe intend to maintain and periodically update the database and

would appreciate hearing of new 14C, TL/OSL or U/Th results as

they become available. We also ask users to notify us of any errors

(on email: [email protected] or [email protected]).

AcknowledgementsAustArch1 incorporates material from published or private

databases by Matt Cupper, Fiona Hook, Mike Smith, Sean

Ulm and Peter Veth. We thank colleagues who have supplied

unpublished data: Neale Draper, Giles Hamm, Fiona Hook, Roger

Luebbers, June Ross, Kathryn Przywolnik and Esmee Webb.

ReferencesSmith, M.A., A.N. Williams, C.S.M. Turney and M.L. Cupper 2008 Human-

environment interactions in Australian drylands: Exploratory time-series

analysis of archaeological records. The Holocene 18(3):397-409.

Thackway, R. and I.D. Cresswell (eds) 1995 An Interim Biogeographic

Regionalisation for Australia: A Framework for Establishing the National

System of Reserves. Canberra: Australian Nature Conservation Agency.

Williams, A.N., C.M. Santoro, M.A. Smith and C. Latorre in press The impact of ENSO

in the Atacama Desert and Australian arid zone: Exploratory time-series analysis

of archaeological records. Chungara: Revista de Antropología Chilena 40.

1 Fenner School of Environment and Society, Australian National University, Canberra, ACT 0200, Australia

² Centre for Historical Research, National Museum of Australia, GPO Box 1901, Canberra, ACT 2601, Australia

3 School of Geography, Archaeology and Earth Resources, University of Exeter, Exeter, Devon, EX4 4RJ, United Kingdom

4 School of Earth Sciences, University of Melbourne, VIC 3010, Australia

The Australian Academy of the Humanities: 2007 Fellows(Source: http://www.humanities.org.au)

At the Annual General Meeting of the Australian Academy of the

Humanities on 17 November 2007, an outstanding scholar in

archaeology was elected Fellow: Sue O’Connor. Fellows elected

to the Academy are residents of Australia who have achieved the

highest distinction in scholarship in the humanities.

Dr Susan O’Connor is Head of the Department of Archaeology

and Natural History at the Research School of Pacifi c and Asian

Studies at the Australian National University. She is currently

conducting a major archaeological research project in East

Timor investigating the cultural and environmental changes

that took place across the Neolithic transition and into the

Metal Age. This project complements and builds on previous

research in the eastern Maluku region and Wallacea and current

research in Papua New Guinea which is investigating Pleistocene

colonisation in Island Southeast Asia and Greater Australia and

subsequent patterns of migration, interaction and exchange.

Her publications include various papers in Terra Australis,

Archaeology in Oceania and Antiquity.


Backfi ll

Lectures

TERCENTENARY OF THE SOCIETY OF ANTIQUARIES OF LONDON LECTURES/GOLSON LECTURE 2008

Merlin’s Magic Circles: Stonehenge and the Use of the Preseli Bluestones

Professor Tim DarvillSchool of Conservation Sciences and Centre for Archaeology,

Anthropology and Heritage, Bournemouth University

6:30pm, Tuesday 13 May 2008, Sydney Grammar School, College

Street, Sydney

7:30pm, Thursday 15 May 2008, Manning Clark Centre, Union

Court, Australian National University (Golson Lecture 2008)

4pm, Monday 19 May 2008, Age Theatre, Melbourne Museum

Thursday 22 May 2008, University of Auckland, Auckland

Friday 23 May 2008, University of Otago

Stonehenge in central southern England is known the world over

as an iconic symbol of Europe’s prehistoric past. In this lecture

Professor Timothy Darvill of Bournemouth University, will show

that while Stonehenge’s origins as a ceremonial monument were

conventional enough, its later history was exceptional. Key to the

transformation was the arrival of about 80 pillars of Bluestone

rock brought a distance of around 250km from the Preseli Hills

of southwest Wales to Salisbury Plain. But why were these stones

important? And what did they mean to Neolithic people? Using

archaeological evidence from Stonehenge itself and from recent

work in the Preseli Hills, and folklore and oral tradition dating

back to the thirteenth century AD, a new picture of Stonehenge

is emerging in which the stones themselves can be seen to have

perceived magical properties connected with healing. Their reuse

in later and ever more elaborate structures at Stonehenge show

something of their power and signifi cance and illustrate how

the landscape of the Preseli Hills is constructed in microcosm at

Stonehenge. People were attracted to the area from continental

Europe, and what started out as a local focus became a celebrated

place for prehistoric pilgrimage.

On 5 December 1707 three men met at the Bear Tavern (or some

say the Young Devils Tavern) in the Strand in London to discuss

matters related to antiquity and British history; thus was founded

the Society of Antiquaries of London, the second Royal Society in

Britain. By 1717 there were 23 members and the Society has a

continuous history from that date as the oldest learned society in the

world concerned with the study of the material past. As part of the

Tercentenary celebrations there was an exhibition of artefacts from

the Society’s collections, held at the Royal Academy, a book ‘Visions

of Antiquity: The Society of Antiquaries of London 1707-2007’

edited by Susan Pearce, and a series of distinguished lectures that

continue through much of 2008. The Australasian Fellows (currently

28 in Australia and 6 in New Zealand) constitute the largest group

of Fellows of the Society beyond Europe after the United States, and

so Fellow Matthew Spriggs has coordinated the organisation of a

lecture tour in Australia and New Zealand by Professor Tim Darvill

of Bournemouth University as our contribution to the Tercentenary

celebrations.

The biennial Jack Golson Lecture honours Emeritus Professor

Jack Golson, Foundation Professor of Prehistory in the Research

School of Pacifi c Studies. Jack Golson came to the Australian

National University in 1961, following a brief period of outstanding

research in the archaeology of New Zealand. In 1969 he became

the Foundation Professor of Prehistory. His research on the nature

of Lapita culture and on the origins and spread of agriculture,

especially in Papua New Guinea, has brought these matters to

worldwide attention. Amongst various honours, he holds an

honorary doctorate from the University of Papua New Guinea and

is an Offi cer in the Order of Australia.

Conferences

AUSTRALIAN ARCHAEOLOGICAL ASSOCIATION ANNUAL CONFERENCE 2008

3–6 December 2008Noosa, QueenslandAAA2008 will be held over three days at the Australis Noosa

Lakes Resort in Noosaville, Queensland. A welcome evening will

be held on Wednesday 3 December, and a post-conference tour

is planned for Sunday 7 December. Noosa is home to some of

the best beaches in Australia, so bring your togs. Noosa is also

known for excellent dining and its natural beauty, including

the nearby Fraser Island World Heritage Area and Great Sandy

Region. Sessions are invited on current issues and research

in Australian archaeology, as well as sessions of interest and

relevance to Australian archaeologists more generally. This may

include overseas work being undertaken by Australia-based

archaeologists.

Details: http://www.socialscience.uq.edu.au/index.

html?page=74856


1. GeneralAustralian Archaeology, the official publication of the



in all fields of archaeology and other subjects relevant to



(5000–8000 words), Short Reports (1000–3000), Obituaries

(500–2000), Thesis Abstracts (200–500), Book Reviews (500–

2000) and Backfill (which includes letters, conference details,

announcements and other material of interest to members).

Australian Archaeology is published twice a year, in June

and December.

2. Submission of ContributionsSubmissions that do not conform to these Notes to Contributors

may be returned to authors for correction before they are

processed. All contributions must be typed, double line spaced,

using 12 point Times New Roman font or similar. Do not use

more than three heading levels. Do not use footnotes. Do not

use double spaces after full stops at the end of sentences. Number

all pages submitted consecutively. For further guidance on style

refer to the most recent issue of Australian Archaeology. A 150–

200 word abstract must be included for articles. The abstract

should be a complete, concise summary of the paper. A cover

page must be included listing contribution title and full names,

affi liations and postal addresses for correspondence (including

email) of all authors. Do not include author names on pages

other than the cover page.

3. Processing of ContributionsThe Editors will acknowledge receipt of all contributions. The

Editors and external peer reviewers will review all contributions

submitted as articles and short reports. As for any refereed

journal, authors may be asked to make revisions to their

manuscript. If substantial revision is required manuscripts

may be rereviewed before a decision to publish is made. Once

a paper is accepted in its final form, page-proofs will be sent

to the senior author for checking. Proof Approval and Author

Agreement forms will be sent with the page-proofs and must

be completed and returned before publication can proceed.

Final acceptance of manuscripts for publication is at the

discretion of the Editors. For paper submission instructions

see below.

4. CitationsReferences should be cited in text by author’s surname,

publication year and page (e.g. Smith 1988:45). For three or

more authors ‘et al.’ (with italics) should be used after the fi rst

surname (e.g. David et al. 1994). If multiple references are cited

they should be ordered alphabetically and then by publication

year, with authors’ names separated by a semicolon (e.g. Appleby

1990:19-25; Childe 1952; David 1988; David and Chant 1995;

David et al. 1994, 1999; White and O’Connell 1982:42, 50).

5. ReferencesType the References starting on a new page. Include all and

only those references cited in the paper. Do not cite papers in

preparation. Papers may be cited as ‘in press’ where they have

been accepted for publication. For general publication categories

the format should follow the examples below. Please pay

particular attention to capitalisation, punctuation and spacing.

Submissions that do not conform to these referencing guidelines

will be returned to authors for correction.

Journal ArticlesBird, C.F.M. and D. Frankel 1991 Problems in constructing a prehistoric regional

sequence: Holocene south-east Australia. World Archaeology 23(2):179-192.

Book ChaptersCraib, J.L. and G.R. Mangold 1999 Storm in a test pit: Effects of cyclonic storms

on coastal archaeological sites in western Micronesia. In J. Hall and I.J.

McNiven (eds), Australian Coastal Archaeology, pp.299-306. Research

Papers in Archaeology and Natural History 31. Canberra: ANH Publications,

Department of Archaeology and Natural History, Research School of Pacifi c

and Asian Studies, Australian National University.

BooksLourandos, H. 1997 Continent of Hunter-Gatherers: New Perspectives in Australian

Prehistory. Cambridge: Cambridge University Press.

Edited BooksHall, J. and I.J. McNiven (eds) 1999 Australian Coastal Archaeology. Research

Papers in Archaeology and Natural History 31. Canberra: ANH Publications,

Department of Archaeology and Natural History, Research School of Pacifi c

and Asian Studies, Australian National University.

MonographsWickler, S. 2001 The Prehistory of Buka: A Stepping Stone Island in the Northern

Solomons. Terra Australis 16. Canberra: Department of Archaeology and Natural

History and Centre for Archaeological Research, Australian National University.

ThesesDavid, B. 1994 A Space-Time Odyssey: Rock Art and Regionalisation in North

Queensland Prehistory. Unpublished PhD thesis, Department of Anthropology

and Sociology, University of Queensland, Brisbane.

Unpublished ReportsSmith, J.R. and H.J. Hall 1996 Beaudesert Shire Regional Archaeological Project.

Unpublished report to the Australian Heritage Commission, Canberra.

Internet ResourcesAustralian Bureau of Statistics 1996 Education: Participation in Education: The

Education of Indigenous People. Retrieved 6 November 2003 from http://www.

abs.gov.au/ausstats.

Listserver CommunicationsRoss, A. 2004 Useless Australian archaeology graduates. Message posted to the

AUSARCH-L listserver, 31 May 2004, archived at http://mailman.anu.edu.au/

mailman/listinfo/ausarch-l.

NOTES TO CONTRIBUTORS


6. Initial SubmissionIn the fi rst instance, papers should be submitted by email with

the contribution as a single attachment, including text, fi gures and

tables, using Microsoft® WORD (.doc), Rich Text Format (.rtf)

or Adobe® Portable Document Format (.pdf). Specifi cations for

tables and fi gures should follow the guidelines below.

7. Final Paper Submission InstructionsOnce accepted for publication, fi nal versions of papers

(including fi gures and tables) should be submitted as email

attachments or on clearly labelled CDs or DVDs in PC or

Macintosh format. Specifi cations for tables and fi gures should

follow the guidelines below.

8. TextText should be submitted using Microsoft® WORD (.doc) or

Rich Text Format (.rtf).

9. TablesType each table (including a caption) on a separate page at the

end of the manuscript, not in the body of the text. Number and

refer to tables in the text with Arabic numerals (i.e. Table 1 etc).

Tables should be submitted using Microsoft® WORD (.doc) or

Rich Text Format (.rtf).

10. Figures (Line Drawings and Photographs)Line drawings and photographs should both be referred to as

Figures, and numbered consecutively with Arabic numerals

(i.e. Figure 1 etc). Figures should be submitted on separate

pages at the end of the manuscript, not in the body of the text.

Figure captions should be typed on a separate page, not on

the fi gures. For fi nal submission, fi gures should be submitted

electronically as separate fi les as TIFF, JPEG (maximum quality)

or EPS (with preview) fi les. If fi gures cannot be submitted as

computer fi les, please submit as black-and-white line drawings

or as high contrast, glossy black-and-white prints. All fi gures

will be printed as greyscale images. Figures should be submitted

at fi nal size, ready for the printing process. Figures should be

submitted at resolutions of 600 dpi at fi nal size. That is, the fi gure

and its caption should be sized to fi t either within the margins of

a double columned page (170mm) or a single column (81mm).

The available space within margins for the full length of a page is

250mm (this leaves room for a one line caption). Please choose

appropriate letter size, line thickness and shading/stippling.

Photocopies of drawings are not acceptable. If not your own

work, you must acknowledge the origin of all fi gures and, where

applicable, it is the author’s responsibility to obtain written

unfettered permission to publish from the copyright owner of

the original.

11. Supplementary InformationSupplementary information relating to articles may be accepted

for online publication where such materials are directly

relevant to the arguments presented in a paper. Such additional

information may include, for example, detailed data tables where

there is only space for a summary table in the printed version.

At the initial submission stage authors should clearly identify

material as supplementary information. Supplementary material

will normally be restricted to tables and fi gures, but may also

include video or audio fi les. Supplementary tables and fi gures

should be numbered consecutively in the form Table S1, Figure

S1 etc. All supplementary information will be peer-reviewed, at

the same time as the normal refereeing process for the paper.

12. Reporting Radiocarbon Ages and CalibrationConventional radiocarbon ages should be reported as ‘BP’ and

calibrated ages as ‘cal BP’. Report laboratory number, material

dated, calibration method used and any corrections made (e.g.

marine reservoir correction).

13. CopyrightAuthors are responsible for ensuring that any material that

has infl uenced the research or writing has been properly

cited and credited both in the text and in the list of references.

Contributors are responsible for gaining copyright clearance

on fi gures, photographs or lengthy quotes used in their

manuscript that have been published elsewhere. Once accepted

for publication in Australian Archaeology, the author retains

copyright in the work and may publish or authorise others

to publish the entire work or any part thereof, provided that

due acknowledgement is made in any further publication that

the work was originally published in Australian Archaeology.

An article will not be published until the signed Author

Agreement has been completed and returned to the Editors by

the contributor.

14. OtherContributors are encouraged to contact the Editors if they have

any questions or concerns about the submission process.

15. Correspondence and SubmissionsAll correspondence and submissions should be addressed to:


PO Box 6088

St Lucia QLD 4067

AUSTRALIA


Notes to Contributors

AUSTRALIAN ARCHAEOLOGICAL ASSOCIATION INC.Offi ce Bearers for 2008

Position Name Address

Executive

President Ian McNiven School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Secretary Tim Denham School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Treasurer Bruno David School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Membership Secretary Jeremy Ash School of Geography & Environmental Science, Monash University,

Clayton, VIC 3800

Assistant Membership

Secretary

Liam Brady Centre for Australian Indigenous Studies, Monash University, Clayton,

VIC 3800

Public Offi cer Sally Brockwell Centre for Archaeological Research, Research School of Pacifi c & Asian

Studies, Australian National University, Canberra, ACT 0200

Webmaster Samantha Bolton Archaeology, M405, University of Western Australia, 35 Stirling Highway,

Crawley, WA 6009

Media Liaison Offi cer Peter Veth National Centre for Indigenous Studies, Australian National University,

Canberra, ACT 0200

Media Liaison Offi cer Wayne Brennan PO Box 217, Katoomba, NSW 2780

Australian Archaeology Editors

Editor Sean Ulm Aboriginal & Torres Strait Islander Studies Unit, University of Queensland,

Brisbane, QLD 4072

Editor Annie Ross School of Social Science, University of Queensland, Brisbane, QLD 4072

Editorial Assistant Linda Terry School of Social Science, University of Queensland, Brisbane, QLD 4072

Short Reports Editor Catherine Westcott Department of Main Roads, GPO Box 1412, Brisbane, QLD 4001

Short Reports Editor Lara Lamb School of Humanities & Communications, University of Southern


Review Editor Chris Clarkson School of Social Science, University of Queensland, Brisbane, QLD 4072

Review Editor Jill Reid Department of Main Roads, GPO Box 1412, Brisbane, QLD 4001

Thesis Abstract Editor Stephen Nichols School of Social Science, University of Queensland, Brisbane QLD 4072

State Representatives

Australian Capital Territory Kyle Turner 22 Donaldson Street, Braddon, ACT 2620

New South Wales Jodie Benton OzArk Environmental & Heritage Management Pty Ltd, PO Box 2069,

Dubbo, NSW 2830

Northern Territory Daryl Guse School of Archaeology & Anthropology, Australian National University,

Canberra, ACT 0200

Queensland Lara Lamb School of Humanities & Communications, University of Southern


South Australia Lynley Wallis Department of Archaeology, Flinders University, GPO Box 2100, Adelaide,

SA 5001

Tasmania Denise Gaughwin Forest Practices Authority, PO Box 180, Kings Meadows, TAS 7249

Victoria Nicola Stern Department of Archaeology, La Trobe University, Bundoora, VIC 3086

Western Australia Stuart Rapley Archae-aus Pty Ltd, PO Box 177, South Fremantle, WA 6162

Australian Archaeology, the offi cial publication of the



in all fi elds of archaeology and other subjects relevant to



(5000-8000 words), Short Reports (1000-3000), Obituaries

(500-2000), Thesis Abstracts (200-500), Book Reviews

(500-2000) and Backfi ll (which includes letters, conference

details, announcements and other material of interest to

members). Australian Archaeology is published twice a year,

in June and December.

Subscriptions are available to individuals through

membership of the Australian Archaeological Association

Inc. or to organisations through institutional subscription.

Subscription application/renewal forms are available at

http://www.australianarchaeologicalassociation.com.au.

Graphic Design: Lovehate Design

Printing: Screen Offset Printing

Cover: Excavations at Glen Thirsty 1, August 2004, looking

southeast (L to R: Mike Smith and Sharon Overend)

(Photograph: M.A. Smith).

All correspondence and submissions should be addressed to:


PO Box 6088

St Lucia QLD 4067

AUSTRALIA


URL: http://www.australianarchaeologicalassociation.com.au

The views expressed in this journal are not necessarily those of

the Australian Archaeological Association Inc. or the Editors.

© Australian Archaeological Association Inc., 2008

ISSN 0312-2417

Editors

Sean Ulm University of Queensland

Annie Ross University of Queensland

Editorial Advisory Board

Val Attenbrow Australian Museum

Margaret Conkey University of California, Berkeley

Bruno David Monash University

Andrew Fairbairn University of Queensland

Richard Fullagar Scarp Archaeology

Martin Gibbs University of Sydney

Luke Godwin Central Queensland Cultural Heritage

Management Pty Ltd

Simon Holdaway University of Auckland

Susan Lawrence La Trobe University

Jo McDonald JMcD CHM Pty Ltd

& Australian National University

Tim Murray La Trobe University

Jim O’Connell University of Utah

Fiona Petchey University of Waikato

Paul Rainbird University of Wales

Lynette Russell Monash University

Robin Torrence Australian Museum

Peter Veth Australian National University

Editorial Assistant

Linda Terry University of Queensland

Short Report Editors

Lara Lamb University of Southern Queensland

Catherine Westcott Department of Main Roads (Qld)

Book Review Editors

Chris Clarkson University of Queensland

Jill Reid Department of Main Roads (Qld)

Thesis Abstract EditorStephen Nichols University of Queensland

number 66 June 2008

EditorialSean Ulm & Annie Ross iiARTICLESThe Upihoi Find: Wrecked Wooden Bevaia (Lagatoi) Hulls of Epemeavo Village, Gulf Province, Papua New GuineaBruno David, Nick Araho, Alois Kuaso, Ian Moffat & Nigel Tapper 1Tigershark Rockshelter (Baidamau Mudh): Seascape and Settlement Reconfi gurations on the Sacred Islet of Pulu, Western Zenadh Kes (Torres Strait)Ian J. McNiven, Joe Crouch, Marshall Weisler, Noel Kemp, Lucía Clayton Martínez, John Stanisic, Meredith Orr, Liam Brady, Scott Hocknull & Walter Boles 15Dating of Bush Turkey Rockshelter 3 in the Calvert Ranges establishes Early Holocene Occupation of the Little Sandy Desert, Western AustraliaPeter Veth, Jo McDonald & Beth White 33Glen Thirsty: The History and Archaeology of a Desert WellM.A. Smith & J. Ross 45


BOOK REVIEWSRenewing Women’s Business: A Documentary by Julie Drew & Wardaman Aboriginal CorporationReviewed by Sally Babidge 64Coastal Themes: An Archaeology of the Southern Curtis Coast, Queensland by Sean UlmReviewed by Bryce Barker 65The Social Archaeology of Australian Indigenous Societies edited by Bruno David, Bryce Barker & Ian J. McNivenReviewed by Richard Cosgrove 67Neolithic by Susan Foster McCarterReviewed by Phillip C. Edwards 68Archaeology of Ancient Australia by Peter HiscockReviewed by Brian Fagan 69Artifact Classifi cation: A Conceptual and Methodological Approach by Dwight W. ReadReviewed by David Frankel 71Whalers and Free Men: Life on Tasmania’s Colonial Whaling Stations by Susan LawenceReviewed by Martin Gibbs 72Inauthentic Archaeologies: Public Uses and Abuses of the Past by Troy LovataReviewed by Denis Gojak 72Lithics in the Land of the Lightning Brothers: The Archaeology of Wardaman Country, Northern Territory by Chris Clarkson Reviewed by Richard A. Gould 74Box Offi ce Archaeology: Refi ning Hollywood’s Portrayals of the Past edited by Julie M. SchablitskyReviewed by Peter Hiscock 75Salvage Excavation of Six Sites Along Caddies, Seconds Ponds, Smalls and Cattai Creeks in the Rouse Hill Development Area, NSW by Jo McDonald Cultural Heritage Management Pty LtdReviewed by Fiona Hook 76Archaeological Theory and the Politics of Cultural Heritage by Laurajane SmithReviewed by Thomas F. King 77The Archaeologist’s Fieldwork Companion by Barbara Ann KipferReviewed by Ian J. McNiven 78Landscapes, Rock-Art and the Dreaming: An Archaeology of Preunderstanding by Bruno DavidReviewed by Tim Murray 79A Companion to Social Archaeology edited by Lynn Meskell & Robert W. Preucel Reviewed by Alistair Paterson 81

In this issue

nu

mb

er 66 2008

THESIS ABSTRACTS 83

OBITUARIESLester Richard Hiatt (1931–2008) 86

BACKFILLMinutes of the 2007 Annual General Meeting of the Australian Archaeological Association Inc. 87



AustArch1: A Database of 14C and Luminescence Ages from Archaeological Sites in the Australian Arid ZoneA.N. Williams, M.A. Smith, C.S.M Turney and M.L Cupper 99The Australian Academy of the Humanities: 2007 Fellows 99

Lectures 100

Conferences 100


ISSN 0312-2417

thesis abstracts 83 obituaries

Documents