FISHERS, GATHERERS AND HUNTERS ON THE MORETON FRINGE:
Reconsidering the Prehistoric Aboriginal Marine Fishery in Southeast Queensland, Australia
SEAN GEOFFREY FRANCIS ULM
A thesis submitted in partial fulfilment of the requirements for the degree of Bachelor of Arts with Honours in the Department of Anthropology and Sociology, The University of Queensland
October 1995 The archaeological structure of a culture should, and in my opinion does, reflect all other structures, for example, kinship, economic, and political. All are abstracted from the events which occur as part of the normal functioning of a cultural system.
' Lewis Binford (1964:425)
To explain, since everypiece of matter in the Universe is in some way affected by every other piece of matter inthe Un iverse, it is in theory possible to extrapolate the whole of creation - every sun, every planet, their orbits, their composition and their economic and social history - from, say, one small piece of fairy cake.
Douglas Adams (1984:202)
11 '
The work presented in this thesis is, to the best of my knowledge and belief, original except as acknowledged in the text. The material has not been previously submitted, either inwhole or in part, for a degree at this or any other university.
S.G. F. Ulm October, 1995 Brisbane, Australia
Ill TABLE OF CONTENTS
Page
Table of Contents ...... -'...... 1v List of Figures ...... vu List of Tables ...... ix Abstract ...... x
Acknowledgements ...... XI
CH APTER I. INTRODUCING THE PROBLEM
Introduction ...... I
Background to the Study ...... 2
Rationale ...... 6
Aims of the Study ...... 7
Research Design ...... 8
Thesis Organisation ...... , ...... 9
CH APTER 2. REVI EWI NG WALTE RS' MODEL : Context and Place in the Literature
Introduction ...... 11
An Outline ofWalters' Model ...... 11
Theoretical Context and Competing Models ...... 18
Summary ...... 27
CHA PTER 3. THESTUDY REGION : Environmental History, Time-Lags and Marginal Landscapes
Introduction ...... ' ...... 28
The Study Region ...... 29
Environmental History ...... 32
Time-Lags ...... 35
Marginal Landscapes ...... 3 7
Terrestrial Resources ...... 3 8
Shellfish ...... 40
Discussion ...... 42
Summary ...... 43
iv CHAPTER 4. CHRONOLO GY IN SOUTHEASTQU EENSLAND: Exploring the Temporal Structure of the Archaeological Record
Introduction ...... 4 5
Radiocarbon Data ...... 52
Site Definition ...... 54
Sample Structure ...... 54
Methods ...... 57
Middens ...... 60
Rockshelters ...... 62
Cemeteries ...... 63
Discussion ...... 63
Summary ...... 66
CHAPTER 5. FI SHINGIN THE ARCHAEOLO GI CAL RECOR D OF SOUTHEAST QUEENSLA ND: Reconsidering the Evidence
. Introduction ...... '...... 67 Review of Sites 1: Sites Used by W alters ...... 67
St He lena Island ...... 70
Minner Dint ...... 71
Wall en Wallen Creek ...... 72
NRS 7, 8 and IO ...... 73
To ulkerrie ...... 75 Sandstone Point ...... 78 Summary of Wa fters' Sites ...... 80
Review of Sites 2: Other Sites ...... 81
Booral Shell Mound ...... 81
Broadbeach Burial Ground ...... 82
Cameron Point Site 62 ...... 82
Double Island Point Site I ...... 83
First Ridge I9B ...... 83
King's Bore Sandblow 97 ...... '...... 83 Little Sandhills ...... 84
New Brisbane Airp ort ...... 84
Saint-Smith Midden ...... 85 Te ewah Beach Site 26 ...... 85 Tin Can BaySite 75B ...... 86 White Patch Site 3 ...... 86 Non-Coastal Sites ...... 87
Summaryof Other Sites ...... 87
Discussion ...... 87
Summary ...... 89
V CHA PTER 6. DISCUSSION ANDIMP LI CATI ONS: A Review of the Evidence and Related Issues
Introduction ...... 91
A Review of the Evidence ...... 92
Ti me-Lags and Marginality ...... 93
Chronology ...... 94
Intensified Fish Production ...... 96
Relat ed Issues ...... 97
Geomorphology ...... 98
Taphonomy ...... 100
Research Designs ...... 101
Excavation Strategies ...... 103
Recovery Te chniques ...... 105
AnalyticalMethods ...... 106
Summary ...... I08
CHAPTER 7. SUMMARY AND CON CLUS ipNS: Directions and Implications for Future Research
Introduction ...... ll0
Waiters' Model: A Reassessment ...... 110
Directions for Future Research ...... 113
Summary ...... ll6
REFERENCES CITED ...... 118
Vl LIST OF FIGURES
Figure Page
1 The Moreton Bay region showing sites used by Waiters in the model of fishery intensification(Wa iters 1992a:l70) ...... 14
2 Settlement pattern ofMoreton Bay(Wa iters 1987:215) ...... 16
3 Radiocarbon dated sites in theMoreton Bay area since sea-level stabilisation. Where more than one date is available for a site, the basal or nearest to basal date has been used(after Walters 1987:214) ...... 17
4 Recovery of fish remains against time for Moreton Bay archaeological sites (afterWaiters 1989:220) ...... 17
5 The study region, showing major physiographic zones ...... 30
6 Approximate position of the shoreline ofMoreton' Bay at 6,000 BP showing areas of recent progradation and reconstructions of the palaeo- Brisbane and palaeo-Pine River channels based on seismic profiling (afterHekel et a/. 1979: 8; Jones 1992:31) ...... 34
7 Southeast Queensland, showing radiocarbon dated archaeological sites .. .. . 51
8 Distribution of dates by laboratory ...... 55
9 Frequency of major site types and proportion of dates for each major site type ...... 56
10 Frequency of sites with different numbers of radiocarbon dates ...... 57
11 Estimated number of dated sites occupied in each 500-year period ...... 61
12 Number of sites with central radiocarbon dates falling in each 600-year period, measured at 200-year intervals ...... 61
13 Number of new sites established in each 500-year period using earliest dates 61
14 Estimated number of dated midden sites occupied in each 5 00-year period. Series 1 excludes sites first occupied in the last 200 years ...... 62
15 Number of identified specimens (NISP) of fishremains by excavation unit, St Helena Island, Test Pit 1 (afterWaiters 1987:343; cf. Alfredson 1984a:63) 71
16 Number of identified specimens (NISP) of fish remains by excavation unit, Wallen Wallen Creek, M28-B (afterWaiters 1987:246) ...... 73
VII 17 Number of identified specimens (NISP) of fish remains by excavation unit, NRS 7(after Waiters 1987:341) ...... 75
18 Number of identified specimens (NISP) of fish remains by excavation unit, NRS 10 (afterWaiters 1987:342) ...... 75
19 Number of identified specimens (NISP) of fish remains by excavation unit, Toulkerrie-W9A(after Waiters 1987:339) ...... 77
20 Number of identified specimens (NISP) of fish remains by excavation unit, Toulkerrie-W9B (afterWaiters 1987:340) ...... 77
21 Comparison of abundance of fish remains at Toulkerrie where NlSP data available (afterHall and Bowen 1989; Waiters 1987) ...... 78
22 Number of identified specimens (NISP) of fish remains by excavation unit, Sandstone Point 1-A(after Nolan 1986:85) ...... 79
23 Number of identified specimens (NISP) Qf fish remains in analysed squares from SandstonePoint (Nolan 1986:85-88) ...... 80
24 Number of identified specimens (NISP) of fishremains by excavation unit, Booral Shell Mound Square A (afterFrankland 1990:58) ...... 81
25 Estimated distance of dated sites from present shoreline (n=58) ...... 103
26 Abundance of fishremains calculated using both NlSP andMNl methods, Booral Shell Mound Square A (afterFrankland 1990:59, 60). A logarithmic scale is employed so that small MN1 figures are not artificially depressed . . . 107
V111 LIST OF TABLES
Table Page
1 Holocene archaeological sites from southeast Queensland (afterWaiters 198 9:218 ) ...... 12
2 Summary of fish bone abundance data used by Waiters (afterWaiters 198 7:233) ...... 13
3 Radiocarbon dates from archaeological sites in southeast Queensland (n= 140) ...... , ...... 46
4 Summary of midden excavations conducted on the southeast Queensland coast ...... 68
5 Weights and percentages of fauna! categories from Minner Dint (Hall 1980b: 105-106) ...... 72
' 6 Summary of regional fish bone data ...... 88
IX ABSTRACT
In this thesis I present a critical examination ofWalters' (198 7, 1989, 1992a, 1992c) model of late-Holocene intensification of Aboriginal marine fishingin southeast Queensland, Australia.
I demonstrate significantproblems in three premises central to his interpretation of prehistoric cultural change in the region. Firstly, environmental, ethnohistorical and archaeological evidence do not support the proposition that the coastal lowlands were a marginal landscape for human occupation at any time in the Holocene or that a time-lag occurred between sea-level stabilisation and Aboriginal occupation of the coast. Nor is there any palaeoecological evidence to support Waiters' argument that periods of greater sedimentation occurring around
' 3,000 BP caused increases in marine resource productivity. Secondly, even if this enrichment did occur it does not correlate with changes documented in the archaeological record from this time. The occupational chronology demonstrates that significant increases in the number of occupied sites and the rate of site establishment does not occur until around 1,000 BP, some
2,000 years after the proposed enrichment of Moreton Bay. Finally, there is no consistent pattern of increase through time in the quantity of fish remains recovered from archaeological sites in the region. In interpreting this evidence I discuss major taphonomic issues and research biases which have played a significant role in structuring the archaeological database for the region. The Holocene archaeological record of southeast Queensland emerges as much more complex and variable than is generally portrayed.
X
I ACKNOWLEDGEMENTS
This thesis has been a long time coming. Somehow all sorts of things managed to distract me along the way and I think I'm as surprised as everyone else that it finally got finished.
First, I am indebted to my supervisors, Dr Jay Hall and Dr Anne Ross (Department of
Anthrop ology and Sociology, The University of Queensland) for nurturing my love of people and the past with constructive advice, support and encouragement. Most of all I thank them for having more confidence in me than I had in myself Sp ecial thanks to Dr Ian Lilley
(A boriginal and Torres Strait Islander Studies Unit, The University of Queensland) as a friend,
colleague and de facto supervisor. '
I especially acknowledge the assistance of the following individuals for generously making unpublished data from their research in southeast Queensland available for use in this thesis:
Dr Jay Hall (Department of Anthropology and Sociology, The University of Queensland), Dr
Peter Hiscock (Department of Anthropology , Northern Territory University), Dr Ian McNiven
(Department of Classics and Archaeology, The University of Melbourne), Dr Mike Morwood
(Department of Archaeology and Palaeoanthropology, The University of New England),
Robert Neal (Cultural Heritage Section, Queensland Department of Environment and Heritage) and Norma Richardson (M useum and Art Gallery of the Northern Territory ).
For valuable advice, comments and suggestions which contributed to the improvement of this thesis I wish to thank Bryce Barker, Chris Clarkson, Dr Bruno David, Dr Peter Dwyer, Dr
Luke Godwin, Garrick Hitchcock, Dr Harry Lourandos, Paul Mclnnes, Dr Ian McNiven, Dr
Monica Minnegal, Jon Prangnell, John Richter (who also drew the maps), Dr Richard Robins and Mike Rowland. I especially thank Deborah Brian for commenting on many drafts.
XI Various staff of the Department of Anthropology and Sociology, The University of
Queensland, have given invaluable guidance and advice to me over the years and in particular
I would like to thankAnna Bartos, Judy Bieg, Dr Laurence Goldman, Dr David Hyndman, Dr
Leonn Satterthwait and Dr Nancy Williams.
The Honours Course Controllers of the past two years, Dr Harry Lourandos, Dr Anne Ross
and Professor Bruce Rigsby, deserve thanks for their continuing patience and good humour.
The Honours students of the past two years also provided much-needed support and encouragement, especially Deborah Brian, Chris Clarkson, Sebastian Crangle, Zoe Ellerman,
Cindy McRae, Catriona Murray and John Richter.
'
This project would not have been possible without the support and friendship of past and present colleagues and students of the Aboriginal and Torres Strait Islander Studies Unit, The
University of Queensland.
In completing this thesis I have been fortunate in having the support of many friends and family
whose love and encouragement pulled me through. I thank all of you for enduring my ep isodic
mood swings, social absences, general distractedness and much, much more. Finally, I owe
much to Jules for her support, encouragement and understanding throughout the growth of the
ideas presented in this thesis.
xii CHAPTER ONE
INTRODUCING THE PROBLEM
Taken together, [the] evidence does indeed seem to suggest that "something happened" in Australia about 2000 BC. But when each piece of evidence is examined individually, the apparent pattern becomes blurred (White 1994:225).
Introduction
Inthis thesis I present a critical examination of Walters' (I987, 1989, 1992a, 1992c) model of
late-Holocene intensificationof Aboriginal marine fishing in southeast Queensland, Australia.
In particular, the study focuses on an evaluation of three indices employed by Waiters to
demonstrate cultural change in the region toward� more intensive production and productivity:
(i) initial occupation of a marginal environment, (ii) increases in the rate of site establishment,
and (iii) intensified marine resource exploitation. Several investigators (e.g. Amold 1993;
Frankland 1990; Hall 1995;McNiven 1991a) have called attention to potential problems in the
evidential basis and theoretical assumptions underlying the model. These previous studies,
however, have not provided a comprehensive consideration of the model with reference to the
regional archaeological database ( cf Frankland 1990).
I critically examine the primary assumptions underlying Waiters' model through a reassessment
of archaeological, ethnohistorical and environmental evidence. I further test Waiters'
interpretations against an independent database generated through a regional synthesis of the
fish bone data and of the radiocarbon chronology of Aboriginal occupation of southeast
Queensland. Finally, I assess the implications of my results for the reliability of Walters' model,
discuss some of the factors which structure the available evidence, and suggest several areas
for future research.
1 2
Background to the Study
Ethnohistoric accounts and Aboriginal oral histories attest to the prominence of marine fishing in the recent lifeways of Aboriginal people in coastal southeast Queensland (e.g. Campbell
1875; Curtis 1838; Fairholme 1856; Flinders 1814; Gregor 1846; Hall 1984b; Mathew 1910;
Petrie 1904; Russell 1888; Thomson 1967; Uniacke in Steele 1972; Watkins 1891; Welsby
1917; Winterbotham 1957). These sources describe Aboriginal fishing activities in considerable detail and include observations of associated material culture and production strategies.
Archaeological investigations have also reflected this apparent maritime emphasis, yielding diverse assemblages of cultural material which indicate the broad use of littoral and marine ' resources over the last 5,000 years (e.g. A!fredson 1984a; Bowen 1989; Frankland 1990;
Haglund-Calley and Quinnell 1973; Hall 1980a, 1982, 1987, 1990; Hall and Bowen 1989; Hall and Lilley 1987; Lauer 1977, 1979; McNiven 1984, 1990a, 1993; Neal and Stock 1986; Nolan
1986; Robins 1983; Waiters 1987; Waiters et al. 1987).
Aspects of prehistoric Aboriginal marine fishing in southeast Queensland were addressed by
Waiters (1987) inhis doctoral thesis, entitled Another Ke ttle of Fish: ThePrehistoric More ton
Bay Fishery. Focusing on the fish bone component of excavated'assemblages, Waiters conducted an intensive investigation into the operation of the Moreton Bay Aboriginal marine fishery in the context of socially-oriented explanations of Australian Holocene culture change
(e.g. Lourandos 1980a, 1980b, 1983, 1985). Indeed, Waiters (1989:222) viewed his investigations as an attempt "to provide the kind of archaeological evidence sought by
Lourandos (1985:413), concerning rates of Holocene activity and management practices in particular regions of Australia". 3
In his dissertation and subsequent publications, Waiters(198 7, 198 9, 1992a, 1992c) developed an explanatory model linking increased discard of fish remains in the lat e-Holocene to intensnying regional social net works. He argued that there was a time-Jag of some 4,000 years bet ween mid-Holocene sea-level st abilisat ion and human occupation. This time-Jag is at t ributed to the marginalit y of the coast al landscape prior to Iate-Holocene increases in marine product ivit y induced by anthropogenic changes in fluvial sediment at ion. In the absence of significant terrestrialresources, the development of sp ecialised production systems predicated on marine fish resources is argued to have been necessary for successfulpermanent occupation of the coast al lowlands. Furthermore, Waiters argued that the adoption of sp ecialised marine ' fishingre quired a transit ion from mobile hunting and gat hering of terrestrial resources to more sedentary marine resource-based production involving radical restructuring of ex isting social st ructures and ideological systems, accompanied by significant population growth.
This not ion of lat e-Holocene development and intensification of marine fishing is consistent wit h models of mid- toIat e-Holocene socio-economic intensification in Aboriginal societies developed by Lourandos (e. g. 198 0a, 198 0b, 198 3, 198 5, 198 7, 198 8, 1993; Lourandos and
Ross 1994) and others (e. g. David 1994; David and Chant 1995; Flood et al. 198 7; Ross 198 4,
198 5; Williams 198 7, 198 8). These models are based on the ident ificat ion of a broad complex of changes in the Aust ralian archaeological record from the mid-Holocene, including changes in stone artefact assemblages, art styles and fishing technology , the occupat ion of new sites and more intensive use of sites, the occupat ion of marginal areas and the intensive utilisation of new foods such as cycads, cereals and some fish resources (e.g. Beat on 1982; Bawdier 198 1; Flood
198 0, 1995; Hiscock 1994; Lourandos 198 0a, 198 3, 1988; Lourandos and Ross 1994;
Mulvaney 1975; Smit h 1986; White and O'Connell 1982). 4
The evidence from southeast Queensland comprises one of the largest regional archaeological data sets available in Australia, incorporating some 59 dated sites (Ulm et al. in press).
Patterning in the archaeological record of the region, including apparent increases in site establishment and changes in resource use, has often been citedas an example of pancontinental processes involving structural change in production and productivity in Aboriginal societies in the late-Holocene (e. g. David 1994; Lourandos 1993; Lourandos and Ross 1994; Morwood
198 6, 1987; Waiters 198 9, 1992a). These indicators of change have received widespread application in similar regional Australian studies (e. g.Barker 198 9, 1991; Bowdler 1976, 1981,
1995; David 1994; Horsfall 198 7; Lourandos 1980b, 198 5; Luebbers 1978 ; Ross 1981, 198 4, '
198 5; Smith 1993; Veth 1993a; Williams 198 7, 1988).
There is strong evidence that fish resources have been an important component of food production strategies in coastal and riverine contexts since initial occupation of the continent
(e.g. Alienet al. 198 9; Balme 1990, 1995). Recent research has confirmed Pleistocene use of marine resources throughout Australia at sites which have always been in close proximity to the sea (e.g. Morse 198 8, 1993a, 1993b; O'Connor 1990, 1994; O'Connor and Sullivan
1994b; Smith and Sharp 1993; Veth 1993b) . Early marine resource'use, however, is poorly documented in most regions owing to a dearth of coastal sites dating to the late-Pleistocene and
early-Holocene, many of which have presumably been inundated by marine transgression (cf.
Beaton 198 5) . It is therefore difficult to assess the distinctiveness of post-transgressive marine- based Holocene economies. As a result, research into coastal economic systems in Australia has generally focused on the explanation of mid- to late-Holocene change, although the role
of marine resource use even in these relatively recent economies is not fully understood (e.g.
Gaughwin and Fullagar 1995; McDonald 1992). 5
Explanations of mid- to late-Holocene change in coastal resource use in Australia have variously focused on one or a combination of the following approaches:
1. site preservation (e.g. Bird 1992, 1995; Godfrey 1989; Head 1983, 1986, 1987;
O'Connor and Sullivan 1994a; O'Connor and Veth 1993; Rowland 1983, 1989;
Sullivan 1982a);
2. environmental factors, particularly coastal productivity (e.g. Bailey 1983; Beaton 1985;
Morwood 1986, 1987; Nicholson and Cane 1991; Rowland 1983; Waiters 1987);
3. population growth and changes in demographic structure (e.g. Beaton 1985, 1990; Hall
and Hiscock 1988; Hughes and Lampert '1982; Luebbers 1978; Rowland 1980);
4. changes in social structure (e.g. Barker 1989, 1991; Lourandos 1983); and
5. technological arguments (e.g. Beaton 1985; Jones 1984; Sullivan 1982b; Vanderwal
1978).
The model proposed by Waiters is one of several recent syntheses which address issues of late-
Holocene culture change in southeast Queensland (see also Hal11982, 1987, 1990; Hall and
Hiscock 1988; McNiven 1990a, !991a, 1992d; Morwood 1986, 1987). Many of these explanations have, inone way or another, relied on the notion of a time-Jag between sea-level stabilisation and the establishment of stable and abundant marine ecosystems to account for the timing of apparent changes in the archaeological record. The models of Alfredson (1983,
1984a, !984b), Hall (1982, 1987), Hall and Hiscock (1988), Hall and Robins (1984), Waiters
(1989, 1992a) and Morwood (1986, 1987) all rely on an environmental functionalist explanation, positing the late development of tidal flats and their associated mangrove, sea- grass and fauna!communities as a major factor inthe restructuring of Aboriginal settlement and subsistence systems in the late-Holocene. 6
Rationale
I undertook this study in response to four main considerations. First, Waiters' model was selected as a primary focus for critical examination as it provides the most cogent expression of assumptions and interpretations which are common to other approaches to modelling prehistoric culture change in southeast Queensland. A reassessment of the validity of the assumptions and interpretations which underlie Waiters' model therefore provides a basis for evaluating the strengths and weaknesses of other models predicated on similar arguments. As one of few wide-ranging regional explanatory models available (see Hall 198 7; Hall and
Hiscock 198 8; McNiven 1990a; Morwood 198 7), Waiters' view of late-Holocene change has ' become an important referential framework for other studies in the region (e.g. Frankland 1990;
Hall 1995; Hall and Lilley 1987; McNiven 1991a; Waiters et al. 1987). To date, no detailed critique of the model has been undertaken to assess its reliability as an heuristic device in the explanation of culture change in the region.
Second, this examination is critical because the changes in occupation and marine resource use modelled by Waiters are viewed as part of wider processes of alterations in subsistence and settlement strategies in many areas of Australia in the late-Holocene (e.g.Lourandos and Ross
1994). Interpretations of the archaeological record of southeast Queensland have assumed an important role in academic discourse on late-Holocene cultural change in Australia. Both proponents of intensification arguments (e. g. David 1994; Lourandos 1993; Lourandos and
Ross 1994; McDonald 1994) and their opponents (e. g. Bird and Frankel 1991a; Hall and
Hiscock 1988)have cited archaeological evidence from southeast Queensland to support quite different interpretations of prehistoric cultural change. 7
Third, the research presented here is a continuation of the Moreton Region Archaeological
Project (MRAP) - Stage II, which "concerns the delineation and explanation of perceived changes inthe region's archaeological record" (H all and Hiscock 1988:4). It makes an original contribution towards this goal by synthesising and analysing two major regional archaeological data sets, based on fish remains and on radiocarbon determinations.
A final consideration is that recent evidence (F rankland 1990; Dr J. Hall, Department of
Anthropology and Sociology, The University of Queensland, pers. comm., 1995; McNiven et al. n.d.) suggests that marine fishingmay have been a significant subsistence focus in southeast
Queensland by the mid-Holocene, and not after 2,000 BP as Waiters argued. The implications of this evidence for Waiters' and other models of coastal resource use in the region needs to be evaluated.
Aims of the Study
Inorder to address the issues outlined above, I have identified the following aims for this study:
1. To critically assess the model of late-Holocene intensification of Aboriginal marine
fishing posited by Waiters for southeast Queensland, by
a) determining the applicability of Walters' characterisation of the coastal lowlands
as a marginal landscape,
b) testing the model against a synthesis of the radiocarbon chronology of human
occupation, and
c) testing the model against a synthesis of regional fish bone data.
2. To address the implications of my results inrespect of Walters' model in particular, and
to the study of southeast Queensland prehistory in general, based upon an assessment
of the taphonomic and research biases which structure the evidence. 8
Research Design
In order to evaluate the reliability of the W alters model I first reduced it to its constituent
elements. I identifiedthree key components: (i) assumptions about landscape marginality and
time-lags in environmental productivity, (ii) interpretations concerning the timing of change in
the region based on radiocarbon dates and finally, (iii) interpretations about the nature of changes in marine resource production based on analyses of fishremains from coastal sites.
I wanted to reassess the validity of these assumptions and interpretations by examining each
in light of a review of the literature and of the archaeological evidence from southeast
Queensland. '
First, I undertook a thorough search of palaeoecological, geomorphological and ecological
literature to re-examine assumptions of landscape marginality and time-lags in marine resource productivity. My aim was to gain a more detailed understanding of the environmental evidence
as a basis for assessing its ability to support Waiters' arguments. This review also provides an
environmental context for reassessing the timing and nature of late-Holocene change.
Second, I combined the radiocarbon dates available to Waiters with additional dates from published sources, unpublished theses and personal communications to establish a database for investigating regional trends in the chronology of southeast Queensland. In analysing this
chronologyI employed recently developed analytical techniques which are capable of providing
a more sophisticated understanding of temporal change than those used by Waiters and other
investigators in the region. Sp ecific analytical methods employed in the analysis of the
radiocarbon chronology are discussed in Chapter Four. 9
Third, although I had initially intended to synthesise the archaeological evidence for marine fishingby integratingresults from analyses of fishbone assemblages into a single database, on reviewing the evidence itbecame clear that the lack of comparability between excavations and analyses from different sites prevented meaningful quantitative treatment of the data and presented real problems for building a regional understanding of the use of fish resources.
Consequendy, I considered it more appropriate to conduct a site-by-site review of evidence for fishing and to synthesise the regional data on this basis. Some further limitations to our understanding of fish bone representation are discussed in Chapter Six.
Finally, I assess the implications of my results for'Waiters' interpretation and other models of cultural change in southeast Queensland and identifY key areas for future research arising out of this discussion.
Thesis Organisation
In Chapter Two, I present a detailed outline of Waiters' model and identifY its mam assumptions and propositions. The relevant literature is sy stematically reviewed and discussed, situating the model in the broader context of competing frameworks for explicating prehistoric cultural change in the study region.
Through a review of environmental and palaeoenvironmental evidence from the study region
I examine three key components of Waiters' model in Chapter Three: (i) that the coastal lowlands constitute a marginal environment, (ii) that there was a significant time-Jag between sea-level stabilisation and the establishment of productive marine environments, and (iii) that the creation of these ecosystems was induced by anthropogenic modification of environmental productivity. 10
Chapter Four presents an analysis of the radiocarbon chronology of the region. Occupational chronologies based on 140 radiocarbon detenninations from some 59 sites are used to examine patterns of Holocene regional land use. The limitations of this data set are discussed with reference to sample structure and analytical techniques. This synthesis provides the basis for an overview of the chronology of human occupation of the region.
Chapter Five presents a review of the evidence for Aboriginal marine fishing in southeast
Queensland. I demonstrate that much of the evidence cited in support of Waiters' model is equivocal at best, and that the interpretation of the timing and nature of the establishment and development of the marine fishery in the region needs to be revised.
InChapter Six, the scope of the study is broadened to consider the implications of my results for interpretations of late-Holocene change in southeast Queensland Aboriginal societies. In particular, this discussion focuses on the limitations of the archaeological database in light of the operation of taphonomic processes and research biases.
In Chapter Seven I conclude the thesis with a summation of the main findings of my research and, finally, I consider the implications of this research for present and future archaeological investigations in this region and elsewhere. CHAPTER TWO
REVIEWING WALTERS' MODEL: Context and Place in the Literatnre
B y examination of the whole in re ciprocal interaction with it s parts, from religion, powe r and prestige , through marriage arrangements and kinship networks, exchange, settlement, subsistence, burning, cle aring, erosion, sedimentation, mud and sand flat formation, establishment of se a grass meadows, to the distribution and abundance of fish, fishing techniques, production, distribution and consumpt ion, some unde rstanding of the history of a complex syste m can be gleaned (Waiters 1992a:177).
Introduction
Waite rs' model for late-Holocene intensification of Aboriginal marine fishing in southe ast
' Queensland is based on a combination of ethnohistorical, archaeological and ecological data.
This chapter consists of three major components. First is a detailed outline ofWalters' position to expose the logic of his argument and to identifY key assumpt ions and inte rp retations
underpinning the model. Second, I present a brie f review ofthe lite rature in order to situate the model in the broade r conte xt of competing models of prehistoric culture change in
southeast Queensland. Finally, I briefly discuss criticisms which have be en directe d at the
mode l by ot her investigators.
An Outline of Waiters' Model
Waiters (198 7, 1989, 1992a, 1992c) argued that an intensification of Aboriginal marine fishery production took place over the last 1,000 years in southeast Queensland. He noted that very
few sites in the region date to before the mid-Holocene and even then "t here is virtually no
coastal occupation at all prior to 2000 BP" (W aiters 198 7:216). According to Waiters,
exponential growth in the numbe r of coastal sites occurred in the last 2,000 ye ars, with no
major changes in the number of inland sites occupied (T able 1).
11 12
Table I. Holocene archaeological sites from southeast Queensland (after Waiters 1989:218).
Thousands of years BP
Inland I! li 13 12 li 5 4 4 I 2
Coastal 32 10 0 3 5 0 0 0 0 0
Further, Waiters (1989, 1992a) argued that the increase in the establishment of coastal sites coincided with a widespread decline in the rate of stone artefact discard on both coastal and inland sites, and a concomitant increase in the discard of marine fishremains on coastal sites ' and the accumulation of bone, sediment and charcoal in hinterland sites.
Waiters (1987, 1989, 1992a, 1992c) used ecological studies (e.g. Covacevich and Ingram
1975; Dwyer et al. 1979a; Dwyer et al. 1979b; Ingram and Corben 1975; Kikkawa 1975) to argue that the southeast Queensland coastal regionhad a depauperateterrestrial fauna! resource base. He contrasts this with the diversity and abundance of the marine biomass (e.g. Anon.
1981; Hall 1982; Pollock 1980; Thomson 1975), characterising the terrestrial coastal lowlands as a marginal environment. He noted that:
For fisher-gatherer-hunters ... the marine resources of the Holocene Moreton Bay area would have been more worthwhile as a basis for an economy than would be the terrestrial fauna. Fishery resources in particular, would have been of great importance in providing the foundation of subsistence practices (Waiters 1987:71).
Waiters (1992a:168; also 1987:70) held that afterinitial "occupation of the Wallum, Aboriginal hunting would have quickly reduced levels of abundance in the terrestrial animal populations" and he dismissed recent accounts (e.g. Martin 1975; Osawa 1984) of abundance in some taxa as "adequately explained as an artefact of the easing of human predation on these populations 13 afterthe dislocation of Aboriginalhunting" (Waiters 1987:248). Shellfish communities are also dismissed as an insignificant resource on the basis of a study by Catterall and Poiner (1984) which indicated that some species of shellfishmay not be resilient to exploitation. From this,
Waiters (1987:71) concluded that it therefore "seems unlikely that a subsistence economy of a reasonably large population of humans could be founded upon shellfish".
Archaeological evidence for development of the marine fishery is derived from the analysis of eight coastal shell middens in a variety of ecological settings around Moreton Bay: Sandstone
Point on the mainland foreshore of Deception Bay; St Helena Island in southernMoreton Bay;
Toulkerrie and Wallen Wallen Creek on the bay-side of the dune-barrier islands of Moreton and
North Stradbroke Islands; and Minner Dint, NRS 7, NRS 8 and NRS 10 from the exposed easterncoasts of Moreton and North Stradbroke Islands (Figure 1; Table 2).
Table 2. Summary of fishbone abundance data used by Waiters (afterWaiters 1987:233).
SSP StH wwc T MD 7 8 10
NISP 905 576 296 4,073 190 126 61 87
NISP/m2 60,000 2,304 1,184 5,431 95 168 15 364
NISP/m3 46,000 4,608 1,184 9,051 190 840 75 1,820
Key: SSP =Sandstone Point; StH = St Helena Island; WWC = Wallen Wallen Creek; T = Toulkerrie; MD= MinnerDint; 7 =NRS 7; 8 =NRS 8; 10 =NRS 10.
Although mangrove communities and flinging coral reefs appear to have been established soon afterthe Holocene marine transgression (Hekel et al. 1979), Waiters (1989:220) noted no real increase in the number of coastal sites before 2,000 BP. He suggested, following Beaton
(1985), that there was a significant time-Jag between sea-level stabilisation and Aboriginal occupation of the coastal fringe:
! I 14 153"30' LEGEND � Fluvral De/la- quartz lithic sand
r.:., l Prodelta- mud r-=:.1 Minimal Deposition Zone b::;d muddy sands f·(::-1 Tidal Delta- quartzose sand r:::J • Coral Reefs- reefat carbonate LOCATION MAP Wallum Bribie Island D o !"Aoreton Bay Sites ' 27'00
27"30'
South Stradbroke Island
5 10 15 20 25 '
��IOmetres
Figure 1. The Moreton Bay region showing sites used by Waiters in the model of fishery intensification C:VValters 1992a: 170). 15
Archaeological research in the Moreton Bay area has demonstrated a time-Jag of 4,000 years between the middle Holocene high sea level stand and human occupation (Waiters 1987:213).
Significantly, Waiters posited that coastal settlement was coincident with the emergence of specialised fishing economies:
The evidence indicates that fishing, on a scale detectable in the archaeological record, began around 2,000 BP, some 4,000 years afterthe fo rmation of the modem bay (Waiters 1989:220; also 1992c:35).
Waiters (1989:221) argued that permanent occupation of the coastal lowlands was dependent on the exploitation of marinefish resources: "after 2000 BP, the coastal Wallum ceased to be marginal, as a subsistence regime was founded on the marine fishery". '
Waiters (1989:220; also 1987:318-320, 1992a) identified three main indicators of intensifYing marine fishing activities in Moreton Bay:
1. The establishment of new fishing territories by eastward expansion of people to the
islands ofMoreton Bay (Figure 2).
2. An exponential increase through time in the number of sites exhibiting marine fish
remains (Figure 3).
3. Increased intensity of fishremain discard through time (Figure 4).
He (Waiters 1987:3 17) argued that although "environmental changes ... have occurred during the Holocene, they are unable to account fo r the timing of fishery establishment or intensification. Hypotheses which ascribe causal roles for these phenomena to environmental change ... must be rejected". Despite this assertion, in subsequent publications he (Walters
1989, 1992a) explained the apparent time-Jag in human occupation after sea-level stabilisation by relating the high ecological productivity ofMoreton Bay to anthropogenic changes in marine 16
productivity induced by Aboriginal firing regimes in the hinterland and subsequent episodic
erosion and increased fluvial sediment discharge(following Hughes and Sullivan 1981; see also
Hall 1990):
it was Aboriginal burning, resulting from an intensified drive for increased social production, which led to greater erosion, which in turn led to sedimentation, which in its turn led to the establishment of rich inshore fishing grounds (Waiters 1992a:176).
0 Deception Bay 2 Little Sandhills Tou_lkerrie Deception Bay 1 NRS 7 Bribie 3 NRSB Minner Dint SSP-1 NRS10 1 First Ridge WWC One Tree ' (upper St.Helena midden) 2 Brown's Road
3
4 0. "' -;;- >. � Platypus 5 Rockshelter • • ·- . ... Bushranger Cave 6 Moreton Bay forms
10 Brisbane River valley occupied (WWC)
100
bay fills and empties alternately - present vegetation pattern established 1000
Transect across bay area (1\'•E)
Figure 2. Settlement pattern ofMoreton Bay (Waiters 1987:215). 17 20
15
00 2 ·;;; � 0 � " .Cl E 10 " z
5
0
0 2 4 6
KYrs BP '
Figure 3. Radiocarbon dated sites in the Moreton Bay area since sea-level stabilisation. Where more than one date is available for a site, the basal or nearest to basal date has been used (after Waiters 1987:214).
50
40 0 0 0 c. 00 c " E 30 ·o " a. 00 .c Jg "0 " "' 20
:g" 31 0 � .Cl E " 10 z
0 .
0 1000 2000 3000
Age (years B.P.) Figure 4. Recovery of fish remains against time for Moreton Bay archaeological sites (after Waiters 1989:220). 18
W alters suggested that these environmental changes provided a background against which
societies underwentorganisational changes:
This "reorganisation of people through space" took place as certain bands moved out of the river valleys of the inland and took up residence in the formerly marginal environment of the coast. Political and social life intensified as people established networks between old countries and new ... Such regional demographic adjustment would imply that social, political and religious networks would have become of greater significance through time, and that social and political life in southeast Queensland intensifiedand became more complex through the middle and late Holocene (Waiters 1992a: 175).
Insum, the modelproposed by Waiters consists of three major elements: (i) the notion of the marginality of the coastal lowlands and a time-Jag in human occupation, (ii) an exponential
' increase in the number of sites through time and, (iii) an exponential increase in the discard of fishremains through time.
Theoretical Context and Competing Models
Waiters' model is best understood in the broader context of archaeological investigation in southeast Queensland, set against a background of major theoretical developments in
Australian archaeology in the late 1970s and early 1980s. Waiters (1989:222) identified his own explanatory position with that of Lourandos, whose seminal works on the interpretation of Australian prehistory employed a socially-oriented approach derived from neo-Marxist anthropological theoryto explain transformations documented in the archaeological record of the mid- to late-Holocene (Lourandos 1980b, 1983, 1985, 1988, 1993; Lourandos and Ross
1994; see also Bender 1978, 1981).
In reaction to the static models emphasising environmental factors which once dominated interpretations of Australian prehistory (e.g. Birdsell 1977; Jones 1977), Lourandos posited social dynamics as primary agents of cultural change in hunter-gatherer societies. He argued 19 that processes of intensification were responsible for socioeconomic modifications wh ich resulted in increases in production and productivity in Aboriginal societies in the late-Holocene.
Evidence for the operation of these processes may include increased intensity of site use, increased rates of establishment of new sites, greater use of marginal environments, changes in resource use and development of complex alliance networks (Lourandos 1985:391).
Over the last 10 years, general syntheses of regional archaeological patterns in southeast
Queensland have emerged which emphasise significant increases in site numbers and use after about 2,000 BP, especially in the coastal region (Hall 1987; Hall and Hiscock 1988; McNiven ' 1990a, 1991a; Morwood 1986, 1987; Waiters 1987, 1989). These syntheses are based largely on the results of numerous archaeological studies conducted under the auspices of the Moreton
Region Archaeological Project (MRAP) (e.g. Alfredson 1983; Bowen 1989; Crooks 1982;
Draper 1978; Hall 1980a, 1982, 1984a, 1986, 1987; Heather 1983; Lilley 1978, 1984;
McNiven 1990a; Nolan 1986; Richardson 1979; Robins 1983; Waiters 1979, 1987). This large regional data set has provided the basis for contesting various archaeological interpretative frameworks.
Interpretations of regional cultural change involving changes in coastal occupation in southeast
Queensland have essentially polarised around two views. In a similar argument to Waiters',
Hall (e.g. 1982, 1987, 1990, in press; see also Hall and Bowen 1989; Hall and Hiscock 1988;
Hall and Lilley 1987; Hall and Robins 1984) has related more intensive use of marine resources in the late-Holocene to changes in marine productivity caused by mud flat progradation which he also linked to the operation of Aboriginal firingregimes. Where Hall's interpretation departs dramatically from that offered by Waiters is in the nature of the inferred antecedents of the complex social sy stems of the recent prehistoric past. Hall maintains that people were always 20 resident on the coast and simply followed the transgressing coastline, subsequently annexing new or previously little used inland areas. Hall argued that after sea-level stabilisation, people began to more intensively utilise the rich estuarine communities which developed as a response to mid-Holocene mud flat progradation, with the emergence of a marine resource-based subsistence economy. Hall also draws a general contrast between the abundance of marine resources and the paucity of terrestrial resources but confines the marginality argument to offshore island contexts (see Hall and Robins 1984).
The explanatoryfram ework adopted by McNiven for the northern coast of the study region is closely aligned with that posited by Hall. McNiven argued that the evidence for first occupation of the Cooloola region at about 5,500 BP is associated with sea-level stabilisation and suggested that this represents "a localized adaptation of an extant marine-terrestrial settlement-subsistence system which had slowly been moving westwards across much of the present continental shelf with the transgressing coastline" (McNiven 1990a:376). McNiven
(1991a:15) invoked a similar marginality argument to that adopted by Waiters arguing that:
Following the general patternfor coastal southeast Queensland, the potential terrestrial animal resource base of eastern Cooloola is poor, exhibiting low abundance and low diversity ... As Dwyer, Hocking and Willmer (1979:81) note, 'the nutritional rewards' for hunter-gatherers exploiting Cooloola's mammal resource base, particularly macropods, 'would have been scant'. In contrast, marine habitats flanking easternCooloola exhibit a rich suite of edible shellfish and fishresources.
Morwood provides the main alternative positionto that proposed by Hall and by McNiven.
Morwood ( 1986, 1987) explained Holocene cultural changes in terms of increasing social complexity which developed as a response to the post-transgressive configuration of key regional resources (particularly marine fish and bunya nuts). Morwood (1987) contended that before 6,000 BP populations were highly mobile and of low density, reflecting the generally
I 21 low productivity of regional environments prior to sea-level stabilisation. Although Morwood
(1987:343) conceded that "there may have been human exploitation of coastal resources" before the mid-Holocene, he argued that "conditions during times of fluctuating sea levels do not appear to have been conducive to high, littoral productivity". After 6,000 BP, however, the fo rmation of Moreton Bay brought about a dramatic increase in marine resource productivity which "provided scope fo r an accelerating increase in population in the region, especially after4000 BP" (Morwood 1987:343).
On the basis of occupational sequences from 25 sites, Morwood developed a regional
' chronology and suggested that:
The growth of site numbers over time fr om 4500 bp (5000 BP) appears to be of logarithmic rather than linear type, and provides a general measure of population increase ... This increase occurs in all major environmental zones and post-dates the stabilization of sea level by 1500 years. Certainly, the evidence does not suggest that a sizable population of coastal dwellers was 'pushed back' to the present coastline by the last marine transgression (Morwood 1987:343).
Although Morwood (1987:346) noted evidence fo r marine resource exploitation around the mid-Holocene, he maintained that "intensive and specialized use of marine fo ods did not occur until much later". In particular, he emphasised, like Waiters, that the apparent elaboration of social complexity in the region in the late-Holocene was related to the development of the marine fishery (see Waiters 1992b:29).
These two positions - Hall and McNiven on the one hand and Morwood and Waiters on the other - reflectthe polarisation of general explanations fo r the development of coastal resource use: (i) that people always lived on the coast but that the evidence has been destroyed by transgressive marine events, or (ii) that marine resources were only recently incorporated into human economic systems because of their low ranking relative to terrestrial resources (Osborn 22
1977, 1980), or because of changes in the availability of marine resources (Beaton 1985).
Waiters' argument fa lls into this finalcategory, or what Moseley and Feldman (1988:126) refer to as the 'calorie-free sea' hypothesis.
Walters uncritically accepted Beaton' s (1985) proposition that Aboriginal occupation of the
Australiancoastline occurred after sea-level stabilisation. In fact, one of the aims of his study was "to determine the coastal occupation time-Jag fo r the region" (Waiters 1987: 173). The notion of a time-Jag between sea-level stabilisation and coastal occupation, related to the slow development of littoral resources, has been persistent in the explanation of changes in coastal resource use in Australian prehistory (e.g. Callaghah 1980a, 1980b; Lampert and Hughes 1974;
McCarthy 1943a, 1943b). This time-Jag model inherently assumes that occupation was dependent upon the existence of large (and specifically estuarine) shellfishcommunities (see
Beaton 1985). Assumptions about the affects of marine transgression on marine fauna and florahave been challenged by recent research which demonstrates the resilience and durability of marine ecosystems over long periods of time and when subject to inundation (e.g. Hutchings and Saenger 1987; Quinn and Beumer 1984; Woodroffe et al. 1988). Even if some species of marine fa una and flora were adversely affected by changing sea-levels,'the littoral zone must at all times in the past have supported a broad range of species. Barker (1989, 1991), for example, has emphasised the persistence of motile marine species (such as fish, dugongs, and turtles) throughout periods of dramatic environmental change. The effects of marine transgression on sessile fauna such as shellfish are as yet poorly understood (see Jablonski
1980). Furthermore, many recent archaeological studies have overthrown the general applicability of the time-Jag model, demonstrating occupation of the coast and use of marine resources throughout the marine transgression (e.g. Barker 1991; Morse 1993a). 23
Even before Waiters' doctoral thesis was accepted in 1987 the existence of an occupational time-Jag in Moreton Bay was called into question. Nolan (1986) considered that the chronology of Wallen Wallen Creek (Neal and Stock 1986), demonstrating continuous occupation from the late Pleistocene, effectively rejected W alters' claim "that the coastal island
Wallum was not occupied during the period 6,000 - 2,000 B.P." (Nolan 1986:5). Also, both
Hall and Lilley's (1987) investigations at the New Brisbane Airport site and Waiters' (Waiters et al. 1987) own investigations at the Hope Island site demonstrated coastal occupation in the region by around the mid-Holocene. Waiters accommodated the Hope Island evidence into his
model by arguing that the "site was not a place of' fishdiscard" and the absence of fish remains could be explained if it was a "women's site" (Waiters et al. 1987:92).
Although various explanatory frameworks have been proposed to account for patterns in the regional archaeological record, curiously there has been little formal exchange of views in the literature. The only exception to this is the response to Morwood's (1987) model for increasing social complexity which was heavily criticised by Hall and Hiscock (1988), McNiven
(1990a) and Waiters (1992b). The main comments about Waiters' model are primarily concerned with the evidential base rather than with theoretical issues.
For example, McNiven (1991a:21) argued that the model proposed by Waiters underestimated the "role of fishing in the mid-Holocene due to an analytical bias towards fish remains recovered from late Holocene shell middens". He suggested that fish remains would only have a high survival potential ifincorporated into a shell matrix and concluded that "as most coastal shell middens in southeast Queensland date to the last 1000 years ... most preserved fish bones will similarly date to this period and few valid inferences can be made concerning fishing conditions prior to 1000 BP" (McNiven 1991a:21). 24
Waiters (1992c:36-37) objected to this criticism, countering that the prominence of late-
Holocene shell middens in the sample "is not really so much a bias, as it is a reflection of the data available". Walters argued that evidence from the four coastal sites which spanned the mid- to late-Holocene period (Hope Island,New Brisbane Airport, Teewah Beach Site 26 and
Wallen Wallen Creek) demonstrated that a developed fishery is not evident until late-Holocene times.
Hall (1990: 178) proposed two alternative explanations for the apparent pattern of fishbone representation insoutheast Queensland sites: (i) "either abundant fishremains in midden sites dating to this early Holocene Period are there but have not yet been found" or (ii) "for some reason, fishwere not readily available inthese kind of quantities to Aboriginal populations until this time".
This same dilemma has been fa ced by several researchers in Australian coastal studies. For example, Lampert (1981) adopted the firstexplanation for the near-absence of archaeological evidence for marine resource use on the South Australian coast west of Eyre Peninsula in comparison with its abundance in southeastern Australia:
Of the two significant marine resources exploited by Aborigines, molluscs and vertebrate fish, the former give rise to large quantities of durable and easily visible waste (Parmalee andKlippel 1974), the latter to fragile waste that is low in bulk and unlikely to endure except in well-protected sites . . . Thus, particularly where open sites are concerned, a change in emphasis from fishing to shellfish gathering gives rise to an entirely different order of archaeological visibility (Lampert 1981:181).
Nicholson and Cane (1991:10), in summarising Lampert's (1981) argument, have also noted that "the orientation of coastal groups towards fishingmay be virtually invisible as fish remains do not preserve well unless incorporated into a deposit". Dortch et al. (1984) document a similar situation in southwest Western Australia where ethnohistoric literature suggested 25
significant use of marine fish resources and only incidental use of shellfish although there is
virtually no archaeological evidence for coastal occupation in open contexts.
Also in criticism of Waiters' position, Hall (in press) suggested that the model may work but that it is not supported by the available evidence. Hall argued that the entire model is
essentially constructed on the evidence derived from a single site, Sandstone Point, which
cannot be considered representative of southeast Queensland coastal sites. Hall and Hiscock
(1988: 12) also regarded Sandstone Point as anomalous and cautioned its use as a basis for
generalisations of regional cultural change. They do, however, support Waiters' (1987)
' argument that relatively more intensive fishing is evident in both the Sandstone Point and
Toulkerrie sites in the Iate-Holocene (Hall and Hiscock 1988:15). They suggested that the
presence of bevelled pounders in sites afterabout 1,500 BP may similarly indicate changes in
plant food production towards specialisation at this time, but that "there is no evidence that
convinces us that dramatic shiftsin social complexity accompanied such specialisation" (Hall
and Hiscock 1988:15). Itcould be argued that major changes in resource production could be
expected to be associated with significantchanges in social organisation. Whether or not this
altered social formation was any more complex than the previous one is difficult to determine.
To date, Frankland's (1990) study inthe Great Sandy Strait region is the only detailed research
which has specificallyaimed to examine and test Waiters' model against an independent body
of evidence. Although her (Frankland I 990) analysis of the Booral Shell Mound fauna!
assemblage demonstrated major fish discard before 3,000 BP, she only discussed the
significance of these results for Waiters' model in a very general way, suggesting a revision of
the timing of the changes posited by Waiters, from 2,000 BP to at least 3,000 BP.
I 26
Few comments have been made in the literature in relation to Waiters' theoretical position which could be characterised as eclectic at best. Hitchcock (1994:36) cited Waiters' (1992a) model as an example of an approach which successfully incorporated a neo-Marxist dialectical account of cultural change "in which the evolution of dynamic human societies is argued to result from the resolution of contradictions within and between social systems and their environments". I suggest that identification of such contradictions requires not only detailed modelling of social responses to environmental change, but consideration of the scale and impact of these changes.
' Arnold (1993) accused Waiters of misusing a Boserupian concept of intensification:
Boserup's (1965) concept of intensification... has been misused by those who argue that phenomena such as simple increases in amounts or types of foods exploited or area of land cultivated represent legitimate labor reorganization and elite involvement (e.g., for Australia, Waiters 1989 and Williams 1987). It is essential to distinguish greater resource use associated directly with population growth ... from real organizational changes such as the appearance of differential control over labor and increasing per capita expenditures of Iabor on resources procurement or other activities ... The latter represent legitimate intensifications of lab or .. . and changes in economic organization (Arnold 1993:84-85, original emphasis).
Inthis criticism, Arnold misrepresents Waiters' position by decontextualising elements of the model. Waiters does not associate changes in marine resource utilisation with absolute population growth, although he does not dismiss the possibility that this was an associated phenomenon. Rather, his model suggests a reorganisation of economic activities, stimulated by competitive social relations, towards higher levels of production and productivity. This process would presumably involve formalised differentiation of labour organisation and control, especially in the context of production for inter-group gatherings (see Hitchcock 1994). 27
Nolan (1986:97-98) generally criticised arguments which relied on increases in marine productivity to explain changes in the archaeological record:
The major difficultywith the argument for increased marine productivity is that there is no paleoecological evidence to independently demonstrate this at Sandstone Point or elsewhere. Use of archaeological evidence to support the environmental argument involves a very circular logic, and is based on the questionable assumption that Aborigines exploited resources directly in proportion to their availability. If such a correlation is to be made, exponential increases in coastal fauna! assemblages necessarily imply dramatic increases of similar magnitude in the resource base; and where no evidence of coastal exploitation is found, lack of a suitable resource base must be inferred.
Although Waiters' model emphasises the importance of social relations in promoting more
intensive production of marine resources, this 'is a thin veil over the deterministic elements of the model which emphasise a mechanistic economic response to apparent increases in marine resource availability. These issues are further discussed in the next chapter.
Summary
The review of the literature related to Waiters' model undertaken in this chapter reveals three major concerns. First, although investigators agree that significant changes in coastal resource use in southeast Queensland occurred in the Iate-Holocene, they attribute it to different causes.
Second, Waiters' model and other models proposed in explanation of prehistoric cultural change in southeast Queensland share many common elements: (i) time-lags in the use of coastal landscapes and marine resources and the notion of marginality of landscapes for human occupation, (ii) the use of site chronologies to infer changes in the intensity of the use of the region, and (iii) the significanceof changes in marine resource production. In the next chapter
I critically evaluate the first of these assumptions through a review of environmental, ethnohistoric and archaeological data. CHAPTER THREE
THE STUDY REGION: Environmental History, Time-Lags and Marginal Landscapes
Though resource rich for a fishingpeople, the bay and its environs have to be considered as a marginal environment for terrestrially oriented hunting and gathering populations. This is so for the entire Holocene history of the region (Waiters 1992a: l72).
Introduction
Waiters' portrayal of the coastal lowlands as a marginal environment is a key concept in his argument that permanent occupation of coastal southeast Queensland was dependent on the systematic exploitation of marine fishresource_s. Two main inter-related arguments are used to support this claim. First, ecological studies are used to argue that the region has a depauperate terrestrial fauna! resource base and that shellfish communitiesare not resilient to exploitation. Second, Waiters argued fo r a significant time-Jag between sea-level stabilisation and human occupation of the coast, related to the late development of stable and abundant marine resources.
In the first section of this chapter I present an overview of the coastal geomorphology and palaeoecology of southeast Queensland and a description of its major physiographic zones and modernfa una! communities. I then discuss the implications of this environmental information for interpretations of time-lags in marine productivity and concepts oflandscape marginality.
The evidence presented in this chapter illustrates the inappropriateness of generalising ecologically-deterministic adaptive models to landscapes with considerable diversity (see Veth
1993a).
28 29
The Study Region
Southeast Queensland encompasses an area of some 55,000 km2on the central east coast of
Australia (Figure 5). The region is bounded by the Pacific Ocean to the east, the Great
Dividing Range to the west, the Macpherson Range and Tweed River to the south, and the
Burnett River basin to the north (Morwood 1987; Waiters 1992b). Massive triangular embayments dominate the coastal landscape, with Hervey Bay in the north and Moreton Bay in the south. The large sand accumulations which form Moreton, Stradbroke and Fraser
Islands and the Cooloola sandmass are also major fe atures of the region. The region lies between latitudes 24°00' and 28°30' south and longitudes 152°00' and 153°35' east. '
The southeast Queensland region can be divided into three broad physiographic provinces
(Coaldrake 1961; Hall 1982; Hall and Hiscock 1988; Lilley 1984; see Figure 5):
1. the coastal lowlands (or 'wallum');
2. the subcoastallowlands; and
3. the subcoastal highlands.
The coastal lowlands form a discontinuous belt oflow undulating country, generally less than
30 m in elevation, fringing the coast and including the barrier islands. Typically, these environments are characterised by siliceous sands and silts, and by dry sclerophyll fo rests, vine forests, heath and sedge lands, and extensive intertidal mangrove communities (Coaldrake
1961; Dowling 1986; Mather 1976; Specht 1992). Palynological data from Moreton Bay indicate that the vegetation structure of the coastal lowlands is of considerable antiquity (at least 37,000 years; Bell 1979).
I 30
ISLAND
Hi/ %>!'/ Coastal lowlands J\?'.:>:.,:.1 Subcoastal highlands Depths are in metres
Scale
0
BRIBIE ISLAND \ • ORFro( ND �: \ r NORTH STRADBROKE ISLAND
:nz
Figure 5. The study region, showing major physiographic zones. 31
The subcoasta1 lowlands, extendingwest of the D'Aguilar and BeechrnontRanges, generally exhibit shallow leached soils which support a variety of fringing, open and closed forests (Hall and Hiscock 1988; Hiscock 1988). The subcoastal highlands are characterised by closed fo rest and patches of subtropical rainforest (Edmonds 1986). Rainfall throughout the region averages
885 mm, dominated by summer rainfall strongly biased towards the coastal lowlands but with marked variability in volume and distribution (Coaldrake 1961; Mather 1976).
The ecological complexity and diversity of southeast Queensland reflects its status as a transitional zone between tropical and temperate provinces (Keast 1981; Mather 1976). The
' diversity of coastal lowlands ecosystems is illustrated in its 500 species of flora and considerable range of fauna! taxa (Coaldrake 1961). For example, there are 58 species of terrestrial mammals, of which over 50% inhabit coastal lowlands environments (Mather 1976).
The wallum supports over 440 species of birds, including many migratory species (Mather
1976) which congregate along coastal and estuarine waterways (e.g. Welsby in Thomson
1967:427). Reptiles are also well represented, with over 100 species recorded (Covacevich and
Ingram 1975; Mather 1976).
The diverse marine environments of southeast Queensland support over 300 species of fishes in addition to a variety of marine mammals, reptiles and molluscs (e.g. Ellway and Hegerl 1972;
Shine et al. 1973; Thomson 1975; Waiters 1987). The volume of the commercial fishery in
Moreton Bay (around 1,400,000 kg per annum) attests to the present-day abundance of the fisheryresource which is dominated by mullet, bream, whiting, tailor, flathead, mackerel, ray, shark and dart (Queensland Department of Environment and Conservation 1989). 32
Environmental History
The coastal southeast Queensland region assumed its modem form approximately 6,000 years ago when sea-level stabilised fo llowing the last marine transgression (Flood 1981, 1984; Hekel et al. 1979; Ward 1986; Ward and Hacker 1982; Ward et al. 1977). Between about 58,000
BP and 10,000 BP, the floorof Moreton Bay was exposed as dry land, with rivers flowing across its surface (Jones 1992). At around 18,000 BP, the shoreline was up to 40 km east of its present position (Jones 1992; Wi llmott 1992). This means that in the southeast Queensland region alone, approximately 20,000 km2 of land on the continental shelfwas submerged by transgressing seas in the terminal Pleistocene and early-Holocene. '
Geomorphological evidence suggests that there may have been minor variations in sea-level between 6,000 BP and about 3,000 BP. Flood (1984) suggested that the postglacial marine transgression ceased around 6,000-6,500 BP, with regional sea-levels at least 1 to 1.5 m higher than at present (see also Jones 1992). At 4,000 BP, sea-level was approximately 1 m higher and was at least 0.5 m higher 3,000 years ago. Flood (1984:127) argued that the "major phase of seaward pro gradation of the coastal shoreline occurred approximately 3, 000 years ago".
To what degree these changes in sea-level are attributable to eustatic factors, tectonic instability and/or changes in sedimentation is debatable (e.g. Chappell 1982), but the net effect of sea- level instability in the period between about 6,000 and 3,000 BP would have been altered local erosional and depositional patterns (Flood 1981, 1984; Ward and Hacker 1982). Although sea- level fluctuations since 3,000 BP appear to have had little impact on the major landforms of the region, the location of shorelines and the distribution of intertidal and subtidal flats have changed significantly in response to coastal progradation, infilling, changing sedimentary budgets, erosion and the development offiinging coral reefs (e.g. Hekel et al. 1979; Figure 6). 33
A radiocarbon date of 6,250 ± 135 BP (SUA-857) obtained on in situ coral material (Favia sp.) from Peel Island, and a uranium series date of 6,300 ± 500 BP for coral fromthe northern tip ofBribie Island indicate the early development offringing coral reef systems throughout the
MoretonBay area inthe final stages of marine transgression (Hekel et al. 1979:17; Ward et al.
1977). This suggeststhat marine fauna associated with these coral communities may also have been present by this time.
Several investigators have noted a change in the dominant coral species at Peel and Mud
Islands from the clean water Acropora species to the mud-resistant Favia species since 3,710
' ± 250 BP (e.g. Flood 1984:130; Hekel et al. 1979). Hekel et al. (1979) suggested that this shift in species dominance was caused by changing patternsof sedimentation, possibly induced by a change in the direction of the main channel of the Brisbane River, a climatic change or a fall in sea-level. Waiters (1989), however, argued that the increased sedimentation was directly attributable to Aboriginal firing regimes and was a prerequisite fo r the establishment of abundant marine resources. This point is furtherdiscussed later in this chapter.
Palaeoecological data indicate that a variety of marine communities were in place by the end of the marine transgression. For example, radiocarbon determinations on samples of Anadara trapezia from beach-ridge deposits and stranded intertidal flats fr inging Deception Bay indicate the development of intertidal and subtidal flatssupporting shellfish communities by about 6,500
BP (Flood 1981:21; Morwood 1987:339; Nolan 1986:23-24). Similarly, fo ssil Saccostrea commercia/is beds located south of Tingalpa demonstrate the presence of oyster communities by the mid-Holocene (Hekel et al. 1979:9; Nolan 1986:30-3 1). Extensive relict Holocene shellbeds near the mouth of the present Maroochy River also demonstrate the establishment of a variety of estuarine molluscan fauna by about 7,500 BP (Wood and King 1974). 34
BRIBIE ISLAND
ISLAND
NORTH STRADBROKE ISLAND
,. ...• ,,.,,.,,:.:! Tidal delta sand
'"''"'''"' River Mud � River sand SOUTI-I c:=:J Minimal sediment cover STRADBROKE Fringing coral reef t«'/i<'·l ISLAND k>.r&'t:%'$! Prograded shoreline - Palco-rivcr channel Scale 0 25 km
Figure 6. Approximate position of the shoreline of Moreton Bay at 6,000 BP showing areas of recent progradation and reconstructions of the palaeo-Brisbane and palaeo-Pine River channels based on seismic profiling (after Hekel et al. 1979:8; Jones 1992:31) . 35
The presence of more mobile marine species is less likely to be revealed by palaeoecological research but may be attested to by archaeological evidence of their exploitation. For instance,
Morwood (1987:339) noted that dugong remains were present in the Wallen Wallen Creek assemblage by at least 4,290 ± 90 BP (SUA-2466) and linked this with the early development ofseagrass beds and associated fa una! communities (Bell and Pollard 1989; Heinsohn 1991;
Kirkman 1975; Lanyon et al. 1989; Young and Kirkman 1975). Independent evidence from pollen cores also indicates the establishment ofmangrove communities in the region by the mid-
Holocene (Bell 1979).
' Time-Lags
The review presented above demonstrates that elements of the biologically diverse marine environments which characterise modem southeast Queensland were in place by the mid-
Holocene. Although there is geomorphological evidence fo r changes in sedimentation regimes throughout the last 10,000 years, it is more difficultto determine the impact of these changes on marine resource availability or abundance. In fact, there is a general failure in models such as Waiters' to identifYthe biological consequences of transgressive events (cf Jablonski 1980).
Thus models which incorporate the notion of a time-lag in marine productivity in explaining
c late-Holocene patterning of the archaeological record in southeast Queensland are problematic
(Alfredson 1983, 1984a, 1984b; Hall 1982; Hall and Robins 1984; Waiters 1987, 1989, 1992a).
Waiters (1992a: 176) argued that it was not until the late-Holocene that cumulative fluvial sedimentation resulted in the creation of "large areas of mud and sand flats covered with shallow turbid waters and sea grass beds" (Waiters 1992a: 176). Adopting arguments presented by Hughes and Sullivan (1981) fo r eastern New South Wales, Waiters (see also Hall 1990) posited an anthropogenic cause fo r these changes in ecological productivity. He suggested that 36 intensifying Aboriginal firing in the hinterland resulted in episodic erosion and subsequent increases in fluvial sedimentation. These late-Holocene developments were associated by
Waiters (1992a:l 76) with the "evolution of fish stocks on a scale akin to those which today fo rm the basis oflarge contemporary commercial fisheries".
I raise two primary objections to this model of landscape and seascape development fo r
Moreton Bay. Firstly, recent geomorphological studies in easternNew South Wales (Proser
1987; Young et al. 1986) have demonstrated that alluvial infilling exhibits no consistent regional pattern, such as would be expected to result from large-scale climatic events or intensifying Aboriginal firing regimes. Rather, the alluvial chronology shows a random distribution of infilling events that are not confined to the late-Holocene. The non-clustered distribution ofalluvial chronologies is suggestive of individual alluvial histories controlled by local factors and independent oflarge-scale events either natural or cultural (Proser 1987:201).
Generally, extreme rainfall episodes, rather than small-scale erosional events, are seen as important determinants of river sediment load (Harris 1995). Nanson (1986) and Young et al.
(1986) have argued that the concentration of dates fo r alluvial deposition in the Holocene, which is identified by Hughes and Sullivan (1981) as resulting fr om Aboriginal firing, actually reflects an analytic bias in the representation of more recent events owing to the destruction of organic materials suitable fo r dating in oxidising environments, the redeposition of alluvium in localised sedimentary processes, and the complete removal of material in catastrophic stripping events.
Secondly, sediment sources other than rivers have been shown to be important in the development of intertidal and subtidal flats in Moreton Bay (Harris et al. 1992). Longshore drift has transported oceanic sands into the Bay, and today, the main sediment sources are the 37 continental shelf and dune-barrier islands (Maxwell 1970). In fa ct, fluvially-derived muds are restricted to the westernBay (Jones and Stephens 1981; Maxwell 1970), while tidal flats in the easternBay are composed almost exclusively of quartzose sand (Harris et al. 1992; Jones 1992; see Figure 6). Nevertheless, fluvial sedimentation has long been active in the regionand is responsible for the massive progradation around the Brisbane River delta (Hekel et al. I 979;
Jones 1992). While rates of sedimentation and shoreline progradation have varied through time, affecting the local distribution and habitat of many marine species, it must be recognised that these processes are localised and on-going.
The above discussion does not seek to deny the role of Aboriginal societies as active agents in landscape and seascape development or changes in biological productivity, but cultural modification oflandscapes needs to be demonstrated, not assumed or simply asserted. Recent studies in southeastern Australia (Dodson et al. 1993) demonstrated the complexity of inter relationships between fire regimes, vegetation response, erosion and eutrophication.
IdentifYingthe nature and scale of prehistoric anthropogenic influences on the landscape is a complex enterprise which requires detailed studies at a local level, not at the level of generalisation employed by Waiters (see Head 1994).
Marginal Landscapes
Coaldrake's (1961) seminal study of the coastal lowlands of southeast Queensland provided the basis for ecological models characterising the wallum as a marginal landscape (e.g. Hall
1982, 1987; McNiven 1990a, 1991a; Waiters 1987, 1989, 1992a, 1992c). Coaldrake (1961) defined the coastal lowlands as an area exhibiting "common fe atures such as assured rainfall, soils with a degree of similarity in morphology and lack offertility, and plant communities with marked floristic and structural affinities", and divided the region into 20 'landscapes' which 38 share broadly similar plant assemblages and soils. The vegetation structure of the coastal lowlands is described as a mosaic environment constantly in a state of flux. Thus the coastal lowlands may be more accurately characterised as a heterogeneous environment, sharing some structural attributes, but exhibiting a variety of potential resource zones.
Waiters (1989, 1992a) argued that the ecological structure of the coastal lowlands and a late increase in marine productivity were major factors inhibiting permanent occupation of the coast. Infa ct, the coastal lowlands have an extremely narrow and restricted spatial distribution
(see Figure 5) and thus it is difficult to believe that this landscape system presented a significant barrier to human occupation of the coast. Wahers' generalised representation of the coastal lowlands as a single marginal environmental zone engendering a single mechanistic human response ignores the variability of wallum environments and the diversity of Aboriginal land management practices. Another major objection is that although the coastal lowlands of southeast Queensland exhibit a low overall abundance of terrestrial fauna, these landscapes interface with extremely productive intertidal, estuarine and marine ecosystems.
Terrestrial Resources
Ecological studies undertaken in the 1970s (e.g. Dwyer et al. 1979a; Dwyer et al. 1979b;
Kikkawa 1975) characterised the coastal lowlands as having a depauperate terrestrial fauna! resource base linked to the almost uniformly low nutrient status of soils. Thus, although the region supports a relatively high diversity of fauna! species, the abundance of each taxa is generally low. It is possible, however, that large terrestrial mammals in particular were more abundant in the recent prehistoric past than at present ( cf Dwyer et al. 1979a), especially if
Aboriginal firing regimes were operating (see Coaldrake 1961). Waiters (1987:318) himself conceded that "burning and opening up of wallum scrublands may have established grazing 39 conditions which allowed macropods to flourish". Coastal low1ands environments may have been well suited to fire management regimes, given the documented pattern of grassy understorey development and maintenance in response to fire (Coaldrake 1961). Further, the validity of extrapolations of the results of these ecological studies to the pre-European period is dubious, given that they were conducted in the 1970s after major European landscape modification and the disruption of indigenous management regimes which may have dramatically impacted on the distribution and abundance of some taxa.
Further problems with this marginality model fo r the coastal lowlands come from the ethnohistoric literature which suggests that co'astal peoples utilised a broad range of marine, littoral and terrestrial resources fromthis zone. Aswell as marine fishing, fern-root processing and shellfish gathering, terrestrial resource exploitation also emerges as an important component of coastal economies:
Aside fr om the hunting of marine animals men were engaged in terrestrial hunting. Macropods were caught by spear or were despatched after being herded into nets staked out at strategic points ... Return boomerangs were used to scare birds into nets stretched between trees; emus were caught in nets in the manner described fo r macropods. Black swans were easily hunted during their summer moult when they were flightless and other aquatic birds (e.g. ducks) were often grabbed by the fe et from under the water. Flying foxes were knocked out oftrees with throwing clubs and sticks. Possums and koalas were caught during the day by scaling trees and cutting the animals out of hollow places .. . Snakes, lizards, tortoises, goannas and grubs were also sought. Honey (' sugar bag') was prized and was especially plentiful on the offshore islands (Hall 1982:86).
Several accounts document the use oflarge nets in the co-operative hunting of macro pods (e.g.
Fairholme 1856; Petrie 1904) in the coastal lowlands. The considerable labour invested in the manufacture and maintenance of these fa cilities suggests that this form offood production was an important activity (Satterthwait 1986, 1987). Clearly, we would not expect this kind of large scale hunting activity if, as Waiters (1992a: 168; also 1987:70, 317) argued, "Aboriginal 40 hunting ... quickly reduced levels of abundance in the terrestrial animal populations" after initial occupation of the coastal lowlands.
While marine fishingwas obviously important in recent Aboriginal economies of the region, as is evident in oral histories and the ethnographic literature, I suggest that the role of marine fish resources has probably been overemphasised by recorder bias and that important, but perhaps less visible, production activities associated with terrestrial resources (such as fernroot production) have been underemphasised (see Durbidge and Covacevich 1981; Whalley 1987).
This proposition appears to be supported by evidence from stable carbon isotope analysis of
' skeletal material from the Broadbeach Burial Ground. Studies conducted by Hobson and
Collier (1984, 1986; Collier and Hobson 1987) indicated that marine resources probably accounted fo r less than half of the dietary intake of these people:
The study of stable-carbon isotope ratios of dietary species reported here reveals that the sea was not the major provider of fo od to coastal people in southeastern Queensland, an environment abundant in marine fo od resources ... We have clear evidence that fo r at least one coastal economy the land was more important than the sea in the provision of fo od (Collier and Hobson 1987:563; added emphasis).
If we can assume that this sample is representative of the coastal population as a whole, it indicates that in the late-Holocene, with apparently high population densities in the region (e.g.
Hall 1982, 1987; Hall and Hiscock 1988; Morwood 1987; Waiters 1989, 1992a) and presumably greater pressure on resources, the majority of subsistence production requirements were met by terrestrial resources.
Shellfish
On the basis of a study by Catterall and Poiner (1984), Waiters (1987, 1989, 1992a) argued that Moreton Bay shellfish communities were not resilient to Aboriginal exploitation. He used 41 this argument to bolster his representation of marine fishproduction as a necessary prerequisite for permanent coastal settlement, concluding that "a subsistence economy of a reasonably large population of humans could scarcely be founded upon shellfish" (Waiters 1989:218).
Catterall and Poiner's (1984, 1987) study was based on data collected from two locations on the western coast ofNorth Stradbroke Island. The combined sample from these sites consisted of 2,142 bivalves and gastropods larger than 10 mm, collected within 78 m3 of randomly distributed quadrats of 0.5 x 1 m, excavated to a depth of 80 mm. Their assessment of resilience to exploitation was based on the assumption that subtidal and buried individuals are less vulnerable to human predation. Size at maturity, mobility and duration of the pelagic larval phase were also considered.
There are several objections to Waiters' unqualifieduse of this study to minimise the potential role of shellfish in the prehistoric subsistence economy. First, the sample was very small considering that there are approximately 1,000 km2 of intertidal and subtidal flats in the
Moreton Bay area. In addition, the random sampling procedure employed excluded clumps of spatially-aggregated shellfishthat may have been targeted by gatherers (Catterall and Poiner
[1987: 119] note that dense clumps of hairy mussel, Tr ichomya hirsuta, fell immediately outside the sampled zone).
Second, the resilience of species was assessed on the basis of biological life-history and habitat attributes without consideration of factors such as the duration, fr equency and intensity of shellfish gathering events or the possibility of human modification of shellfish availability.
Deliberate management of shellfish has been documented ethnographically which resulted in the manipulation of the population structure of shellfish communities towards greater 42 productivity by removing competition and by moving shellfish to repopulate depleted beds or to colonise new areas (Meehan 1982; Ross and Coghill n.d.).
Finally, and most importantly, only shellfish in low energy contexts were considered by
Catterall and Poiner (1984, 1987). Large midden accumulations on the ocean coasts of southeast Queensland consist predominantly of pipi (Donax deltoides), which form large spatially aggregated communities inthe intertidal zone of high energy beaches (e.g. Hall 1980b;
McNiven 1985; Richardson 1984). Luebbers (1994), in fa ct, argued that Donax deltoides are highly resilient to exploitation.
'
It is also worth noting that Waiters appears to misrepresent the results of Catterall and Poiner' s
(1984, 1987) study in that, contrary to his reading of the work, not all shellfish species are characterised as vulnerable to over-exploitation. In fact, of the five types of shellfish considered by the study, three (Anadara trapezia, Strombus luhuanus and the family
Veneridae) were thought to have "little risk oflocal depletion due to traditional gathering"
(Catterall and Poiner 1987:121). Only oyster (Saccostrea commercia/is) and whelk (Pyrazus ebeninus) were found to have low resilience to exploitation.
Although shellfish may never have been a primary dietary component throughout the year, they may have fo rmed a significant back-up resource. Recent re-evaluations of coastal economies
(e.g. Erlandson 1988, 1991; Meehan 1977, 1982) have demonstrated that shellfish may even out irregularities in fo od production, particularly as a source of protein.
Discussion
Pre-European subsistence strategies ofAboriginal societies in southeast Queensland have been modelled by several investigators to reflect a specialised marine economy operating in an 43 otherwise marginal environment (e.g. Draper 1978, 1980; Hall 1982; Hardley 1975; McNiven
1992c; Waiters 1987). However, I argue that although based on the coast, people derived a substantial part of their subsistence requirements fr om terrestrial resources. Fish, which are often regarded as a staple fo od resource in Aboriginal economies in the region, may have been secondary to other resources. In light of ethnohistoric accounts and the stable carbon isotopes analysis of the Broadbeach Burial Ground skeletal material, I suggest that current models emphasising the primacy of marine resources, particularly fish, may require significant revision.
Environmental generalisations characterising the coastal lowlands as a marginal landscape have
' uncritically entered the literature as established ecological assumptions which have informed models of subsistence and settlement systems in the region (e.g. Hall 1982, 1987; Hall and
Robins 1984; McNiven 1990a, 1991a; Morwood 1986, 1987; Nicholson and Cane 1994;
Waiters 1987, 1989, 1992a, 1992c). These models have relied on a very generalised view of coastal lowlands landscapes, characterising them as uniformly marginal and thereby reducing the dynamic nature of human-environment interaction to one of homogeneous structure and mechanistic response. I have demonstrated that such marginality fo r the coastal lowlands of southeast Queensland is by no means proven.
Summary
Problems in Waiters' approach to the landscape component of his model limit the reliability of the conclusions reached in his study. His model fo r increases in marine productivity induced by anthropogenic fluvial sedimentation is clearly simplistic and does not take into account the variability and impact of changing sedimentation regimes. The palaeoecological data reviewed above suggest that there was no significant time-Jag in the establishment of marine communities after sea-level stabilisation. 44
The review of palaeoecological, environmental, ethnohistoric and palaeoanthropological evidence presented inthis chapter calls into question the validity of simplistic characterisations of the coastal lowlands as marginal landscapes fo r human occupation (e.g. McNiven 1990a,
1991a; Neal and Stock 1986; Waiters 1987, 1989, 1992a). I argue that the coastal lowlands are productive landscapes, particularly as they interface with extremely productive intertidal and marine ecosystems. Finally, I suggest that Waiters may have underestimated the role of shellfish communities and terrestrial plant and animal resources in the prehistoric economy. In sum, there is no reason to believe that at any time during the Holocene the hinterland was any more favourable than the coastal lowlands fo r human occupation. ' CHAPTER FOUR
CHRONOLOGY IN SOUTHEAST QUEENSLAND: Exploring the Temporal Structure of the Archaeological Record
Radiocarbon dating provides the only reliable basis for building a local chronological sequence, but even here some critical methodological issues need to be addressed (Bird and Frankel I 991 a:2).
Introduction
Inthis chapter I present an analysis of the radiocarbon chronology of pre-European Aboriginal
occupation of southeast Queensland. Analytical methods are outlined and their limitations
briefly discussed with reference to sample structure and reliability. The analysis of these data
' provides the basis for evaluatingWaiters' claimfor a time-Jag of 4,000 years between sea-level
stabilisation and permanent occupation of coastal southeast Queensland, when exponential
growth of coastal sites occurred with no major shifts in the number of inland sites.
The radiocarbon date sample employed in this analysis was constructed through a survey of
published and unpublished sources which report research in the study region. This search
involved systematically reviewing Queensland Archaeological Research and relevant
unpublished theses submitted at The Un iversity of Queensland and The University of New
England as well as references contained in these works. In addition, individual researchers
were contacted to seek access to otherwise unavailable data. This search resulted in the
location of 140 dates from 59 sites (Table 3; Figure 7), including several dates which were not
previously reported. Although some dates were not available for inclusion, it is estimated that this sample represents as much as 90% of determinations made on archaeological materials in
southeast Queensland (DrJ. Hall,Department of Anthropology and Sociology, The University
of University, pers. comm., 1995).
45 46
Table 3. Radiocarbon dates fromarchaeological sites in southeast Queensland (n=l40).
Site C-14 Age Lab. No. Sample Reference Calibrated Age/sa
Aranarawai Beach Ridge II 1880 ± 80 Beta-24538 charcoal Neal 1989 1939(1735)1551
Bishop's Peak 1420 ±60 Beta-16298 charcoal Edmonds 1986 1380(1290)1174
2620 ±90 Beta-16299 charcoal Edmonds 1986 2852(2742)2357
Boonah Shelter 1300 ± 60' Beta-2 5205 charcoal Morwood pers comm 1291(1174)1005
1960 ± 110' Beta-2 1276 charcoal Morwood 1987 2124(1866)1568
3240 ±50 Beta-25204 charcoal Morwood pers comm 3540(3392)3276
Booral Homestead Midden 2400 ± 110 Beta-38414 charcoal Frankland 1990 2743(2348)2122
2420 ±80 Beta-32045 charcoal Frankland 1990 2723(2352)2 157
Booral Shell Mound 980 ±60 Beta-37394 charcoal Frankland 1990 954(904,853,835,806,797)720
1750 ±60 Beta-36303 charcoal Frankland 1990 1730(1603,1590,1576)1503
2480 ± 60 Beta-36304 charcoal Frankland 1990 2736(2460,2384,2377)2341
2660 ±60 Beta-38415 charcoal Frankland 1990 2844(27 50)2507
2790 ± 80 Beta-38242 charcoal Frankland 1990 3063(2846,2819)2744
2950 ± 60 Beta-32046 charcoal Frankland 1990 3216(3061,3050,3013)2865
Bribie Island 9 200 ± 80 Beta-56565 charcoal Smith 1992 421(267 ,205,143,17,0*)0*
3280 ± 80 Beta-56566 charcoal Smith 1992 3634(3462)3270
Broadbeach Burial Ground 26 ± 86' V-161 charcoal Haglund 1976 278(0*)0*
450 ± 70 ANU-67 charcoal Haglund 1976 543(488)299
1086 ±91' V-157 charcoal Haglund 1976 1221(969)786
1156 ± 110' V-162 charcoal Haglund 1976 1290(1061)796
' 1290 ± 70 ANU-68 charcoal Haglund 1976 1294(1171)982
Brooyar Shelter 2650 ±60 Beta-23344 charcoal McNiven 1988 2794(2748)2496
Brown's Road 2030 ±70 Beta-3077 charcoal Kelly 1982 2116(1931)1747
Bushranger's Cave 700 ± 60 Beta-42843 charcoal Hall pers. comm. 681(648)539
1980 ±70 Beta-42844 charcoal Hall pers. comm. 2009(1874)1710
2090 ± 90 Beta-18897 charcoal Hall & Hiscock 1988 2307(1989)1814
3460 ± 80 Beta-44134 charcoal Hall pers. comm. 3865(3683,3670,3638)2467
4720 ±100 Beta-4851 charcoal Hall & Hiscock 1988 5607 (5447 ,5404,5328)5050
5540 ± 100 Beta-4852 charcoal Hall & Hiscock 1988 6481(6294)6036
5820 ± 110 Beta-42845 charcoal Hall pers. comm. 6855(6624,6590,6566)6313
8090 ± 110 Beta-42846 charcoal Hall pers. comm. 9252(8979)8551
9270 ± lOO' Beta-42847 charcoal Hall pers. comm. I 0435(1 0278,1026 1 ,I 0 1 95)9984 47
Table 3. Radiocarbon dates fro m archaeological sites in southeast Queensland (n=l40) (cont.).
Site C-14 Age Lab. No. Sample Reference Calibrated Age/sa
Cameron Point Site 62 190± 50 Beta-34400 charcoal McNiven 199lb 293(264,2 12,14 I ,2 I ,0*)0*
950 ±60 Beta-34401 charcoal McNiven 199Ib 937(788)675
Christmas Creek 1000 ± 70 Beta-54588 charcoal Bonica !992 978(913)720
1990 ±90 Beta-54589 charcoal Bonica 1992 2116(1878)1635
3270 ± 60 Beta-44841 charcoal Bonica !992 3577(3460)3345
Cribb Island 390 ±50 Beta-3227 shell Ward & Hacker 1981 modem
1300 ± 90 Beta-3226 shell Ward & Hacker I 98 I 1035(857)654
Deception Bay 1 1000 ± 140 LJ-949 shell Kelly 1982 833(563)322
Deception Bay 2 760 ± 140 LJ-952 shell Kelly 1982 627(420)0
Deeeption Bay 3 995 ± 70 SUA-1257 shell Nolan 1986 683(558)472
Deception Bay 4 650 ± 130 SUA- 1256 shell Nolan i986 517(280)0
Deception Bay 5 755 ± 135 SUA- 1255 shell Flood 1981 619(410)0
Double Island Point I 160 ±90 Beta-34059 charcoal McNiven I 990a 309(248,232, I 3 I ,90,33,0*)0*
First Ridge 1150 ±70 Beta-1 946 shell Kelly 1982 869(680)557
Fraser Island 217/15 1960 ± 110 Beta-1698 shell Kelly 1982 1788(1510)1274
Fraser Island 796/54 1835 ± 85 Beta-1701 shell Kelly 1982 1562(1356)1214
1965 ± 100 Beta-! 700 shell Kelly 1982 1767(1513)1287
Fraser Island 799/54 1270± 80 Beta-1699 shell Kelly 1982 970(797)649
Gatton Shelter 1090± 70 Beta-5897 charcoal Morwood 1986 1070(945)788
3030 ±90 Beta-5898 charcoal Morwood 1986 3375(3203,3192,3 162)2876
3820 ± 120 Beta-15811 charcoal Morwood 1986 '4 508(4 142,4 1 04,4097)383 1
Hope Island 1500 ± 80 Beta-20800 charcoal Walters eta/. 1987 1524(1331)1256
2600 ± 70 Beta-20797 charcoal Waiters et al. 1987 2778(2734)2362
3720 ± 70 Beta-20798 charcoal Waiters et al. 1987 4228(3984)3780
4350 ± 220 Beta-20799 charcoal Waiters et al. 1987 5566(4860)4277
Kabali Road Scatter 70 ±70 Beta-19424 shell McNiven 1990a modem
King's Bore Site 97 2290 ± 80 Beta-30402 charcoal McNiven 1992d 2359(2314,2224,221 3)20 I 6
3560 ± 100 Beta-25510 charcoal McNiven 1992d 4082(3824,3785,3749 ,3734)3488
Little Sandhills 102.1 ± 0.7% Beta-1947 charcoal lv1RAPFiles 0*(0*)0*
Maidenwell Shelter 1210 ± 100 SUA- 1915 charcoal Morwood 1986 1288(1064)9 18
4300 ± 70 Beta-6924 charcoal Morwood 1986 4980(4835)4571
Maroochy River 2 610±70 Beta-27085 shell McNiven 1989 423(260)0 48
Table 3. Radiocarbon dates fromarchaeological sites in southeast Queensland (n=l40) (cont.).
Site C-l4Age Lab. No. Sample Reference Calibrated Age/sa
Maroochy River 4 170 ±60 Beta-27087 shell McNiven 1989 mcxlern
840 ±60 Beta-27086 shell McNiven 1989 550(474)3 13
Minner Dint 520 ± 75 I- 11095 charcoal Hall 1980b 641(514)319
Moon Point 1470 ± 80 Beta-2609 shell Kelly 1982 1206(995)841
New Brisbane Airport 1 120 ± 60 SUA-2 179 charcoal Hall & Lilley 1987 1080(966)9 13
39 10± 80 Beta-23345 charcoal Hall & Lilley 1987 4513(4277)3994
4830 ± 110 Beta-33342 charcoal Hall pers. comm. 5738(5577,5511,5493)5296
NEMoreton Is. 1 430 ±60 Beta-24473 shell Lilley & Hall 1988 modem
NEMoreton Is. 5 380 ± 60 Beta-24470 shell Lilley & Hall 1988 modem
NEMoreton Is. 16 400 ± 60 Beta-24471 shell Lil!ey & Hall 1988 modern
NEMoreton Is. 21A 320 ±50 Beta-24472 shell Lilley & Hall 1988 modem
NEMoreton Is. 27 440 ±60 Beta-24473 shel! Lilley & Hall 1988 modem
NRS Site 8 470 ±60 ANU-3336 shell Richardson 1984 264(70,10)0
NRS Site 19 450 ± 60 ANU-3337 shell Richardson 1984 modern
NRS Site22 410± 50 ANU-3338 shell Richardson 1984 modern
One-Tree 1620 ±60 Beta-3073 charcoal Ha!l 1982 1570(1491 ,1472, 1440,1419) 1328
Platypus Rockshelter 560 ± 60 Beta-3076 charcoal Ke!ly 1982 643(529)476
2420 ±90 I- 11094 charcoal Ke!ly 1982 2736(2352)2 151
2480 ± 70 Beta-3075 charcoal Ke!ly 1982 2743(2460,2384,2377)2334
3850 ± 170 SUA- 1502 charcoal Ke!ly 1982 4809(422 1 ,4207 ,4153)3698
4540 ± 80 Beta-3074 charcoal Hall & Hiscock 1988 5444(5246,5186,5122,5056)4867
Polka Point 765 ±45 V-22 charcoal Bermingham 1966 719(663)567
1560 ± 80 Beta-28487 shell Neal 1989 1279(1 107)924
1750 ±90 Beta-28488 charcoal Neal 1989 1 826(1603, 1590,1576) 1402
3020 ± 100 Beta-24540 charcoal Neal 1989 3380(3200,3 197,3 1 54)2860
Saint-Smith Midden 260 ±60 Wk-3424 charcoal Hall 1995 430(284)0*
960 ± 60 Wk-3423 charcoal Hall 1995 940(886,875,823,817, 791 )693
Sandstone Point 103 ±0.8% Beta-15808 charcoal Nolan 1986 modern
320 ±50 Beta-15806/A charcoal Nolan 1986 469(303)0*
340 ± 50 Beta- 15806/B charcoal Nolan 1986 479(309)0*
500 ±50 SUA-2358 charcoal Waiters 1987 546(509)334
620 ± 95 SUA-478 charcoal Ke!ly 1982 674(550)472 49
Table 3. Radiocarbon dates fro m archaeological sites in southeast Queensland (n=l40) (cont.).
Site C-14Age Lab. No. Sample Reference Calibrated Agefs•
Sandstone Point 740 ±50 SUA-2357 charcoal Waiters 1987 705(657)554
740 ±80 Beta-15809 charcoal Nolan 1986 737(657)539
780 ±95 SUA-479 charcoal Hall 1982 899(666)542
810± 80 Beta-16837 charcoal Nolan 1986 899(672)557
1190 ± 100 Beta-15805 charcoal Nolan 1986 1283(1 059)800
1500±110 Beta-15807 charcoal Nolan 1986 1554(133 1)1167
1600 ± 80 Beta-16838 charcoal Nolan 1986 1611(1412)1296
1990 ± 60 Beta-15810/A charcoal Nolan 1986 2000(1878)1726
2290 ± 100 Beta-1 58 10/B charcoal Nolan 1986 2469(23 14,2224,22 13)1990
Seary-Broutha Site 100.5 ±0.9% Beta-19423 shell McNiven 1990a modem
Spitfire Creek 450 ±70 SUA-2082 shell Neal 1984 modem
560 ±70 SUA-2081 shell Neal l984 331(230)0
St Helena Island 1370 ±60 Beta-6 140 shell Alfred son 1984a 1050(916)754
2240 ± 70 Beta-6 141 shell Alfredson 1984a 2008(1834)1647
Talgai Skull 1 1650 ± lOO' - sediment Oakley el al 1975 13851 (13538) 1 3285
Teewah Beach 5D 100.1 ±0.9% Beta-19422 shell McNiven 1990a modern
Teewah Beach 26 340 ± 70 Beta-30401 charcoal McNiven 1990a 504(309)0*
950 ± 100 Beta-25511 charcoal McNiven 1990a 984(788)659
1070± 70 Beta-30399 charcoal McNiven 1990a I 065(936)782
3140 ± 100 Beta-30400 charcoal McNiven 1990a 3540(334 1 ,3281 ,3278)2994
4780 ± 80 Beta-25512 charcoal McNiven 1990a 5642(5562,5549,5468)5299
Tin Can Bay 75B 700 ± 70 Beta-19421 shell McNiven 1991b 485(3 13)158
Toulkerrie 310±80 Beta-32049 charcoal Hall & Bowen 1989 502(299)0*
350 ± 80 Beta-32796 charcoal Hall & Bowen 1989 512(312)0*
370 ± 75 I- 1 1096 charcoal Hall 1982 515(425,392,3 19)0*
2150±80 Beta-32048 charcoal Hall & Bowen 1989 2323(2063)1878
2290 ± 80 Beta-32047 charcoal Hall & Bowen 1989 2359(23 14,2224,22 13)2016
Wallen Wallen Creek 1070 ± 50 SUA-2461 shell Neal & Stock 1986 719(637)530
1230 ±50 SUA-2465 shell Neal & Stock 1986 906(755)654
1520 ±70 SUA-2236 shell Neal & Stock 1986 1241(1058)911
2120 ± 70 SUA-2462 charcoal Neal & Stock 1986 2303(2007)1874
4290 ± 90 SUA-2466 charcoal Neal & Stock 1986 5027(4833)4531 50
Table 3. Radiocarbon dates fromarch aeological sites in southeast Queensland (n=l40) (cont.).
Site C- 14 Age Lab. No. Sample Reference Calibrated Age/s"
Wallen Wallen Creek 6950 ± 80 SUA-2343 charcoal Neal & Stock 1986 7693(7673)7656
9760 ± 140' SUA-2467 charcoal Neal & Stock 1986 11319(10957)10390
9810 ± 130' SUA-2251 charcoal Neal & Stock 1986 11488(10977)10480
!1990 ± 70' SUA-2344R charcoal Neal & Stock 1986 14233(13931)13672
13040 ±220' SUA-2463 charcoal Neal & Stock 1986 16113(15437)14661
13650 ± 240' SUA-2342 charcoal Neal & Stock 1986 16925(16307)15609
16420 ± 260' SUA-2464 charcoal Neal & Stock 1986 20044(19279)18712
17500 ± 900' SUA-2235 charcoal Neal & Stock 1986 18549(20770)0*
20560 ±250' SUA-2341 charcoal Neal & Stock 1986 no calibration possible
Webber Swamp 100 101.1 ± 0.8% Beta- 19425 shell McNiven 1990a modem
' WhitePatch Site 3' 450± 90 SUA-480 charcoal Crooks 1982 553(488)287
670 ±95 SUA-481 charcoal Crooks 1982 721 (640,597,565)504
NOTES a Dates werecalibrated using the following procedure (after Stuiver and Reimer 1993): 1. Conventional radiocarbon ages were calibrated using the CALIB (Version 3.03) computer program (Stuiver and Reimer 1993). 2. Atmospheric samples were calibrated using the hi-decal atmospheric calibration curve with no laboratory error multiplier. 40 years were subtracted from atmospheric samples to correct for C-14 variations between northern and southern hemispheres.
3. Marine samples were calibrated using the marine calibration model with a .6.Rregional correction value of-5 ± 35. 4. The calibrated ages reported span the 2 sigma calibrated age range. b Excluded from analysis. See text. c These dates were corrected for isotopic fractionation using a correction estimate of-24 ± 2_� as recommended by Stuiver and Reimer (1993). d Excluded fr om analysis - calibrated age/s over 10,000 BP. See text. e Also referred to in the literature as Bribie Island Site 3.
0* Represents a 'negative' or 'modern' age BP. 51
• Site ''') L '''i\J Coastal lowlands 1,>:-;','''d Subcoastal highlands
Depths are in metres
Scale
0
2/A & 27
:n<-
Figure 7. Southeast Queensland, showing radiocarbon dated archaeological sites. 52
Radiocarbon Data
Table 3 lists the radiocarbon data used in this analysis and includes the site name (as reported), mean and standard deviation of conventional C-14 ages, laboratory code, sample material, citation and calibrated agels. Where it was not possible to cite a published source, unpublished theses, reports and personal communications are acknowledged. All conventional radiocarbon ages were converted to calendar years using the CALm (Version 3.03) computer program
(Stuiver and Reimer 1993). C-14 ages inthe second column are as reported without correction for marine reservoir effect or other error, with the exception of three dates from the
Broadbeach Burial Ground, which are listed with a correction estimate fo r isotopic ' fractionation (Stuiver and Reimer 1993). Five dates in the sample (from Deception Bay sites
1-5) were initially undertaken as part of geomorphological research and later identified to be fr om cultural deposits (see Nolan 1986).
Although various 'chronometric hygiene' criteria have been implemented in archaeological studies based on radiocarbon dates (e.g. Anderson 1991; Meltzer and Mead 1985; Spriggs and
Anderson 1993), I have chosen to only exclude determinations which are clearly problematic
on the basis of their archaeological associations. The date for the Talgai Skull listed in Table
3 is one of 14 determinations obtained on calcareous sediments reportedly associated with the human cranial remains (Hendy et al. 1952; Oakley et al. 1975; Smith and Sharp 1993). There
is no firm evidence for directly relating these determinations with the actual antiquity of the remains. Because of the equivocal nature of these dates they have been excluded from the
analysis. The three dates from Boonah Shelter are inverted, although Morwood (Department
of Archaeology and Palaeoanthropology, The University ofNew England, pers. comm., 1995)
suggests that the age of 3,240 ± 50 BP (Beta-25204) is accurate and that the other two 53 detenninations (Beta-21276 and Beta-25205) were undertaken on intrusive organic material and should be discarded. I have chosen to follow Morwood's recommendation in excluding these two detenninations. A single detennination in the Wallen Wallen Creek sequence (SUA-
2344R) is clearly anomalous and is thought to have been caused by ground water contamination (Neal and Stock 1986:620).
Only calibrated dates are used in the fo llowing analysis to ensure comparability of the time scale used in assessing changes in temporal patterns of human occupation (see Ward 1994 for a discussion). Only one determination (SUA-2341) from southeast Queensland is presently ' beyond the limits of available calibration data. Electron Spin Resonance (ESR) dating is the only other technique which has been employed to date archaeological deposits in the region with limited success (see Alfredson 1984a:128-132; Caddie et al. 1984), although both thermoluminescence (TL) techniques and uranium series dating have been used in geomorphological applications (Hekel et al. 1979; Tejan-Kella et al. 1990; Ward et al. 1977).
In this analysis I only include determinations obtained using radiocarbon dating techniques.
Asthe purpose of this chapter is to examine Waiters' model fo r Holocene change in southeast
Queensland, only calibrated ages under 10,000 years are considered in this analysis. The exclusion of older dates does not significantly alter the sample structure as only eight unequivocal determinations from two sites (Bushranger's Cave and Wall en Wall en Creek) exceed 10,000 BP. The exclusion of problematic determinations (one from Talgai Skull, two from Boonah Rockshelter, and one from Wallen Wallen Creek) and the eight determinations which exceed 10,000 BP reduces the sample used in this analysis to 128 radiocarbon dates from
58 sites. 54
Site Definition
Site definitionis an important consideration, as the 'site' fo rms a basic analytical category for determining the frequency of site occupation at various time scales. Investigators in southeast
Queensland have generally adopted a traditional site-based approach in both description and analysis (cf. McNiven 1985, 1992a), where a site is defined as "a locus, or concentration, of cultural materials, or a concentration of evidence of human activity'' (Sullivan 1983 :2). This approach has major limitations in accommodating non-discrete, spatially continuous distributions of cultural material across the landscape (e.g. Dunnell and Dancey 1983; Foley
1981a, 1981b). '
Most dated deposits included in the sample are separated by significant distance, which alleviates this definitional problem to an extent. Several sites included in this analysis, however, are closely juxtaposed. The dated deposits at Toulkerrie, Little Sandhills and One Tree on southern Moreton Island (Figure 7) are all located within three kilometres of each other and could be argued to fo rm part of the same discontinuous shell midden deposit. Similarly, the fiveNortheastern (NE) Moreton Island sites and the three NRS sites from Stradbroke Island could be argued to belong to single site complexes.
Sample Structure
The comparability and reliability of the radiocarbon data set used in this analysis is enhanced by two features. Firstly, the majority of determinations (71.88%, n=92) were made at a single laboratory, thereby minimising the effects of inter-laboratory variability in sample preparation and counting procedures (Figure 8). Secondly, only a small proportion of determinations in the sample (n=37) were made on marine shell, with over 70% made on charcoal (n=9 1). This bias is important in confined areas such as Moreton Bay where there is the possibility of older 55 terrestrial carbon from fluvial activity being incorporated into shell structures (see Little 1993).
Studies of marine reservoir effect havenot been undertaken in the immediate study area (see
Bowman and Harvey 1983; Gillespie and Polach 1979; Head et al. 1983). Researchers have generally corrected radiometric determinations obtained on marine shell by simply subtracting
450 years (e.g. Alfredson 1984a) and some have then calibrated determinations using the bi- decal atmospheric calibration curve (e.g. McNiven 1989). Applying this standard correction value, however, does not take into account the fact that oceanic reservoir ages are not constant, but rather fluctuate through time with changes in reservoir parameters such as "input and output fluxes and exchange with the atmosphere" (Stuiver et al. 1986:980). In this analysis I ' have applied the global marine calibration model with a generalised <>R regional correction value of -5 ± 35 suggested by Stuiver and Braziunas (1993) fo r marine shell determinations.
This <>Rvalue is the weighted mean ofthe difference between the known historical age and the conventional C-14 age for marine shell samples from the region based on results presented by
Gillespie and Polach (1979; see Stuiver et al. 1986:Table 1).
Wk
ANU
V I
Key: Beta - Beta Analytic; I - Teledyne Isotopes; V-Victoria Radiocarbon Laboratmy; ANU-Australian National University; U-University of Califor nia; SUA - Sydney University; Wk-Waikato.
Figure 8. Distribution of dates by laboratory. 56
Three broad typesof dated archaeological sites occur inthe study region: (i) coastal middens,
(ii) subcoastal rockshelters, and (iii) cemeteries (Figure 9). Many other site types (such as unstratified artefact scatters, earthen circles, fishtraps and rock art sites) are found in southeast
Queensland but none has been successfully dated with radiometric techniques (see Hall and
Hiscock 1988).
Cemetery
a% of Sites Rockshelter l!ll %of Dates
'
Midden
0 10 20 30 40 50 60 70 80 90
Figure 9. Frequency of major site types and proportion of dates fo r each major site type.
Asis clear from Figure 9, coastal middens dominate the sample, comprising over 84% of dated sites and over 76% of radiocarbon determinations. Together, rockshelters and cemeteries account fo r less than 20% of the sample, although they have marginallJ;more radiocarbon dates per site than mid den deposits. The dominance of midden deposits in the sample reflects the general coastal orientation of archaeological research in the region.
Over 55% of sites have only a single date and over 77% of sites have fewer than three dates
(Figure I 0). This is a significant problem both in establishing occupational chronologies fo r individual sites and in assessing the reliability of age determinations. It is generally accepted that three dates is the minimum necessary for constructing a site chronology (e.g. Smith and
Sharp 1993). In this analysis, sites with single determinations are considered to be accurate, in the absence of other data to assess reliability (see Bird and Frankel 199la:I3). 57
13 � Cl) � 11 G> c. Cl) 9 � , 7 0 � .2l 5 E ::J 3 z
1
0 10 20 30 40 50 60
%of sites
Figure 10. Frequency of sites with different numbers of radiocarbon dates.
' Methods
Three main techniques have been employed in Australia in the analysis of occupational
chronologies based on radiocarbon dates: (i) rates of establishment of new sites (e.g.
Attenbrow 1982; Flood et al. 1987; Hughes and Lampert 1982; Lourandos 1983; Williams
1987, 1988), (ii) estimates of the number of sites occupied in specified periods (e.g. Bird and
Frankel 1991a; Lourandos 1993; Lourandos and Ross 1994), and (iii) counts of radiocarbon
dates over time using a moving average (e.g. Alien and Holdaway 1995; Bird and Frankel
1991a; Holdaway and Porch 1995).
Basal dates have commonly been used in regional syntheses to represent the number of new
sites established through time. This technique is useful fo r examining patterns ofland use by
fo cusing on site creation rather than occupation. Bird and Frankel (1991a:4) have criticised
this method fo r, among other things, "assuming continuity of occupation after firstuse, and
lumping together dates fr om sequences and short-term occupations", thereby denying "the
possibility of demonstrating discontinuity or the reuse of sites after a significant gap in
occupation". These criticisms are difficult to sustain, however, as the method does not assume 58 continuity ofsite use since the technique is explicitly based on site establishment, requiring the use of initial dates for site occupation (and possibly reoccupation after long breaks in occupation, see Smith and Sharp 1993).
In analysing the radiocarbon chronology from westernVictoria and southeast South Australia,
Bird and Frankel (1991a, 1991b) used a technique developed by Rick (1987) involving the calculation of a moving average of the number of radiocarbon dates in sliding 600-year intervals measured every 200 years. Several other Australian investigators have recently adopted this style of analysis (see Alien and Holdaway 1995; Holdaway and Porch 1995; Smith and Sharp
1993). Rick's (1987) method is based on the premise "that the number of dates is related to the magnitude of occupation" (Rick 1987:55, original emphasis) and that "all things being equal, more occupation produces more carbon dates" (Rick 1987:56). Implicitly, this method assumes that radiocarbon samples are selected at random:
He suggests that if radiocarbon determinations were obtained randomly from an unbiased archaeological record, the number of radiocarbon determinations would vary with the total number of person years of human existence in a particular region (Holdaway and Porch 1995:74).
AsHoldaway and Porch (1995:74) noted, several ofthe assumptions of the method are difficult to sustain, as in reality, dates are not selected randomly and the availability of charcoal fo r dating is structured by taphonomic fa ctors affecting both the representation of sites and of charcoal within deposits, as well as sampling and research biases.
The usefulness of this method is, therefore, dependent on a detailed understanding of the complex biases structuring the archaeological date sample. In his examination of the Peruvian preceramic radiocarbon record, Rick (1987) provided a detailed consideration of the effects of research design and the effects of taphonomic processes on the sample. Bird and Frankel 59
(1991a) provide only a superficial discussion ofthe biases inherent in their date sample and justifYthe use of the technique on the basis that it is a "finer scale of counting" and "suggests greater or more complex variability" (Bird and Frankel 1991a:4). Bird and Frankel (I 991a: 1 0) interpret the results of this method to support their argument for "short-term adjustments in settlement patternin response to local conditions" rather than "cumulative directional change".
This seems a strange conclusion given that the original aim ofthe technique was to smooth the curve and broaden the temporal influenceof each date (Rick 1987:61 ).
The dependence of this technique on sample structure is clearly illustrated by an example from the present study. In southeast Queensland the majority of research has been conducted in geologically-recent landscapes to resolve questions oflocalised resource use (e.g. Nolan 1986).
Excavations undertaken at Sandstone Point, for example, were located entirely on a late
Holocene beach-ridge system. Sites located on this ridge could therefore be no older than late
Holocene in age. Fourteen radiocarbon determinations were obtained at this site to investigate intra-site variability. These 14 dates represent 11% ofthe entire date set. Clearly, the nature of research strategies has radically affected the structure of the radiocarbon date sample and severely limits the usefulness of this method in southeast Queensland.
The third major technique used in constructing regional chronologies is simply an estimate of the numberof sites occupied in 500-year periods based on interpolations between radiocarbon determinations. Because of the dearth of sites with termination dates, an assumption is made that sites were occupied to the present if there are no major stratigraphic disconformities to suggest otherwise. This may result in an inflation in the number of recently-occupied sites, but given the coarse scale of analysis, this probably has little effect on broad trends. 60
In the present analysis I have chosen to present the regional date set using all three of the methods reviewed above. The first is an estimate ofthe number of dated sites occupied in 500- year periods (Figure 11), the second utilises Rick's (1987) method of counting the number of dates in600-year periods measured at 200-year intervals (Figure 12) and, finally, the dates are expressed as the number of new sites established through time (Figure 13).
Middens
The dramatic increase in the number of middens in the last 1,000 years is exaggerated by the inclusion of 14 sites which were first occupied only within the last 200 years (Figure 14):
' Double Island Point Site 1, Kabali Road Scatter, Little Sandhills, Northeastern (NE) Moreton
Island 1, 5, 16, 21A and 27, NRS 7, 8 and 10, Seary-Broutha Site, Teewah Beach 5D and
Webber Swamp 100 all fall into this category. As first occupation of all of these sites is confined to the post-contact period, they cannot be assumed to represent pre-contact settlement and subsistence strategies. Despite this, however, removal of these sites from the data set does not significantly alter the apparent trend towards more sites occupied through time (Figure 14).
Two other sites included in this category require special comment. The Wallen Wallen Creek site is considered a coastal site by about 10,000 BP as the sea was in close proximity to the site by this time. King's Bore Sandblow Site 100 is essentially an open artefact scatter in structure although since first use the site has always been in close proximity to the ocean; it is therefore considered as a midden for the purposes of analysis (see McNiven 1990a, 1992d). 61
60
50 "' J!l -+-Midden ·u; 40 0 --a- Rockshelter � " 30 .c __.,.._ Cemetery E " 20 -- Total z 10
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 "' 0 "' 0 "' 0 "' 0 "' 0 "' 0 "' 0 "' 0 "' 0 "' 0 � � N N M M ...... "' "' (!) (!) 1'- 1'- 00 00 (j) (j) 0 � Years cal BP
Figure 11. Estimated number of dated sites occupied in each 500-year period.
45 ' 40 "' 35 J!l ·u; 30 -+-Midden 0 25 -m-- Rockshelter � " .c 20 __.,.._ Cemetery E " 15 -- Total z 10 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 .... 00 N (!) 0 .... 00 N (!) 0 .... 00 N (!) 0 .... 00 N (!) 0 .... 00 � � N N N M M ...... "' "' (!) (!) (!) 1'- 1'- 00 00 00 Years cal BP
Figure 12. Number of sites with central radiocarbon dates falling in each 600-year period, measured at 200-year intervals.
18 16 "' 14 " - ·u; 12 -+-Midden 0 10 -m-- Rockshelter � " 8 .c __.,..__ Cemetery E " 6 z 4 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 "' 0 "' 0 "' 0 "' g "' 0 "' 0 "' 0 "' 0 "' 0 "' g � � N N M M ...... "' "' (!) (!) 1'- 1'- 00 00 (j) (j) 0 � Years cal BP
Figure 13. Number of new sites established in each 500-year period using earliest dates. 62
50 45
1/J 40 ;§! 35 1/J 30 0 -+-Series1 � 25 ., -fll- Series2 .c 20 E :I 15 z 10 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 "' 0 "' 0 "' 0 "' � � N N "' "'
Years cal BP
Figure 14. Estimated number of dated midden sites occupied in each 500-year period. Series 2 excludes sites first occupied in the last 200 years. '
Rockshelters
Rockshelter deposits account for only eight (13.8%) of the total number of dated sites. Only one rockshelter, Bushranger's Cave, is dated to before 5,500 BP. Anincrease in the number of dated rockshelter sites occupied is evident between 5,500 and 3,000 BP, from one to eight dated sites. However, because of the small sample size, the significance of this change is difficult to assess, particularly as there is no further increase after 3, 000 BP. All dated rockshelter sites are occupied before 2,500 BP and over half of them are occupied before 4,000
BP. Until the pattern derived from this small sample is replicated by further excavation, it cannot be assumed that the apparent divergence in patterns of site occupation between mid den and rockshelter sites in the late-Holocene is not simply a product of inadequate sampling. In most of the dated rockshelter sites only dates from single 50 cm x 50 cm pits were made and several of these are non-basal, such as at Brooyar Shelter (McNiven 1988) and Boonah Shelter
(Dr M. Morwood, Department of Archaeology and Palaeoanthropology, The University of
New England, pers. comm., 1995). 63
Cemeteries
Although several Aboriginal burial sites are recorded in southeast Queensland (see Haglund
1976; McNiven in press), only two have been excavated: the Broadbeach Burial Ground
(Haglund 1976) and the Double Island Point site (McNiven 1991c). Only the Broadbeach site has been radiometrically dated, although McNiven (1991c) speculated that the Double Island
Point site may be up to 900 years old.
Discussion
Two major points emerge fr om the occupational chronology presented above. First, coastal occupation pre-dates the close of the last marine transgression around 6,000 BP. Waiters' model explicitly precludes widespread occupation and use of the southeast Queensland coastal lowlands before about 2,000 years ago. The occupation of the coastal lowlands throughout the mid- to late-Holocene (and arguably the early-Holocene, as evidenced by the Wallen Wallen
Creek assemblage) is demonstrated by the continuous record of radiocarbon dated sites.
Although there is no simple relationship between number of sites and number of people (e.g.
Ross 1981), the general increase in the number of dated sites through time suggests an expansion in the general magnitude of occupation in the region.
Second, the evidence does not support Waiters' argument that periods of greater sedimentation around 3,000 BP were related to increases in marine productivity and increases in Aboriginal occupation ofthe coast. If this enrichment of marine resources did occur, it does not coincide with any changes in the archaeological record fr om this time. The occupational chronology demonstrates that significant increases in the number of occupied sites and the rate of site establishment does not occur until around 1,000 BP, some 2,000 years after the proposed enrichment ofMoreton Bay. 64
The overwhelming concentration of sites established in the late-Holocene may be partly explicable in terms of the operation of geomorphological processes, such as erosion and progradation, working to differentially destroy and bury sites. In a recent paper, Hall (in press) discussed the possible impact of sea-level rise on the archaeological record of southeast
Queensland. He suggested that many pre-6,000 BP coastal sites have been destroyed by inundation and erosion associated with transgressive processes. The paucity of sites before
6,000 BP may simply be an artefact of rising sea-levels inundating earlier sites. Assuming the relative stability of sea-levels and coastal landforms since 3,000 BP (see Chapter 3) though, it is difficult to explain the dramatic increase in the number of occupied sites which occurs after '
1,000 BP in terms of differential preservation. The possible role of marine transgressions in fo rming this archaeological pattern is further considered in Chapter Six.
Open inland sites, consisting almost exclusively of stone artefacts, account fo r over 99% of recorded non-coastal sites in the study region (Smith 1994). Unfortunately, these sites generally fo rm unstratified deposits and are rarely associated with material suitable for radiometric dating. Attempts to establish a relative dating system based on artefact morphology and correlation with dated rockshelter assemblages may provide an avenue fo r resolution of this problem (see Hiscock 1988), although recent attempts to implement the system have met with limited success owing to problems of obtaining large sample sizes of complete and unretouched flakes(J. Smith, Department of Anthropology and Sociology, The
University of Queensland, pers. comm., 1995).
From 6,000 BP the estimated number of dated sites occupied in each 500-year period increases.
The relatively coarse-grained temporal fr amework used here may mask short-term variability in the use of particular sites, landscapes or even regions. At I, 000 BP there is an apparent 65 increase in the rate of site occupation, although this may be a product of analytical methods in that sites without a termination date are assumed to have been occupied until the last 500 years; this has the cumulative effect of exaggerating the number of sites in the most recent period.
Regardless of the interpretation of occupation in the last 1,000 years, the data clearly do not support Waiters' proposition that the coastal lowlands were not occupied or only little used priorto the late-Holocene, when unprecedented growth in occupation of coastal sites occurred.
It must be noted here that a major problem in accepting Waiters' interpretation of temporal patterns in the regionis his misrepresentation of the archaeological evidence. The data fo r site establishment presented and discussed by Wafters (1989:218; see Table I) does not derive exclusively from southeast Queensland, but rather from the whole of Queensland (cf. Kelly
1982:39-41). The evidence presented in this chapter is not only consistent with models of continuing coastal occupation throughout the marine transgression (such as those proposed by
Hall l987; Hall and Hiscock 1988; McNiven 199la) but suggests the late-Holocene increases in occupation may be related to the operation of existing coastally-oriented social systems.
The evidence presented here fo r rockshelter occupation demonstrates that the number of occupied shelter sites remains stable during the last 3,000 years, and that the pattern of occupation of shelter sites is not synchronous with changes in the number of occupied coastal sites, which does not alter greatly until after 2,000 BP. Significantly, although most rockshelter sites were first occupied between 6,000 and 3,000 BP, all eight were occupied up until the present.
On the basis of the chronological data presented in this chapter, I argue that a model of permanent occupation of the coastal lowlands throughout the Holocene is more appropriate 66 thanWaiters' (1992c:35) argument fo r "transient or itinerant" occupation of the coast prior to
2,000 BP.
Summary
The occupational chronology presented in this chapter provides unequivocal evidence fo r
Aboriginal presence inthe coastal lowlands throughout the Holocene. Increases are evident inthe number of occupied sites throughout the region from 6, 000 BP. An accelerated rate of midden occupation is evident inthe last 1,000 years, although this may be exaggerated by the analytical techniques employed. The implications of these chronological patterns for Waiters' model and other models of culture change in southeast Queensland are discussed more fully in
Chapter Six. The broad chronological framework fo r southeast Queensland prehistory established here provides the background for an analysis of marine fish remains in coastal sites in the region presented in the next chapter. CHAPTER FIVE
FISHING IN THE ARCHAEOLOGICAL RECORD OF SOUTHEAST QUEENSLAND: Reconsidering the Evidence
All I have tried to show, here and elsewhere since the late 1980s, is that on the evidencewe have, the interpretation I have given is the most parsimonious one that adheres to the data (Waiters 1992c:37, original emphasis).
Introduction
Over the past 30 years, archaeological investigations in southeast Queensland have fo cused almost exclusively on the coastal strip (but see Gillieson 1982; Lilley 1978, 1984). Although little of the region has been systematically surveyed, over 1,500 coastal midden sites have been documented (e.g. Hall and Hiscock 1988; McNiven 1985, 1992a, 1993; Robins 1983, 1984a;
Smith 1992; Stockton 1974), 59 ofwhich have been excavated (Table 4). Of these 59 sites,
24 are said to contain fish remains, although only 22 have been reported in published or thesis fo rm and only 18 have been radiometrically dated (Table 4). In order to critically assess the evidence fo r changes in marine fish production through time, each of these 18 sites, which include those used by Waiters to derive his model, are systematically reviewed in this chapter.
This review of the archaeological sites is fo llowed by a brief discussion which synthesises the evidence fo r marine fishing.
Review of Sites 1: Sites Used by Waiters
The first eight sites reviewed in this section are those employed by Waiters in constructing the model of fishery intensification. Two sites in this group, Toulkerrie and Sandstone Point, have been the subject of further research after Waiters' work. This more recent evidence is also considered in this section.
67 68
Table 4. Summary ofmiddenexcavat ions conducted on the southeast Queensland coast.
Site Year Excavator/s Site Area Sieve Site Fish C-14 Area Excavated Size Reported Remains Dated ' ' (m ) (m ) (mm) Present
Aranarawai Beach Ridge li ? R. Neal ? ? ? N ? y
Aranarawai Creek ? R. Neal ? ? ? N ? ?
Bell's Creek 1972 L. Haglund 1,000 0.7S ? y ? N
Booral Homestead Midden 1989 I. McNiven ? 0.2S S.0/2.0 N ? y
Booral Shell Mound 1989 I. McNiven 120 o.s S.0/2.0 y y y
Bribie Island 9 1992 J. Hall soo,ooo 0.7S 6.013.0 N ? y
' , Bribie Island 67 1993 J. Hall 300,000 1.0 6.013.0 N y N
Broadbeach Burial Ground 196S-8 L. Haglund 28S 228 ? y y y
Brown's Road 1980-1 M. Strong s,ooo 8.2S 6.013.0 N N y
Bundall 1968 L. Haglund 2,SOO 7S ? y N N
Caloundra l 196S D. Tugby s,ooo 1.9 ? y N N
Cameron Point Site 62 1988 I. McNiven ? 0.7S 3.0 y y y
Cameron Creek Site 134 1988 I. McNiven 312 16 3.0 y N y
Canalpin Creek ? R. Neal ? ? ? ? ? ?
Cascade Gardens 1970 L. Haglund 700 so ? y y N
Cooloo la Sand Patch 102 1988 I. McNiven 3,000 10 3.0 y N N
Double Island Point Site 1 1988 I. McNiven 20,000 8 3.0 y y y
First Ridge !9A 1980 R. Robins 4 0.7S 6.0/3.0 y N N
First Ridge 19B 1980 R. Robins 4 0.2S 6.0/3.0 y y y
Freshwater Creek Site 1988 I. McNiven 64 1.7S 3.0 y N N
Hope Island 1986 I. Waiters ? 1.2S 3.0 y N y
Kabali Site 1988 I. McNiven ? 1.7S 3.0 y N y
King's Bore Sandblow 97 1988 I. McNiven 10,000 2.5 3.0 y y y
Leisha Track Site 93 1988 I. McNiven ISO 21 3.0 y N N
Little Sandhills 1980 R. Robins 90 90 6.0/3.0 y y y
Madonna ? R. Neal ? ? ? N ? ?
Maroochy River 2 1988 I. McNiven 2,000 0.2S 3.0 y N y
Maroochy River 4 1988 I. McNiven 4,400 1 3.0 y N y
Minner Dint 1978 J. Hall ? 4 3.17S y y y
New Brisbane Airport 1984 J. Hall ? 0.2S 6.0/3.0 y N y 1987 J. Hall ? 1 6.0/3.0 y y y 1988 J. Hall ? ? 6.0/3.0 N y y ' 1989 J. Hall ? ? 6.0/3.0 N y ? 69
Table 4. Summary ofmidden excavations conducted on the southeast Queensland coast (cont.).
Site Year Excavator/s Site Area Sieve Site Fish C-14 Area Excavated Size Reported Remains Dated (m') (m') (mm) Reported
NEMoretoo Islaod 1 1987 1 Lilley ? 0.5 6.013.0 y N y
" NRS5 1983 R. Neal 182 1.0 3.0 N ? ?
NRS7 1983 R. Neat 754 0.75 3.0 y y ?
NRS 8 1983 R. Neat 149,450 6.75 3.0 y y y
NRS 10 1983 R. Neal 5,280 025 3.0 y y ?
NRS 12 1983 R. Neat 2,000 0.5 3.0 N ? ?
' NRS 14 1983 R. Neat 1,800 1.0 3.0 N ? ?
NRS 17 1983 R. Neat 500 2.0 3.0 N ? ?
NRS 19 1983 R. Neal 18,050 0.25 3.0 N ? y
One-Tree 1980 J. Hall 700 ? 1/4/1/8 N ? y
Polka Point 19614 D. Tugby ? ? ? N ? y ? R. Neal ? ? ? N ? y
Saint.Smith Midden 1994 J. Hall 60,000 0.75 6.013.0 y y y
Sandstone Point 1972 L. Hag!und 25,000 6.0 ? y y y 1984 I. Waiters 25,000 0.5 3.0 y y y 1985 J. Hall 25,000 9.5 3.0 y y y
Seary-Broutha Site 1988 I. McNiven 600 0.5 3.0 y N y
St Helena island 1983 G. Alfredson 180 0.5 2.0 y y y
Teewab Beach 5D 1988 I. McNiven ? 7.5 3.0 y N y
Teewab Beach 18 1988 LMcNiven ? 0.75 3.0 y N N
Teewah Beach 26 1988 I. McNiven 100 1.5 3.0 y y y
Teewah Creek 112 1988 I. McNiven 140 0.5 3.0 y N N
Tin Cao Bay 75B 1988 lMcNiven ? 0.5 3.0 y y y
Tou!kerrie 1978 J. Hatl ? 3.5 8.0/3.0 y y y 1985 I. Waiters ? 2.5 3.0 y y N 1989 J. Hall ? 1.75 3.0 y y y
Watlen Watlen Creek 1985 R. Neal ? 4 ? N y y
Webber Swamp 100 1988 LMcNiven ? 0.5 3.0 y N y
White Cliffs Saodblow 98 1988 I. McNiven 600 ID 3.0 y N N
White Patch Site 1 1973 L. Haglund ? 3.0 ? y y N
White Patch Site 2 1973 L. Haglund ? 3.0 ? y y N
White Patch Site 3 1973 L. Haglund 35 4.5 ? y y y
White Patch Site 4 1973 L. Haglund ? 2.0 ? y N N
. White Patch Site 5 1973 L. Haglund ? 9.0 ? y N N 70
St Helena Island
St Helena is a small island in Moreton Bay located approximately 6 km east of the mouth of the Brisbane River (Figure 7). Alfredson (1983, 1984a, 1984b) excavated two 50 cm x 50 cm pits in an extensive stratified midden in the southwest cornerof the island which is estimated to have a volume in the order of900 m3. To date, only one of the excavated pits has been analysed. About 75% of the excavated material was sieved through 2 mm mesh, while the remainder was mechanically dry-sieved in nested Endicott sieves ranging from 125 11m to 16 mm mesh. Of the residue from the Endicott sieves, only about 75% of material smaller than
4 mm was sorted owing to the limitations of time and identifiability.
The fauna! assemblage was dominated in terms of volume by rock oyster (Saccostrea commercia/is), hairy mussel (Trichomya hirsuta) and mud whelk (Pyrazus ebeninus). Fruit bat (Pteropus poliocephalus) was present throughout, as were fish (Figure 15), with seven species identified. Smaller amounts of mud crab (Scylla serrata) shell, dugong (Dugong dugon), turtle and lizard bone were also recovered. Alfredson (1984b:4) noted that both fish and fruit bat bone constituted a much greater proportion of the fauna! assemblage in the lower half of the deposit, although no attempt was made to quantify the bat bone owing to extensive fr agmentation.
Two shell samples were submitted fo r radiocarbon dating, returning ages of 1,370 ± 60 BP
(Beta-6140) for XU9 and a non-basal 2,240 ± 70 BP (Beta-6141) fo r XU13. Alfredson
(1984a:74) used a depth-age curve to extrapolate a basal age fo r the site of about 2, 700 BP. 71
2 -
4 p - '2 ::l 6 c � 8 g! "' 10 u )( w 12
14
16
0 10 20 30 40 50 60 70 80 90
Number of Identified Specimens (NISP)
Figure 15. Number of identified specimens (NISP) of fish remains by excavation unit, St Helena Island, Test Pit 1 (afterWaiters 1987:343; cf Alfredson 1984a:63). '
Alfredson (1984a:73) interprets material from excavation units 2 to 5 as disturbed 19th-century quarryspoil. Even if we disregard this material there is an obvious trend evident in Figure 15 towards the diminution of the quantity of fish remains towards the surface of the deposit which
Waiters (1987:244) could not attribute to diagenetic loss of bone. As there are only two radiocarbon dates from the sequence it is difficult to construct temporal analytical units fo r this site (see Frankland 1990). Waiters, however, treated the excavation units as temporal analytical units. This point is discussed further in Chapter Six.
Minner Dint
Minner Dint is a discontinuous shell midden exposed in a truncated foredune on the east coast of Moreton Island (Figure 7). Two 1 m x 2 m trenches (A and B) were excavated by Hall
(1980b), yielding a near-basal age of 520 ± 75 BP (I-ll095) on charcoal fr om Trench B (Hall
1980b:l00). The site was excavated according to natural stratification and all sediment was sieved through 1/8 inch (3.175 mm) mesh (Hall l980b:IOO). Waiters (1979:37) incorrectly stated that the material was only sieved through 1/4 inch (6.35 mm) mesh. 72
The deposit was dominated by pipi (Donax deltoides), which comprised over 99% by weight of the entire fauna! assemblage. Small amounts of fish, marine snail (Conuber sp.) and cockle
(Anadara sp.) remains were also found (see Table 5). The vertebrate fauna! assemblage consisted entirely of fish remains, representing eight identified species. A total of 190 fish
NISPs were counted. All 22 g of identified fishbone recovered fr om Trench B occurred in excavation unit 3, 10-15 cm below the surface (Waiters 1979:40).
Table 5. Weights and percentages of fauna! categories from Minner Dint (Hall 1980b:105- 106)'.
Category Trench A 0/o of'J'otal Trench B 0/o of Total Weight(g) Wcight (g)
Fish Bone 1.55 0.00 22.00 0.02
Pipi 67,757 99.98 110,107 99.91
Snail 8.26 0.01 69.50 0.06
Cockle - - 2.30 0.00
' This is a corrected version ofHall's (1980b:105) Table 2 which used the weight ofpipi fr agments rather than total pipi weight fr om his Table 5 (Hall 1980b:106). Waiters (1987:180) similarly used the erroneous version.
Wa llen Wa llen Creek
Wallen Wallen Creek is an open midden site located about 400 m inland fr om the present west coast ofNorth Stradbroke Island (Figure 7). The site represents the earliest evidence of human occupation in southeast Queensland, with a cultural sequence spanning at least the last 20,000 years (Neal and Stock 1986:619). Fauna! remains are restricted to the upper deposit which is dated to about the last 4,000 years. A total of fo ur 1 m squares were excavated, although results from only two of these (WWC-X/Z and WWC-M28-B) have been reported (Neal and
Stock 1986; Waiters 1987). Details of fishremains are only available fo r square WWC-M28-B
(Waiters 1987). 73
Radiocarbon dates reported fo r adjacent squares are 1,520 ± 70 BP (SUA-2236) on shell from
49-56 cm depth, 9,810 ± 130 BP (SUA-2251) on charcoal from 136-141 cm depth, and 17,500
+ 900/-800 BP (SUA-2235) on charcoal from 189-200 cm depth. Neither Neal and Stock
(1986) nor Waiters (1987) related the excavation units to measurements of depth, although
Neal (Cultural Heritage Section, Queensland Department ofEnvironment and Heritage, pers.
comm., 1995) has indicated that fish remains in this sequence are restricted to the last 3,000
years. Significantly, Waiters' (1987) own analysis revealed decreasing rather than increasing
fish discard at the site though time, which he could not attribute to diagenetic loss of bone
(Waiters 1987:244; Figure 16). '
1
"' 3 c:: ;:) c:: 5 .!2-
..� 7 � w 9
11
0 10 20 30 40 50 60 Number of Identified Specimens (NISP)
Figure 16. Number of identified specimens (NISP) of fishremains by excavation unit, Wallen Wallen Creek, M28-B (after Waiters 1987:246).
NRS 7, 8 and 10
These three sites are located on the central east coast ofNorth Stradbroke Island (Figures 1
and 7) and form part ofthe Blue Lake Creek site complex (Neal 1984). 'NRS' is an acronym
for 'NormaRichardson Site' (Neal 1984:29). 74
NRS 7 covers an estimated area of 7 54 m2 and is located on a high frontal dune 225 m south of Blue Lake Creek and 50 m west of Eighteen Mile Swamp (see Figure 7). Neal (1984) excavated three contiguous 50 cm x 50 cm pits at the site. Although culturally-sterile deposits were not reached, the upper units appear to contain the densest cultural material. Pipi (Donax deltoides) comprised over 99% by weight of the total assemblage. The remainder included occasional other shells such as oyster (Saccostrea commercia/is), whelk (Pyrazus ebeninus) and hairy mussel (Trichomya hirsuta) with smaller amounts of stone artefacts and bone. A total NISP of 126 fo r fishremains from the site was counted by Waiters (1987:233), although it is unclear whether these remains derive from only one or from all three pits excavated (see '
Figure 17).
NRS 8 is an extensive midden (149,450 m2) located in a large swale behind high dunes to the south ofBiue Lake Creek (Richardson 1984:28-29). Neat (1984) excavated some 6. 75 m2 at the site in a series of 50 cm grid squares. Waiters (1987:233) reported a total NISP of 61 fo r fish remains derived from fo ur of these excavated pits. Data on the vertical distribution of fish remains are not reported.
NRS 10 covers an area of some 5,280 m2 to the north of the confluence ofBiue Lake Creek and Eighteen Mile Swamp (Richardson 1984:28-29). A single 50 cm x 50 cm pit was excavated. The gross composition of the deposit is similar to other NRS sites, with pipi
(Donax deltoides) dominating the fauna! remains. Walters (1987:233) identified a total of87 fish specimens from this site (but note there is an unexplained discrepancy offour fishNISPs between the total of87 presented in Table A4 [Waiters 1987:342] and the 91 in Table All
[Waiters 1987:349]; see Figure 18). 75
.'!::: 4 c: ::J 8 c: 0 12
�"' 16 " >< w 20
24
0 5 10 15 20 25 Number of Identified Specimens (NISP)
Figure 17. Number ofidentified specimens (NISP) of fishremains by excavation unit, NRS 7 (after Waiters 1987:341).
' := 3 c: ::J c: 6 0 = 9 �> "' " >< 12 w 15
0 5 10 15 20 Number of Identified Specimens (NISP)
Figure 18. Number ofidentified specimens (NISP) offish remains by excavation unit, NRS 10 (afterWaiters 1987:342).
Excavated material from all three sites was sieved through 3 mm mesh. Although no dates are published fo r the controlled excavation pits (Neal 1984), Waiters (1989, 1992a) stated that they are of a similar antiquity to marine shell fr om auger samples taken in this general area which indicate that the entire site complex dates to the last 200 years.
Toulkerrie
Toulkerrie is an extensive midden complex situated on low sand ridges on the southwest coast ofMoreton Island (Figure 7). The site was initially investigated by Hall (1984a) in six 50 cm 76 x 50 cm pits and a 2 m x 1 m trench, some of which were later extended. Excavation was conducted in 5 cm units and material wet-sieved through nested 6 mm and 3 mm screens (Dr
J. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1995; Hall [1984a:64] incorrectly stated that all material was sieved through 8 mm and
3 mm mesh, and Waiters [1987:181] restated Hall's erroneous figure after reporting the correct one in a previous work [Waiters 1979:38]). A single radiocarbon determination on charcoal obtained close to the base of the cultural deposit returned an age of370 ± 75 BP (I-11096).
Fish remainswere fo und to dominate the vertebrate fauna! assemblage (95% by weight), with a minimumof204 individuals from 10 species identified from the coarse screen fraction alone ' fr om trenches 4 and 6 (Hall 1984a). Bream (Acanthopagrus australis), tarwhine
(Rhabdosargus sarba) and mullet (Mugil cephalus) were the dominant fish species.
Aspart of his doctoral research, Waiters (1987:189-193) excavated a furtherten 50 cm x 50 cm pits adjacent to Hall's 'Trench 6' to expand the sample of fish remains. All material was wet-sieved through 3 mm mesh. No radiocarbon deterrninations were obtained although
Waiters assumed contemporaneity with the deposits excavated by Hall (1984a). This assumption is problematic, however, as the excavations conducted by Waiters are located at least 20 m away fromthe pit which yielded the 370 ± 75 BP date. The only clear indication of the antiquity of the deposits excavated by Waiters is a number ofEuropean artefacts in the top
5 cm of Trench 6 (Hall 1984a:68). NlSP fishdata is only presented for two pits - W9A and
W9B (Figures 19 and 20). 77
Ill 2 Fm � !: :::> 4 !: � .�- "' > 6 "' u )( w 8
10 ' 0 10 20 30 40 50 60 70 80
Number of Identified Specimens (NISP)
Figure 19. Number of identified specimens (NISP) of fish remains by excavation unit, Toulkerrie-W9A (afterWaiters 1987:339).
'
- - 2 ·;: :::> 4 !: 0 6 �"' u )( w 8 ·. ' 10 � 0 20 40 60 80 100 120
Number of Identified Specimens (NISP)
Figure 20. Number of identified specimens (NISP) of fish remains by excavation unit, Toulkerrie-W9B (afterWaiters 1987:340).
Hall and Bowen (1989; see also Bowen 1989) excavated a furtherseven 50 cm x 50 cm pits
about 100 m south ofthe original excavation in 1989. Sediment was wet-sieved through 3 mm
screens. Only three of these pits have been analysed and reported. Fish dominate the
vertebrate fauna! assemblage with fo ur species identified in the preliminary analysis: bream
(Acanthopagrus australis), tarwhine (Rhabdosargus sarba), snapper (Chrysoph1ys auratus)
and whiting (Sillago sp.). Significantly, low rates of fish bone deposition are evident fr om the
earliest use of the site some time before 2,290 ± 80 BP (Hall and Bowen 1989), although the 78
highest concentrations of fishremains are located in the upper unit which also contains
abundant shellfish remains. Waiters identified a total 57 MNI offish fo r both trenches and Hall
and Bowen (1989) calculated 20,023 NISP fo r Trench 1 and 20,012 NISP fo r Trench 2 (Figure
21).
25000 20023 20012 20000
' 15000 a. (/) 2 10000
5000 266 472 0 Trench 1 Trench 2 W9A W98
Excavation Square
Figure 21. Comparison of abundance of fish remains at Toulkerrie where NISPdata available (afterHall and Bowen 1989; Waiters 1987).
Figure 21 clearly illustrates significant variation in the density of fish remains across the
Toulkerrie site. This intra-site variation has not been taken into account in interpreting fish remains abundance at the site and should act as a caution against extrapolating from small test
pits to entire sites (cf. Hall and Bowen 1989).
Sandstone Point
Sandstone Point is an extensive midden complex covering an area of some 25,000 m2, located
on a series of beach-ridges on the fo reshores of Deception Bay (Figure 7). The site was
originally excavated by Haglund (1974; see Crooks 1982) who recovered a variety of shellfish
remains with some terrestrial and marine mammal bone. Charcoal samples returned
radiocarbon ages of620 ± 95 BP (SUA-478) and 780 ± 95 BP (SUA-479; Crooks 1982:37).
Waiters (1987:202-209) excavated a further two contiguous 50 cm x 50 cm pits at the site 79
(SSP 1-A and 1-B) adjacent to "the most promising looking exposure" (Waiters 1987:204).
Two charcoal samples fromclose to the base of the midden were submitted for radiocarbon dating and returned values of500 ±50 BP (SUA-2358) and 740 ±50 BP (SUA-2357; Waiters
1987:204). Only the first fo ur excavation units {out of37) from one of the pits (SSP 1-A) were analysed in Waiters' (1987) original study owing to time constraints. Despite the small size of this sample, however, Waiters (1987:209, 233) extrapolated the NISP data to give a figure of some 46,000 NISP/m3• Contraryto Frankland's (1990:62) assertion, fullNISP data fo r SSP-1-
A is presented by Nolan (1986:85) who reports a total of35,883 NISP (Figure 22).
1 111!!!11 3
5
7
9
11
13
15
§ 17 c ,g 19 !!
> 25 Zl 29 31 33 � 35 '31 0 15ll = = Figure 22. Number of identified specimens (NISP) of fish remains by excavation unit, Sandstone Point 1-A (afterNolan 1986:85). 80 Hall(Hall et a/. 1987; see also Nolan 1986) excavated a further 38 (SO cm x SO cm) pits at the site in 1985. Incontrast to the previous excavations which were all on the frontal dune, these squares were distributed over the entire site complex. Analysis of fo ur of these assemblages by Nolan (1986) revealed marked intra-site variability in the distribution ofcultural materials and, in particular, of fishbone (Figure 23). 45000 40000 35883 ' 35000 30000 ll. 25000 Figure 23. Number of identified specimens (NISP) of fishremains in analysed squares from Sandstone Point (Nolan 1986:85-88). Summary of Wa iters ' Sites Of the eight sites Waiters employs in his study, three (NRS 7, 8 and 1 0) are dated to the post- contact period and should not necessarily be considered as representative of prehistoric Aboriginal lifeways, two {WallenWallen Creek and St Helena Island) actually exhibit decreases in the rate of fish discard over the last 1,000 years which Waiters (1987) cannot attribute to taphonomic factors, and one site (Minner Dint) has virtually all recovered fish remains from a single excavation unit. Only two sites (Sandstone Point and Toulkerrie) unequivocally support the model, demonstrating trends towards increased deposition of fish remains over the last thousand years. 81 Review of Sites 2: Other Sites This second section systematically reviews the other 12 sites in the region which have both reported fish remains and radiocarbon dates. This review provides a wider database for consideration of the validity ofWalters mod�!. Booral Sh ellMo und The Booral Shell Mound is part of an extensive midden site complex located on the western shores of Great Sandy Strait in the north of the study region (Figure 7). A 50 cm x 100 cm trench was excavated at the highest point of the mound in contiguous 50 cm x 50 cm pits. The site was excavated in 25-litre arbitrary units within stratigraphic units. Excavated sediments were sieved using nested 5 mm and 2 mm screens. Frankland (1990) analysed the total assemblage from one of these pits (BSM-A). The site was occupied shortly before 3,000 BP, after which time large numbers of fishremains from a variety of species were evident, although Frankland (1990) demonstrated decreasing rates of fish bone discard through time (Figure 24). XU Years BP 1 980 1750 10 2480 2660 2790 20 2950 28 250 200 150 100 50 0 50 100 150 200 250 f"''*'l NISPIXU I < Broadbeach Burial Ground The Broadbeach site is a large burial complex located near the Nerang River at the Gold Coast (Figure 7). Haglund (1976; see also Haglund-Calley 1968) conducted extensive excavations at the site, covering a total of 228 m2• The fo cus of the excavation was the recovery of Aboriginal skeletal material for comparative anatomical studies (e.g. Murphy 1991). Consequently, little attention was given to the recovery and analysis ofother cultural materials such as stone artefacts and non-human faunal remains. Some of the excavated faunal material was examined by Bartholomai (1976:90-93). Fish remains were apparently recovered from throughout the deposit and all diagnostic elements were identified as belonging to the Family Sparidae. Only 34 individual fish bone elements were identified, consisting mainly of vertebrae. Since the site apparently had a specialised purpose, the context of fish production and consumption is difficultto assess. Cameron Point Site 62 This site is a large shell midden exposed in an erosion fuce on the eastern shores of Tin Can Bay (Figure 7). McNiven (1990a:200-209, 1991b) excavated three 50 cm x 50 cm pits immediately behind the eroding section, only one of which (SQB) has been analysed and reported. There is a marked vertical discontinuity in the distribution of cultural material with faunal remains restricted to the top 17 cm of the deposit which dates to the last 200 years. All stone artefactual material lies below this upper level. McNiven identified0.15 g of fish bone from the upper level (excavation units 2, 4 and 5). An otolith (0. 1 0 g) from the summer whiting (Sillago ciliata) was the only diagnostic fragment recovered. McNiven (1990a:206) cautioned that the fish remains may have been deposited by 83 dingoes rather than by humans as "dingo fa eces exhibiting fishbones have been observed on the surface adjacent to the site". Double Island Point Site 1 Double Island Point Site 1 is a large shell and stone artefact scatter in an active sandblow 500 m south ofDouble Island Point (Figure 7). McNiven (1990a:l30-141, 1991c) excavated a total area of6 m2 at the site, including two Aboriginal burials. Fish remains accounted for 0.31 g ofthe non-human fauna! assemblage which included a single MNI oftarwhine (Rhabdosargus sarba). First Ridge 19B This site is one of a number of small discrete shell heaps located in the northeastern section of Moreton lsland (Figure7). Robins (1983, 1984b) estimated that site 19B covered an area of 4m2 and he excavated a 50 cm x 50 cm pit in two 10 cm-deep excavation units. All sediment was sieved through 6 mm and 3 mm screens. The recovered assemblage consisted almost entirely ofpipi (Donax de ltoides), with some charcoal and a single whiting otolith. A C-14 date of 1150 ± 70 BP (Beta-1946) was obtained on a shell sample giving a calibrated age of 680 BP. King 's Bore Sandblow 97 King's Bore is a massive site located in a large active sandblow about 250 m inland from Teewah Beach (Figure 7). McNiven (1985, 1990a, 1992d) recorded a large stone artefact scatter covering some 10,000 m2 within the sandblow. A total of 111 m2 was excavated to a depth of5 cm yielding 5,265 g of stone artefacts and 1.22 g of shellfish and fish remains with some fragments identified as pipi (Donax de ltoides) and summer whiting (Sillago ciliata). A 84 further 2.5 m2 was subjected to controlled excavation to investigate in situ deposits, resulting in the recovery of 137 g of stone artefacts and a further 1.37 g of pipi fragments and unidentified fish bone and scale fragments. McNiven (1990a, 1990b, 1992d) argued that all of the recovered fauna! remains derived from carnivore activity given their good preservation and the presence of flesh on some shell and bone fragments. Fish bone was also noted as a component of a disaggregating dingo scat on the site. Little Sandhills The Little Sandhills site is an extensive shell midden and stone artefact scatter located in active sand dunes on southern Moreton Island (Figure 7). Robins (1983, 1984b) conducted excavations over the entire surface of the site, some 90 m2, to a depth of 5 cm. Excavations included the total recovery of 10 small shell piles. Excavated sediment was sieved through 6 mm and 3 mm screens. The combined assemblage consisted of shellfish remains (mostly Donax deltoides) and stone artefacts, as well as small amounts of fish bone and other bone fragments tentatively identified as dugong (Dugong dugon). A radiocarbon date obtained on shell returned a result of 102.1 ± 7% BP modem (Beta-1945), indicating post-contact occupation. Fish remains included 18 diagnostic skeletal elements from Sparidae and whiting (Sillago sp.) with some mullet scales. New Brisbane Airp ort The New Brisbane Airport site is located on a palaeo-shoreline in a prograded section of the Brisbane River delta, some 5 !an from the modem shoreline (Figure 7). The site was excavated by Hall (Hall and Lilley 1987) on fo ur separate occasions, yielding evidence of human occupation before 5,000 BP (Dr J. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1991). Waiters (1992c:35) noted that only a few 85 fragments offish bone were recovered and argues that these have not been demonstrated to be cultural. Hall (1990:180) reported 'significant' numbers of fish remains dating to the mid Holocene and recent analysis has revealed more fragmented fish and shellfishremains encased in the ironstone conglomerate matrix of the lower excavation units, although quantification of the remains has not been attempted (DrJ. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1995). Saint-Smith Midden The Saint-Smith site is an extensive shell midden deposit on the shores of Deception Bay near the Caboolture River (Figure7). Inaddition to an extensive program of auguring to definethe subsurface extent of the deposit, Hall (1995) excavated three 50 cm x 50 cm test pits. Excavation proceeded in 1 0-litre units and sediment was sieved through 6 mm and 3 mm screens. Hall (1995:13) noted significant disturbance in one of these pits (TPI), and excluded it from his analysis. Radiocarb on dates on charcoal indicate that the site spans the last 1,000 years, with a near-basal date of960 ± 60 BP (Wk:-3423)fo r TP2. The cultural assemblage was dominated by shellfish remains with lesser quantities of stone artefacts and bone. Fish remains were recovered from allpits and are associated with shell material. A total of7.5 g of fish bone was recovered from TP2 and 33 g fromTP3. Te ewah Beach Site 26 Teewah Beach Site 26 is an extensive midden site 8 km north of the Noosa River (Figure 7). McNiven (1990a:96-1 09) excavated a 1.5 m2 pit using arbitrary 5 cm units within stratigraphic units. All sediments were sieved through 3 mm mesh. Five radiocarbon dates were obtained on charcoal, ranging from 4,780 ± 80 BP (Beta-25512) to 340 ± 70 BP (Beta-30401). The combined assemblage was dominated by shellfish remains (and Donaxdeltoides in particular) 86 with lesser amounts ofstone artefacts, charcoal and some fish bones. A total of 1.73 g offish bone yielded an MNI of one. Four posterior molars identified as tarwhine (Rhabdosargus sarba) were the only diagnostic fragments. Allfish remains are associated with the upper shell " deposit which dates to the last 1,000 years. Tin Can BaySite 75B This site is exposed in an erosion face on the west coast of Tin Can Bay (Figure 7). McNiven (1990a:l91-200, 199lb) excavated two contiguous 50 cm x 50 cm pits. The base of the midden is dated on shell at 700 ± 70 BP (Beta-19421), producing a calibrated age of313 BP (probably late 18th-century and not modem as McNiven [1990a:l94] inferred). Cultural material consisted primarily of shellfish remains, stone artefacts and a minute quantity of fish bone. The fishremains consisted of a total ofO.Ol g ofunbumt bone and McNiven suggested, once again, that dingoes and not humans were responsible fo r deposition of the material. White Patch Site 3 This site is one of a number of sites excavated by Haglund (see Crooks 1982:60-74) on the southwestern coast ofBribie Island (see Figure 7). Haglund excavated some 4.5 m2 at Site 3, yielding a large fauna! assemblage dominated by whelk (Pyrazus ebeninus), cockle (A nadara sp.), oyster (Saccostrea commercia/is) and mussel (Trichomya hirsuta). Although many of the fish remains were not identified due to fragmentation, a total of 35 skeletal elements were identified to species, mainly bream (Acanthopagrus australis) and tarwhine (Rhabdosargus sarba), with a single element belonging to the fo rk-tailed catfish (Arius sp.). Crooks (1982:64) noted a concentration of fishbone in square B52 representing about 14 bream, 10-15 cm below the surface. Two dates were obtained on charcoal from square C50, returning ages of 450 ± 95 BP (SUA-480) and 670 ± 95 BP (SUA-481). 87 Non-Coastal Sites Fish remains, including sea mullet (Mugil cephalus) were discarded at Platypus Rockshelter as early as 4,200 BP (Novella 1989). Bushranger's Cave has fish remains in basal levels dated to around 10,000 BP (DrJ. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1995). Remains of perch have also been recovered from Gatton Shelter in levels dated to before 3,000 BP (Morwood 1986). Summary of Other Sites What emerges fr om the fo regoing site-by-site description of archaeological fish remains in southeast Queensland is the generally small quantity of fishremains at any site. Of the sites reviewed in this section, only the Booral Shell Mound could be said to have evidence of significant fish discard but this may be related to the use of smaller sieve mesh sizes (2 mm). From this brief overview it is clear that the distribution of fishremains in archaeological sites in southeast Queensland varies considerably across space and through time. Discussion My review reveals fo ur major impediments to accepting the model of late-Holocene fishery development proposed by Waiters. First, some basic predictions of the model are not supported by the quantitative data. Many sites in the region dating to the late-Holocene do not contain evidence of fishing. Of the five sites which do exhibit significant fish discard, three (the Booral Shell Mound, Wallen Wallen Creek and St Helena Island) actually demonstrate de creasing rather than increasing rates of fish bone discard through time (Table 6). Second, the small samples employed by Waiters raise significant problems fo r the examination of regional production strategies over time. The small size of the sample combined with the effect of inadequate recovery procedures considerably reduces confidence in the available data and 88 Table 6. Summary ofregional fish bone data. Site Date First NISP MNI Weight (g) Comments Fishing (cal BP) - - St Helena Island 1834 576 " decreasing discard MinnerDint 510 190 10 23.55 22 gin one XU Wallen Wallen Creek not dated 296 - - assemblage not dated; decreasing discard NRS7 moden1 126 - - modern NRS 8 moden1 61 - - modern ' NRS 10 modern 87 - - modem Toulken·ie 2224 40,050 - - major fishdiscard Sandstone Point 2224 37,754 - - major fish discard Booral Shell Mound 3063 1470 36 - discard before 3,000 BP; decreasing discard Broadbeach Burial 1,200 34 - - production possibly Ground associated with non- subsistence activities Carneron Point Site 62 264 - I 0.15 possible non-human origin, minimal Double Island Point not dated - I 0.31 undated; production possibly associated with non-subsistence activities First Ridge 19B 680 I I - minimal King's Bore Sandblow - - - <1.37 probable non-human otigin, minimal Little Sandhills moden1 18 - - modem; minimal New Brisbane Airport 5,000 few - - fish bonepresent by elements 5,000 BP; minimal Saint-Smith Midden 1,000 - - 40.5 significant discard, confmed last 1,000 years Teewah Beach Site 26 788 - 1 1.73 minimal Tin Can Bay Site 75B 313 - 1 0.01 possible non-human origin, minimal White Patch Site 3 640 - - - minimal 89 thus the reliability ofthe model. Third, because half ofthe fauna! assemblages analysed by Waiters are undated, any calculation of discard rates through time is problematic at best. Only two sites (Sandstone Point and Toulkerrie) give unequivocal support to the model, demonstrating trends towards increased deposition of fishremains over the last thousand years. In short, only 25% of his sample can be legitimately employed in support of his model. One must question the regional applicability of this two-site sample. Ethnohistoric literature characterises the Toorbul Point area (where Sandstone Point is situated) as an important Aboriginal gathering site, at least in the contact period (see Nolan 1986). Thus, major marine fishery production at this site, at least in the recent past, may have taken place within a very specific inter-group production context; that is, it may well be atypical. At Toulkerrie, the sample excavated by Waiters was not dated and the assumption of contemporaneity with nearby occupation dated by Hall (1984a) to about 400 BP should be treated with caution as the later research conducted by Hall and Bowen (1989) revealed significant differences in the timing of initial occupation across the site. Recent excavations at the site revealed an occupational sequence spanning the last 2, 000 years, indicating that people occupied this locality for some time (see Bowen 1989; Hall and Bowen 1989). Summary The regional data relating to the representation of fishremains in coastal sites is summarised in Table 6. Clearly the evidence for late-Holocene intensification of marine fishing based on Waiters' own eight-site sample is equivocal at best. The overall picture presented by these 20 sites is one of considerable variability and the complete absence of any definite directional trend. A number of fa ctors may have influenced the representation of fishremains in southeast 90 Queensland sites including the location of consumption, discard habits, taphonomic fa ctors, recovery techniques and quantitative methods (see Chapter Six). In sum, on the basis of this evidence it is difficult to make definitive statements about prehistoric marine fishing at a regional level. Inthis chapter I synthesised the considerable body of data related to prehistoric marine fishing in southeast Queensland. Many problems and limitations became apparent in this undertaking. Results from many sites areunpublished or do not contain detailed analyses of fauna! remains. Because of variations in recovery techniques and abundance measures, combining the various bodies of data into a single data set is problematic. This lack of comparability between assemblages is a major obstacle in deriving regional patterns in fishingbehaviour. CHAPTER SIX DISCUSSION AND IMPLICATIONS: A Review of the Evidence and Related Issues Regional syntheses are usually normative in that they emphasize some "facts" while ignoring others ... The results of normative thinking are the distortion of · an archaeological record of variation in adaptive responses into a record of homogeneous response (Claassen 1991:249). Introduction The data and discussions presented in the previous three chapters highlight significant problems in the interpretations and assumptions which underpin the model proposed by Waiters as well as other models of prehistoric cultural change in southeast Queensland (Hall 1982, 1987, 1990; Hall and Hiscock 1988; McNiven 1991a). These problems relate both to the evidence itself and to other, largely unconsidered, fa ctors which have been involved in structuring that evidence. In this chapter, I will briefly review the evidence fo r each of the three central elements of Waiters' model: (i) the key concepts of landscape marginality and of a time-Jag between sea- level stabilisation and Aboriginal occupation of the coast; (ii) the timing of the proposed change in the availability and exploitation of marine resources; and (iii) the nature of that change, specificallythe evidence fo r intensification of marine fishing. I will then discuss several of the fa ctors which have played a potentially significant role in structuring this database, but which have yet to be adequately taken into account in interpretations of southeast Queensland prehistory. In evaluating the evidence, key issues such as regional geomorphology and its effects on site survival, as well as smaller-scale taphonomic processes and their effects on the survival of fish bone in particular, need to be taken into consideration. There is also a second set of fa ctors which structure our perceptions of the 91 92 archaeological evidence. These include research designs, excavation strategies, recovery techniques and analytical methods, which together influence the structure of the data sample and the way that it is represented. A Review of the Evidence Basically, Waiters' model is dependent on three central tenets: 1. That the southeast Queensland coastal lowlands constituted a marginal landscape fo r ' human occupation until transformed by environmental change, and that this change only occurred after a 4,000 year time-lag after sea-level stabilisation. 2. That this environmental transformation provided a basis fo r Aboriginal occupation of the coast and exploitation of its now abundant marine resources aft er 2, 000 BP, as evidenced by a dramatic increase in the number of coastal sites at this time. 3. That this exploitation took the fo rm of intensive marine fish production, as evidenced by increasing quantities of fish remains in coastal sites fr om this time. If the evidence fa ils to support any one of these elements, Waiters' model of late-Holocene intensification of marine fishing cannot be sustained. Also, it must be noted that these elements are likewise fundamental to other interpretations of southeast Queensland prehistory (Hall 1982, 1987, 1990; Hall and Hiscock 1988; McNiven 1991a; see Chapter Two). Thus, their fa ilure to support Waiters' interpretation constitutes a call fo r much closer attention to the investigation oflate-Holocene change in the region. I will now brieflyexamine each of these elements, before moving on to discuss some of the general processes responsible for the structure of the evidence and its interpretation. 93 Time-Lags and Marg inality Waiters proposed that the apparently late use of southeast Queensland coastal landscapes was related to their earlier marginality. He posited that after sea-level stabilisation, changes occurred in sedimentation patterns in ano around Moreton Bay, culminating in the establishment of abundant sea-grass and mangrove communities and associated fauna! suites. This alteration in marine productivity is argued to have occurred aftera time-Jag of some 4,000 years, facilitating Aboriginal occupation of the coast at around 2, 000 BP. The review of evidence undertaken in Chapter Three, however, demonstrated that this marginality model could not be sustained on the basis of current environmental, ethnohistoric or archaeological evidence. The argument was largely constructed on the perceived paucity of the terrestrial resource base and the inability of shellfish communities to sustain exploitation. While ecological studies have characterised the coastal lowlands as having a depauperate terrestrial fa una! resource base, they have simultaneously emphasised the diversity and 'patchy' distribution of these resources (e.g. Dwyer et al. 1979a, 1979b; Kikkawa 1975). Similarly, some shellfish species of major importance in Aboriginal economies have been shown to be resilient to exploitation (Catterall and Poiner 1984, 1987; Luebbers 1994). Thus, there is no sound basis fo r the argument that the coastal lowlands were marginal in such a way as to exclude the possibility of their earlier occupation. It appears that a diverse range of fauna! resources, including terrestrial mammals and marine shellfish, were available in the area and that although their abundance may have fluctuated in response to environmental changes, at no time would the coastal lowlands have been as 'depauperate' as Waiters suggests. The notion of a 4,000 year time-Jag between sea-level stabilisation and Aboriginal occupation of the coast is also problematic. 94 Although the major period of increased sedimentation and shoreline progradation in Moreton Bay is dated to around 3,000 BP, there is no change in the estimated number of sites or in the rate of site creation until almost 2, 000 years later. I can only infer from these data that the initial occupation ofthe coast and the marked increase in occupation in the last 1,000 years cannot be related to ecological changes resulting from sea-level stabilisation or from later changes in sedimentation. Whether these later changes are attributable to anthropogenic cause (as argued by Hall [1990] and Waiters [1989, 1992a]) or other processes, or a combination of the two, is a moot point at this stage. The increase in occupation after 1,000 BP correlates neither with major environmental changes argued by Waiters to occur around 3, 000 BP, nor with the establishment or development of shellfish communities which are in evidence before the close of the marine transgression (see Chapter Three). As Mclnnes (1988:78, 81) noted: Rather than assume that some fo rm of "time-Jag" is responsible, the lack of correlation is best interpreted as a lack of direct causal relation ... the fa ilure of simple environmental models to explain the patterns in the archaeological record indicates the need fo r a logic linking the environmental changes with changes in human behaviour. Chronology The second key element of Waiters' model is that Aboriginal occupation of the coastal lowlands occurred around 2, 000 BP. He maintains that this is evidenced by the absence of coastal sites before this time and the exponential growth in the number of coastal sites occupied after this time. My own analysis of the chronological trends in Chapter Four, however, does not support this interpretation. 95 Three important patterns emerge from an analysis ofthe occupational chronology of the region. First, the coastal lowlands were clearly occupied from the time of sea-level stabilisation. Archaeological evidence fromWallen Wallen Creek (Neal and Stock 1986), Teewah Beach Site 26 (McNiven 1991a), Hope Island (Waiters et al. 1987) and the New Brisbane Airport site (Dr J. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1995; Hall and Lilley 1987) indicate that people were resident on the coast around the time of sea-level stabilisation (ea. 6,000 BP). Secondly, the number of coastal sites increases gradually, but consistently, from 6,000 BP to 1,000 BP after which an exponential increase is apparent. Significantly, increases in the estimated number of sites in the last 1,000 years coincide with increased rates of establishment of sites, possibly indicating a reordering of land-use involving more locations on the landscape. Thirdly, few coastal sites dating to the early and rnid-Holocene are recorded in southeast Queensland. The apparent increasing trend in the number of occupied sites through time commencing at 6, 000 BP is suggestive of the potential role of marine transgression in structuring the archaeclogical sample by inundating previous coastal sites (Hall in press). This truncation of the archaeological record creates a bias towards the representation of post-transgressive occupation (Head 1986). The later change at around 1,000 BP, however, is more difficultto ascribe to taphonornic fa ctors and may instead signal some kind of change in the use of coastal landscapes. Whether or not this relates to changes in the use of marine resources is yet to be determined. 96 In tensified Fish Production Waiters' model has its basis in his observation that rates of fishremains discarded on coastal sites increase exponentially in the late-Holocene. My review of all reported sites which yielded fish bone in dated contexts revealed that, generally, the archaeological evidence for fishing throughout the region is extremely limited, and that significant numbers of fish remains have only been recovered from three sites in the region: Sandstone Point, Toulkerrie and Booral. In reality, Waiters' original interpretation of regional change in fish exploitation was based almost entirely on the results fr om the analysis of the Sandstone Point assemblage. AsGaughwin and Fullagar (1995:47) noted, models of"prehistoric change based on atypical sites must be treated with caution". The Sandstone Point site is clearly not typical of coastal archaeological deposits in southeast Queensland. Three main fa ctors set the fo redune deposits at Sandstone Point apart: (i) the cultural deposits on the fo redune accumulated over a relatively shortperiod of time, (ii) there appear to be relatively low rates of mechanical damage to bone elements compared to other sites in the region (Waiters 1987:248) and (iii) quantities of scales were recovered fr om the excavation suggesting favourable preservation conditions (Waiters 1987:235). However, further analysis of the Sandstone Point material by Waiters (see Hiscock and Waiters 1994) may clarify the apparently anomalous status of the site. While Waiters' model emphasises the primacy of fish production as abasis fo r late-Holocene colonisation of the coastal lowlands and the subsequent elaboration of social complexity, both the archaeological and ethnohistorical evidence indicates considerable variability in resource use. Waiters has consistently omitted consideration of several important studies (e.g. Collier and Hobson 1987) which suggest that although marine resources were certainly important, the economic system is perhaps more accurately characterised as a 'mixed economy'. 97 Ethnohistoric records suggest that coastal peoples utilised a broad range ofmarine, littoral and terrestrial resources. Also, while fi shing appears to have been important in recent times, I suggest that the role of marine fish resources has been overemphasised by recorder bias and that important but perhaps less visible production activities associated with terrestrial resources have been underemphasised. This possibility would seem to be confirmed by analyses of skeletal material fr om the Broadbeach Burial Ground which indicate that marine resources probably accounted for less than half of the diet (Collier and Hobson 1987; Hobson and Collier 1984). The review of the evidence fo r marine fishing in southeast Queensland presented in Chapter Five demonstrated that the spatial and temporal distribution of fish remains in coastal sites in the region is highly variable, and not simply characterised by increasing discard of fish remains through time. In fa ct, most coastal sites dating to the late-Holocene do not exhibit fishremains and at others, the abundance of fish remains actually appears to decrease through time. However, it is difficult to confidentlydefine regional patterns in the fishremains evidence at this stage as a combination offactors, such as discard behaviour, taphonomic fa ctors, recovery procedures and quantification techniques, may have a significant impact on the comparability of assemblages and the validity of inter-site comparisons. Related Issues The evidence discussed above cannot be considered without reference to the factors which have influenced its structure. These fa ctors fa ll into two categories. There are those factors which structure the physical evidence itself, and those which influencethe way we view that evidence, primarily sampling problems. In this section, I briefly address six major issues which have influenced the evidence fo r the Holocene occupation of southeast Queensland. The first 98 two are issues relating to large-scale and small-scale taphonomic factors which have affected the survival of the physical evidence, that is, the sites and their contents. The other fo ur issues relate to some of the biases introduced by archaeological research methodologies: research designs, excavation strategies, recovery techniques, and analytical methods. Geomorphology Geomorphological processes (especially the possible impact of sea-level change and erosion on-the representation of archaeological materials) have fe atured prominently in discussions of the southeast Queensland coastal archaeological record (e.g. David 1994; Hall and Lilley 1987). The problem is particularly acute in southeast Queensland because of the absence of major rock fo rmations close to the coast and the dominance of sandy sediments. This has resulted in complex patterns of coastal sedimentation and landscape development operating over both short and long time spans (see Chapter Three). As David (1994:325) notes, in southeast Queensland, "the influence of changing sea levels on site survival has to be considered before occupational trends can be inferred fr om the archaeological record". Sea-level fluctuations, coastal eroswn, cyclonic and storm surge activity and coastal progradation have resulted in differential destruction ofthe coastal archaeological record (e.g. Bird 1992, 1995; Head 1983, 1986, 1987; Rowland 1983, 1989). In the study region, approximately 20,000 km2 of land has been submerged by transgressing seas over the last 18,000 years (see Chapter Three). This fact has obvious implications fo r the representation of coastal archaeological sites pre-dating the close of the transgression (ea. 6,000 BP). Thus, the archaeological record is expected to be truncated and biased towards later Holocene adaptations. 99 Afterreviewing evidence fo r Pleistocene archaeological sites on the Australian continent, Smith and Sharp (1993 :49) noted the "presence of sites on or near the Pleistocene coast, wherever the continental shelf is steep enough to allow their preservation". This trend is also apparent in international studies (e.g. Waselkov 1987). All known Pleistocene and early-Holocene sites in Australia which exhibit use of coastal resources are located in rockshelters situated near palaeoshorelines (e.g. Barker 1989, 1991, 1993; Morse 1988, 1993a, 1993b, Veth 1993b). Thus, in southeast Queensland where the continental shelf is relatively wide, has a gentle slope and no near-coastal rock formations conducive to shelter fo rmation, it is perhaps not surprising that archaeological evidence is lacking fo r coastal occupation prior to the mid-Holocene sea- level stabilisation. This evidence does not demonstrate that people did not occupy the coast prior to this time, as Waiters has implied. Evidence ofPieistocene occupation of the region from Wallen Wallen Creek, Bushranger's Cave, and possibly the Talgai Skull site (see Chapter Four fo r a discussion of dating problems), combined with evidence of coastal occupation synchronous with sea-level stabilisation at Teewah Beach 26, New Brisbane Airport, Hope Island and Wallen Wallen Creek, provides compelling evidence fo r the presence ofpopulations fo llowing the transgressive coastline westwards (see Hall and Hiscock 1988; McNiven 1991a). The simplest explanation fo r the near-absence offish remains and sites before 3,000 BP is that known coastal sites were fo rmed in landscapes which only date to the late-Holocene. Detailed geomorphological studies from the entire region are now required to establish a framework fo r evaluating the structure of this archaeological evidence. In sum, given that all the assemblages used by Walters in his analysis were derived fr om geomorphological contexts which were late- Holocene in age, it is perhaps not surprising that components of these assemblages only date to the Iate-Holocene. 100 Taphonomy A further complication is the potential role of taphonomic fa ctors in structuring archaeological assemblages inthe region. Waiters (1987:256) has repeatedly argued that the "methodological controls" employed inhis investigations "imply"that the patterns of distribution and abundance ... adequately reflected composition of the prehistoric catches". It is important to develop an understanding of how fish bone becomes incorporated into archaeological deposits. The only archaeological marine fish bone samples currently available are from shell midden contexts. This is unlikely to represent the range of fish production and consumption activities recorded in the ethnohistoric literature. For example, several accounts note the immediate consumption of fish at the site of capture on the beach probably below tidal range (see McNiven 1991a). Like all fauna! remains, fishbone is susceptible to a wide variety ofpossible post-depositional modification(see Lyman 1994a). Colley (1990:215), fo r example, noted that fish remains are "particularly vulnerable to the effects of differential preservation because some fish bones are extremely fr agile while others are robust". It is ironic that despite publishing a paper on the role of dogs in the representation of fauna! remains in the archaeological record (Waiters 1984), Waiters does not consider the potential role of domestic dogs in modifYing fish bone accumulations in his southeast Queensland study. His own and other ethnoarchaeological research has established that the remains of small-bodied animals such as fish are especially vulnerable to modificationby dogs, with documented losses of up to 97% of skeletal elements (e.g. Hudson 1993; Kent 1981, 1993; Lyon 1970; Waiters 1984, 1985). Early ethnohistoric accounts (e.g. Flinders 1814; Uniacke in Steele 1972:59) document the presence of numerous dogs at habitation sites and on visits to islands such as Coochie Mudlo (Flinders 1814). There is evidence of scavenging by dogs in the study region by 4,000 BP (see 101 Novella 1989:73) and recent studies of dingo scavenging behaviour in the area demonstrate that fish comprise a significant part of the diet (Twyford 1994, 1995). Thus, domestic dogs may have been a significant agent in structuring the representation of fish remains in archaeological assemblages. Other taphonomic fa ctors influencing the survival of fishbone may include burning and post depositional diagenesis. In fa ct, the two may be related in that the high proportions of burnt bone recovered in the region (Waiters 1987:240) could indicate the loss ofunburnt bone. It is unlikely that chemical attrition alone could account fo r the distribution of fishremains since the majority of excavated sites have a fairly high alkalinity of pH 7.0 to 8.0. Two sites, Cameron Point Site 62 and Tin Can Bay Site 75B, have recorded pH as low as 4.0 and 4.5 respectively which would not be conducive to survival offish bone although at both these sites other fau na! remains (such as shellfish) did not appear to be adversely affected by these conditions (see McNiven 1990a). Further studies are required to define the various taphonomic processes affecting the representation of fishremains in the archaeological record. Research Designs A major limitation in interpretations of coastal occupation in southeast Queensland stems from a poor understanding of the relationships between coastal and inland sites. In part, this has resulted fr om patterns of research fo cused on shell-bearing archaeological sites in close proximity to the present shoreline. The distribution of excavated sites either on the extreme coastal fringe or in inland rockshelters may have seriously biased our knowledge of the variability in site structure and thus representations of the variability of resource use and settlement patterns in the area. 102 The sample of dated archaeological sites in southeast Queensland is clearly dominated by sites located within 500 m of the shoreline (Figure 25). All dated inland rockshelter sites are over 50 km from the shoreline. Only nine sites fa ll between these two extremes: the Broadbeach Burial Ground, Brown's Road, First Ridge, Hope Island, Kabali Road Scatter, Little Sandhills, New Brisbane Airport, Seary-Broutha Site and Webber Swamp Site 100. Both Hope Island and the New Brisbane Airport site are located adj acent to tidal estuaries. The Little Sandhills, Kabali Road Scatter, Webber Swamp and Seary-Broutha sites are all modern and may not reflect pre-European settlement and subsistence patterns. The Broadbeach Burial Ground site appears to have been used fo r a specialised purpose, thus making the context of resource production difficult to determine. This leaves only two sites, Brown's Road and First Ridge, which may be taken to unequivocally represent pre-European Aboriginal settlement and subsistence away from the immediate coastal fringe. The First Ridge site on Moreton Island is only a small site which consists almost entirely ofpipi (Donax de ltoides), and thus has limited potential to inform us about subsistence behaviour away from the coast (see Robins 1983; 1984b). The Brown's Road site has not been reported, although limited excavation and extensive surface collection have yielded numerous bevelled-edged stone artefacts associated with plant-food processing (Gillieson and Hall 1982; Higgins 1988; McNiven 1992b; Richter 1994; Robertson 1994). Significantly, preliminary identification of fauna! materials has revealed the remains of several large macro pods (Dr J. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1995). The Brown's Road site may, therefore, represent a terrestrially-focused component of coastally-oriented settlement-subsistence systems in the area. 103 45 40 35 "' .e 30 lii - 0 25 � ., .a 20 E " 15 z 10 5 0 > 0.5 0.5 -2.0 2. 0-5.0 50 > ' Distance (km) Figure 25. Estimated distance of dated sites fr om present shoreline (n=58). Interestingly, all six coastal sites in the region which date to before 3,000 BP (Booral Shell Mound, Teewah Beach Site 26, Bribie Island Site 9, Polka Point, Wallen Wallen Creek, New Brisbane Airport, King's Bore Sandblow Site 97) also contain more recent cultural deposits dating to the late-Holocene. Perhaps further and more thorough excavation and systematic dating of basal deposits at apparently recent sites may result in the identification of further mid- Holocene deposits (see Breschini and Haversat [1991] fo r a similar phenomenon in coastal California). Excavations at several coastal sites in the region have ceased before reaching culturally-sterile deposits. Post-excavation analysis of material from both Toulkerrie (Bowen 1989; Hall and Bowen 1989) and NRS 8 (Neal 1984) revealed that cultural material was still being recovered in basal excavation units. There is also the possibility that evidence of earlier occupation is separated fr om more recent assemblages by sterile deposits such as at Wallen Wallen Creek (Neal and Stock 1986). Excavation Strategies Any estimation of the value ofWalters' model as an heuristic device must be tempered with the recognition of the extremely small sample employed in his study. Waiters provides no rationale 104 or discussion of sampling adequacy other than the vague statement that all shore types were represented (Waiters 1987:209). In his 1987 study, Waiters conducted independent excavations at only two sites, Sandstone Point and Toulkerrie. All other supporting evidence . is derived from research by other investigators . .In the main, the research designs used by these researchers were directed towards the resolution of issues of chronology and gross assemblage composition and not towards characterising broad changes in production strategies. Of the eight sites employed in Waiters' (1987) original study, five - St Helena Island, Wallen Wallen Creek, NRS 7, NRS 10 and Minner Dint - fa ll into this category. Waiters (1987:202, 204) adopted Jones' (1980) multum in parvo (much in little) principle at Sandstone Point where he excavated only a single 50 cm x 50 cm pit and analysed only the top fo ur excavation units out of a total of37. I suggest that this strategy was inappropriate given his objective of characterising the prehistoric marine fishery in Moreton Bay. Waiters makes the implicit (and dangerous) assumption that midden deposits in coastal southeast Queensland are homogeneous in composition and isotropic in structure. Ethnoarchaeological studies (e.g. Kent 1987) demonstrate that activity areas should not be expected to be distributed in a random or representative manner. This point is particularly important given ethnohistoric accounts that document some fishconsumption at specialised production sites (e.g. Dwyer and Buchanan 1986; Petrie 1904). A simple comparison of the area excavated and the total site area (see Table 4) indicates that the excavated fraction of archaeological deposits is much less than 0.5% fo r the majority of the sites reviewed, with only the Broadbeach Burial Ground providing an exception. The high degreeof spatial heterogeneity evident at the two middens which have been subjected to larger- scale excavation, Sandstone Point (Haglund 1974; Hall et al. 1987; Nolan 1986; Waiters 1987) 105 and Toulkerrie (Bowen 1989; Hall l984a; Hall and Bowen 1989; Waiters 1987), should serve as a caution against accepting results fr om small samples as representative of entire sites (see O'Neil 1993). I suggest that, given the small sample of excavated sites and of recovered fish remains fr om them, the total range ofprehistoric fishing activities in southeast Queensland is unlikely to be adequately represented. Contrary to Waiters' assertion, I argue that the available sample cannot be expected to encompass the full range of intra-site variability, let alone provide sufficient data to enable meaningful inter-site comparisons and insights into regional processes (see Anderson 1973). To date, the types of sampling procedures employed in the excavation of shell midden deposits in southeast Queensland have been unsuited to characterising the internal diversity and composition of sites at any but the coarsest level. Recovery Te chniques Another important sampling consideration is the ability of recovery techniques to yield data compatible with a given research question. Various screening experiments have demonstrated that mesh sizes impose significant biases on fauna! recovery (e.g. Shaffe r 1992; Shaffer and Sanchez 1994; Thomas 1969). Experiments which have included fishbones indicate that mesh sizes between 0.5 mm and 2 mm are required fo r maximum representative recovery (Casteel 1970, 1972, 1976a; Colley 1990; Fitch 1969; Gordon 1993; Reitz 1982). Indeed, Casteel (1976a, 1976b) has argued that column or core sampling is the only reliable method fo r representative recovery of fish remains. With the exception of the two Booral sites and St Helena Island where 2 mm screens were employed, 3 mm mesh is almost universally the smallest screen size used in excavations in the area (Table 4). This factor may have had a significant impact on the representation of fish remains in archaeological samples in the region. 106 In an earlier study, Waiters (1979) analysed a single 661.6 g bulk sample from Toulkerrie which indicated that as much as 80% of the fishremains passed through 3 mm mesh. Given this result it is somewhat surprising that he placed so much credence in the fish bone assemblages he used for his later study which were derived almost exClusively fr om 3 mm sieve residues. This study also demonstrated that the use oflarger mesh sizes biased recovery against some fishtaxa with small diagnostic skeletal elements, such as mullet and whiting (see also Hall 1980b:105-106). Also, the use of different sieve sizes makes it difficult to compare the representation of fish remains between sites. For example, Frankland (1990) attempted to compare the abundance of fish remains at Booral, where a 2 mm screen was employed, with those derived from sites around Moreton Bay where 3 mm screens were used. Analytical Me thods A variety of abundance measures have been employed in the analysis of fish remains from archaeological sites in the region. Waiters (1987) used number of identified specimens (NISP), McNiven (1990a) used both weight and minimum number of individuals (MNl) and Frankland (1990) used MNland NlSP. These abundance measures are not necessarily comparable (e.g. Grayson 1979, 1984). Figure 26 reveals major discrepancies between patterns revealed by NlSP and those given by MNl inthe abundance of fishremains fr om the Booral Shell Mound. NlSP is particularly prone to problems of inter-site comparability given the potential effects of taphonomic fa ctors and differences in the way that the measure is calculated by diffe rent investigators. Lyman (1994b:38) defines NlSP as "the number of identified specimens per taxon ... The taxon can be a subspecies, genus, family, or higher taxonomic category". Although there is some disagreement over whether the term 'specimen' can be equated with 107 1000 ,------, WNISP 100 � E " z 10 1 Excavation Unit Figure 26. Abundance of fish remains calculated using both NISP and MNI methods, Booral Shell Mound Square A (after Frankland 1990:59, 60). A logarithmic scale is employed so that small MNI figures are not artificially depressed. 'element' or 'fragment', there appears to be consensus on the meaning of 'taxa'. All investigators who have used NISP as a measure of fish remains abundance in southeast Queensland have adopted the highest taxonomic category possible - Pisces. The use ofNISP also raises significant problems in Waiters' application of the measure to calculate the abundance of fish remains discarded in various time intervals. The sites in his sample are not necessarily comparable even at a basic level. Direct inter-site comparison of these data is based on an assumption that all fish remains were subject to identical preparation, discard and post-depositional alteration and that the remains of all fishtaxa respond to these factors in the same way (cf Brian 1994). Waiters (1987) has already demonstrated that even in his small sample there were significant differences between fish bone assemblages in patterns of mechanical damage. A major fa iling of Waiters' analysis is the equation of excavation units with temporal units in presenting discard patterns at particular sites. In this fo rm of analysis there is an explicit 108 assumption that each excavation unit represents an equivalent (and therefore comparable) period oftime (Frankel 1988) and a constant rate of accumulation ofboth cultural materials and natural sediments. As Bird and Frankel (1991b:181) note, these arbitrary analytical units "combine material from an unknown number of activities or episodes of occupation into longer- term archaeological units". Although excavation units are useful as a descriptive framework fo r defining assemblage composition and variability, they are inappropriate fo r addressing questions of change through time. These chronological limitations are exacerbated when inter-site comparisons are attempted. Another underlying assumption in comparing site assemblages is that the assemblages were created by activities which are functionally equivalent. This approach denies functional variation between sites. At the Booral Shell Mound site, fo r example, a series of five radiocarbon dates from a single 50 cm x 50 cm pit provided Frankland (1990) with a well-dated sequence fr om which to calculate rates of discard of cultural material in 1 00-year intervals. This analysis demonstrated that in terms of both MNI and NISP, the most intensive period of fish discard at the site occurred between 2950 and 2790 BP (see Figure 24). This fo rm of analysis is not possible fo r most other coastal sites because of poor chronological resolution. Summary In this thesis I have demonstrated the invalidity of some of the fundamental assumptions underlying the model of intensified marine fishery production in southeast Queensland in the late-Holocene. The model of dramatic late-Holocene change in marine resource production presented by Waiters (1987, 1989, 1992a) is based on an inadequate sample of sites combined with a poor understanding and/or consideration of the role of taphonomic processes and research methods in structuring archaeological fa una! assemblages in the region. 109 AsBinf ord (1981 :29) notes, the "accuracy ofour inferential constructions ofthe past is directly dependent on the accuracy of the assumptions or premises serving as the basis of our inferential arguments". I have demonstrated that three of the major assumptions underlying Waiters' model of late-Holocene intensification of marine fishing in southeast Queensland are not supported by the available evidence. The equivocal status of these assumptions undermines the explanatory potential ofthe model. In seeking a model with broad explanatory fe atures Waiters has minimised the significance of major variation in environmental and archaeological patterns in favour of generalised characterisations of human adaptive responses. CHAPTER SEVEN SUMMARY AND CONCLUSIONS: Directions and Implications for Future Research Australian interpretations are to some extent premature and more archaeological investigation is required, as well as theory developed at the middle range to accommodate these issues. For example, while the more recent processes of socio-economic intensification are becoming increasingly evident, their genesis, presumably in the mid-Holocene and earlier, is less clear (Lourandos 1988:160). Introduction This chapter concludes the study with a brief summation of the main arguments presented and a consideration of the implications of these fo r continuing archaeological research in southeast Queensland and Australia generally. Finally, a number of areas fo r future research are identified to clarity and expand upon aspects of this study. Waiters' Model: A Reassessment In this thesis I have demonstrated significant problems in three premises central to Waiters' interpretation of prehistoric culture change in southeast Queensland. Firstly, environmental, ethnohistorical and archaeological evidence does not support Waiters' proposition that the coastal lowlands were a marginal landscape fo r human occupation at any time in the Holocene or that a time-Jag occurred between sea-level stabilisation and Aboriginal occupation of the coast. Nor is there any palaeoecological evidence to support Waiters' argument that periods of greater sedimentation occurring around 3,000 BP were related to increases in marine resource productivity. If such an enrichment did occur, it does not correlate with any changes in the archaeological record fr om this time. The occupational chronology demonstrates that significant increases in the number of occupied sites and the rate of site establishment does not 110 111 occur until around 1,000 BP, some 2,000 years after the proposed enrichment of the Bay. Finally, there is no consistent pattern of increase in the number of fishremains recovered from archaeological sites in the regionthrough time. Infa ct, most excavated coastal sites which date to the last 1,000 years do not exhibit fishrerriains at all. The fishery intensification model proposed by Waiters is constructed on a series of nested propositions. To build an abstract model based on empirical data such as the one proposed by Waiters requires a high level of confidence in the reliability of the database. I have demonstrated that Waiters' model does not provide a reliable framework fo r the explication of prehistoric cultural change in southeast Queensland because of major problems in the assumptions upon which it is fo unded. Other models of cultural change in the study region which share these assumptions, such as those proposed by Hall (1987, 1990), Morwood (1987) and McNiven (1990a, 1991a), are similarly problematic and require revision and qualification oftheir conclusions to accommodate the low levels of confidence which can be placed in some categories of archaeological data from the region. Assessing the significance oflate-Holocene changes in Aboriginal societies is dependent on accurately characterising the context in which these adaptations emerged. In southeast Queensland, analyses of fish production are limited by the lack of direct evidence for subsistence behaviour prior to the late-Holocene. Although there is apparently widespread occupation before this time, there is virtually no associated fa una! material. Terminal Pleistocene fish remains are evident in the basal levels ofBushranger's Cave and mid-Holocene fishing hasbeen documented at the New Brisbane Airport site and Platypus Rockshelter. The Booral Shell Mound documents significant discard of fish remains from the first occupation of 112 the site shortly before 3, 000 BP. Despite intensive research and numerous excavations conducted in the southeast Queensland coastal region, however, little is known of the nature of occupation of the region prior to the late-Holocene. Although current interpretations are limited by the near-absence of fauna! remains dating to before 4,000 BP, accumulating evidence suggests that marine resources (including fish) have been a significantresource in Aboriginal economies in southeast Queensland since at least the mid-Holocene. This proposition can be tested by further excavation. If the argument is sustained, then Aboriginal production systems in southeast Queensland can be seen as conforming to continuous patterns of marine resource use demonstrated elsewhere on the Australian continent, and not anomalous as Waiters (1992c:35) suggests. Lourandos' (1988:160) characterisation of Australian archaeological investigation cited at the beginning of this chapter is particularly appropriate in concluding the present study. Despite almost 30 years of organised archaeological endeavour in southeast Queensland, the antecedents of the complex systems recorded ethnohistorically and documented archaeologically fo r the recent past are little understood. An understanding of the developmental processes giving rise to these late prehistoric systems will require inter disciplinary research fo cused on the location of early- to mid-Holocene coastal archaeological deposits. This will involve detailed modelling of transgressive events and past shorelines in addition to an assessment of the impact of recent landscape modification on the visibility and preservation of sites. 113 Directions fo r Future Research Collaboration with geologists may be an important avenue fo r the investigation of submerged sites fo llowing the success of recent studies of this kind in North America (e.g. Dunbar et al. 1992; Hoyt et al. 1990; Stright 1990). Little research has yet been conducted in Australia on submerged archaeological sites (Dortch 1991; Dortch and Godfrey 1990). This avenue of research is a natural extension of coastal archaeology and is particularly important in the attempt to define the antecedents of post-transgressive coastal systems and as an important ' adjunct to studies ofpre-transgressive occupation of distant offshore islands (e.g. Veth 1993b). In Moreton Bay, Jones (1992:27) has noted the potential fo r identifYing submerged Aboriginal occupation sites along palaeo-river channels located by seismic profiling. Indeed, geological coring of marine deposits in Moreton Bay has already revealed a dense shell deposit of possible cultural origin at the confluence of the palaeo-Pine and palaeo-Caboolture Rivers (Dr J. Hall, Department of Anthropology and Sociology, The University of Queensland, pers. comm., 1994). Seismic surveys have also defined the probable course of the palaeo-Brisbane River channel as running close to the northwest coast ofMoreton Island (see Figure 6). Although parts of this coast have pro graded since 6,000 BP, it is possible that early- to mid-Holocene archaeological deposits are extant in this area, perhaps underlying more recent occupation sites. Also there is a large part of the eastern Bay which is subject to minimal sedimentation and is described as "virtually a submerged old land surface" (Jones 1992:30), enhancing the possibility of locating submerged sites in this area. Monitoring of the industrial extraction of mineral sands and corals fr om Moreton Bay may also reveal submerged archaeological deposits (Dortch 1991; Stephens 1992). 114 Recent parochial arguments in the literature (see McNiven !99la; Waiters 1992c) concerning the antiquity of marine fishing in southeast Queensland and its relative importance in the production economy can be directly evaluated using skeletal evidence. Stable isotopes analyses in palaeodietary studies may "provide an independent database that can improve behavioural predictions based on conventional archaeological and physical anthropological analyses" (Pate 1995:81). Not only does this technique enable direct determination of the relative role of marine and terrestrial resources in the subsistence economy but in some cases can be used to determine the importance of fish as opposed to other marine resources (Hay den et al. 1987). Stable carbon isotopes ratios may also enable independent examination of the emergence of social closure structures hypothesised fo r southeast Queensland (e.g. Hall in press; Hall and Bowen 1989; Hall and Hiscock 1988; McNiven 1991a, in press) by determining the degree of access inland groups had to coastal resources (see, fo r example, Aufderheide et al. 1994; Burton and Price 1990a, 1990b; Pate and Schoeninger 1993; Walker 1986; Walker and DeNiro 1986 fo r various palaeodietary applications of stable isotopes analysis). To address specific questions concerning changes in mobility and production strategies proposed in models fo r southeast Queensland, fu ture excavation strategies fo r the coastal region will need some modification to ensure adequate recovery of small cultural materials. Methods suitable to the recovery of fishand plant remains, such as finescreen sieving (e.g. 0.5 mm - 2.0 mm), bulk sediment sampling and chemical studies of sediments should be employed. Sample size and representativeness are also critical issues which need to be addressed here and elsewhere in Australia. Sites cannot be assumed to be homogeneous in content and structure. To provide meaningful data on intra- and inter-site variability, excavation strategies need not only to cover broader areas but to confidentlyrepresent the full range of content and structure 115 present in the archaeological record. In addition, finer chronological resolution must be established to aid in the interpretation of contemporaneity, periodicity and intensity of site use. The development of a relative dating system for unstratified lithic scatters, or application of direct dating techniques using Accelerator Mass Spectrometric (AMS) dating of residues, is important fo r defining variation in the use of landscapes away from the coast (Bonica 1992; Hiscock 1988). Futther excavation and analysis of sites in the region (in particular Booral Shell Mound, Sandstone Point, Hope Island, New Brisbane Airport and Wallen Wallen Creek as well as previously uninvestigated sites) is necessary to augment our understanding of subsistence behaviour and to increase sample sizes to bring greater resolution to issues of mid-Holocene occupation. Further excavation of sites known to date earlier than the late-Holocene may provide a solid basis fo r future interpretations of mid-Holocene adaptations. Archaeological research in biogeographic zones adj acent to the coastal lowlands may also fu rther contribute towards an understanding of whether the ecological structure of this landscape is an important factor in determining patterns ofhuman settlement and use (see Lilley and Ulm [in press] fo r a discussion of preliminary results of research on the coast in the transitional zone between coastal lowlands and other coastal landscapes to the north). Although substantial archaeological investigations were conducted on the northeast New South Wales coast in the 1970s, the value in understanding the use of coastal lowlands landscapes is eroded by apparent chronological problems and a general lack of detailed reporting (e.g. Campbell 1969, 1972; Coleman 1978; Connah 1975, 1976; McBryde 1982). Future studies in this region may aid in understanding patterning in the archaeological record of in southeast Queensland. 116 The results of the present study suggest that a re-evaluation of the role of marine resources in Ab original economies in southeast Queensland is required. In particular, a broader range of near-coastal sites need to be excavated to define the variability of subsistence behaviour. Studies on the role of plant fo od production in the overall economy are also necessary to develop a balanced context fo r establishing the significance of changes in other subsistence sources (e.g. Gillieson and Hall 1982; Hall et al. 1989; Higgins 1988; Robertson 1994). Also the actual concept of marginality, how it is socially constructed and how it affects actual patterns ofhuman use oflandscapes, requires clearer definition (see Godwin 1983, 1990). Summary Models such as the one constructed by Waiters are valuable devices for directing research effort at a regional scale. Most explanatory frameworks in Australian prehistory tend to fo cus on site- or local-level change and only incidentally make reference to broader frameworks of interpretation. The intensity of archaeological research conducted in southeast Queensland and northeast New South Wales provides a valuable regional database for examining mid- to late- Holocene cultural systems. However, as David and Chant (1995 :iii) note, the utility of these archaeological data for comparisons through space and time is limited because of the lack of comparability in the way in which data have been collected, analysed and reported. While current evidence does not refute Waiters' proposition that marine fish production intensified in the late-Holocene, several of the key premises of his model have been shown to be untenable. Waiters' conceptualisation of the southeast Queensland coastal lowlands as a marginal landscape depauperate in resources has been shown to be a gross oversimplification. Also, it has been demonstrated that both the timing of coastal occupation and the nature of resource use in the regionwere considerably more variable than Waiters' model allows. These - 117 findings are consistent with the emerging pattern of coastal occupation fo r Australasia in general (e.g. Alien et al. 1989; Barker 1991, 1993; Morse 1988, 1993a, 1993b; O'Connor 1989, 1994; Veth 1993b; cf Beaton 1985). The challenge fo r archaeologists working in southeast Queensland is to address the seribus limitations of the archaeological database exposed in this study and thus establish a solid basis fo r modelling regional prehistoric behavioural processes. 118 REFERENCES CITED Adams, D. 1984 The Hitch Hiker 's Guide to the Galaxy: A Tr ilogy in Four Parts. London: Heinemann. Alfredson, G. 1983 St. Helena Island: A changing pattern of exploitation? Australian Archaeology 17:79-86. 1984a An Archaeological Investigation into the Aboriginal use of St. Helena Island, Moreton Bay. Unpublished B.A (Hons) thesis, Department of Anthropology and • Sociology, University of Queensland. 1984b The Aboriginal use of St. Helena Island - The archaeological evidence. In R.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke: New Information on North Stradbroke Island and Surrounding Areas, 1974-1984, pp. 1-8. Brisbane: Boolarong Publications. Alien, J., C. Gosden and J.P. White 1989 Human Pleistocene adaptations in the tropical island Pacific. Antiquity 63:548- 561. Alien, J. and S. Holdaway 1995 The contamination of Pleistocene radiocarbon determinations in Australia. Antiquity 69:101-112. Anderson, A.J. 1973 A critical evaluation of the methodology of rnidden sampling. Ne w Zealand ArchaeologicalAssociation Newsletter 16: 1 19-127. 1991 The chronology of colonization in New Zealand. Antiquity 65:767-795. Anon. 1981 Fish species from inshore areas vital fo r Queensland fishermen. Australian Fisheries 40(1):41-43. Amold, I.E. 1993 Labor and the rise of complex hunter-gatherers. Journal of Anthropological Archaeology 12(3):75-119. 119 Attenbrow, V. I982 The archaeology ofUpper Mangrove Creek: Research inprogress. In S. Bowdler (ed.),Coastal Archaeology in Eastern Australia: Proceedings of the 1980 Valia Conference on Australian Prehistory, pp.67-79. Occasional Papers inPrehistory No. I I. Canberra: Department ofPrehistory, Research School of Pacific Studies, Australian National University. Aufderheide, A. C., M.A. Kelley, M. Rivera, L. Gray, L.L. Tieszen, E. Iversen, H.R. Krouse and A. Carevic I994 Contributions of chemical dietary reconstruction to the assessment of adaptation by ancient highland inunigrants (Atto Rarnirez) to coastal conditions at Pisagua, North Chile. Journal of Archaeological Science 2I:5I5-524. Bailey, G.N. I983 Problems of site fo rmation and the interpretation of spatial and temporal discontinuities in the distribution of coastal mid dens. In P .M. Masters and N. C. Flemrning (eds), Quaternary Coastlines and Marine Archaeology, pp.559-582. London: Academic Press. Balme, J. I990 A Pleistocene Tradition: Aboriginal Fishery on the Lower Darling River, Western New South Wales. 2 vols. Unpublished PhD thesis, Department of Prehistory, Australian National University. I995 30,000 years of fishery in western New South Wales. Archaeology in Oceania 30(1):I-21. Bartholomai, A. I976 Faunal remains excavated from site QI, Broadbeach, southeast Queensland. Appendix B. In L. Haglund, The Broadbeach Aboriginal Burial Ground: An Archaeological Analysis, pp.90-93. St. Lucia: University of Queensland Press. Barker, B.C. I989 Nara Inlet I: A Holocene sequence from the Whitsunday Islands, central Queensland coast. Queensland Archaeological Research 6:53-76. I99I Nara Inlet I: Coastal resource use and the Holocene marine transgression in the Whitsunday Islands, Central Queensland. Archaeology in Oceania 26(3): I 02-I 09. I993 Early human exploitation of island environments within the Great Barrier Reef Marine Park. ReefResearch 3(3): 13-I4. Beaton, J.M. I982 Fire and water: Aspects of Australian Aboriginal management of cycads. Archaeology in Oceania I7(I):5I-58. 120 1985 Evidence fo r a coastal occupation time-lag at Princess Charlotte Bay (North Queensland) and implications for coastal colonisation and population growth theories for Aboriginal Australia. Archaeology in Oceania 20(1 ): 1-20. 1990 The importance of past population fo r prehistory. In B. Meehan and N. White (eds), Hunter-Gatherer Demography: Past and Present, pp.23-40. Oceania Monograph No. 39. University of Sydney. Bell, C. 1979 Reconstruction of vegetation from spectra of pollen and spores in Quaternary sediments from Moreton Bay and Maroochydore. In A. Bailey and N.C. Stevens (eds), NorthernMo reton BaySy mposium: The Proceedings of a Sy mposium held atthe Abet Smith Lecture Theatre, Un iversity of Queensland, September 23-24, 1978, pp.29-3 1. Brisbane: Royal Society of Queensland. Bell, J.D. and D.A. Pollard 1989 Ecology offish assemblages and fisheries associated with seagrasses. In A.W.D. Larkum, A.J. McComb and S.A. Shepherd (eds), Biology of Seagrasses: A Treatise on the Biology of Seagrasses with Sp ecial Reference to the Australian Region, pp.565-609. Aquatic Plant Studies 2. Amsterdam: Elsevier. Bender, B. 1978 Gatherer-hunter to fa rmer: A social perspective. Wo rld Archaeology 10(2):204- 222. 1981 Gatherer-hunter intensification. In A. Sheridan and G. Bailey (eds), Economic Archaeology: Towards an in tegration of Ecological and Social Approaches, pp.149-157. Oxford: British Archaeological Reports, International Series 96. Berrningham, A. 1966 Victoria natural radiocarbon measurements. Radiocarbon 8:507-521. Binford, L.R. 1964 A consideration of archaeological research design. American Antiquity 29(4):425- 441. 1981 Bones: Ancient Men and Modem My ths. New York: Academic Press. Bird, C.F.M. and D. Frankel 1991 a Chronology and explanation in western Victoria and south-east Australia. Archaeology in Oceania 26(1):1-16. 1991b Problems in constructing a prehistoric regional sequence: Holocene southeast Australia. World Archaeology 23 (2) : 179-192. 121 Bird, M.K. 1992 The impact of tropical cyclones on the archaeological record: An Australian example. Archaeology in Oceania 27(2):75-86. 1995 Coastal morphodynamics and the archaeological record: Further evidence from Upstart Bay, north Queensland. Australian Archaeology 40:57-58. Birdsell, J.B. 1977 The recalibration of a paradigm for the first peopling of Australia. In J. Alien, J. Golson and R. Jones (eds), Sunda and Sah ul: Prehistoric Studies in Southeast Asia, Melanesia andAu stralia, pp. 113-167. London: Academic Press. Bonica, C.L. , 1992 A Technological Analysis of the Lithic Assemblage from Christmas Creek Rockshelter, southeast Queensland. Unpublished B.A.(Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Boserup, E. 1965 The Conditions of Agricultural Growth. Chicago: Aldine. Bowdler, S. 1976 Hook, line and dilly bag: An interpretation of an Australian coastal shell midden. Mankind 10(4):248-258. 1981 Hunters inthe highlands: Aboriginal adaptations inthe eastern Australian uplands. Archaeology in Oceania 16(2):99-111. 1995 The excavation of two small rockshelters at Monkey Mia, Shark Bay, Western Australia. Australian Archaeology 40:1-12. Bowen, G. 1989 A Model fo r Moreton Island Prehistory: Colonisation, Settlement and Subsistence. Unpublished B.A. (lions) thesis, Department of Anthropology and Sociology, University of Queensland. Bowman, G. and N. Harvey 1983 Radiocarbon dating marine shells in South Australia. Australian Archaeology 113-123. Breschini, G.S. and T. Haversat 1991 Early Holocene occupation ofthe central California coast. In J.M. Erlandson and R.H. Colten (eds), Hunter-Gatherers of Early Holocene Coastal California, pp.125-132. Perspectives inCalif ornia Archaeology Vol. 1. Los Angeles: Institute of Archaeology, University of California. 122 Brian, D. 1994 Shall I Compare Thee to a Fish?: A Comparative Taphonomic Analysis of Vertebrate Remains fromNara Inlet ArtSite, Hook Island, Central Queensland Coast. Unpublished B.A (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Burton, J.H. and T.D. Price 1990a Paleodietary applications of barium values in bone. In E. Pernicka and G.A Wagner (eds), Proceedings of the 27th In ternational Sy mposium on Archaeometry, Heideberg, /990, pp.787-795. Base!: Berkhauser Verlag. 1990b The ratio of barium to strontium as a paleodietary indicator of consumption of marine resources. Jo urnal of Archaeological Science 17:547-557. Caddie, D.A, H.J. Hall, D.S. Hunter and P.J. Pomery 1984 The application of electron spin resonance dating to Holocene bone. Australian Archaeology 19:75-85. Callaghan, M. 1980a Some previously unconsidered environmental fa ctors of relevance to south coast prehistory. Australian Archaeology 11:43-49. 1980b An Analysis of the Molluscan Remains from the Macleay River Midden Sites. Unpublished B.A. (Hons) thesis, Department of Prehistory and Archaeology, University ofNew England. Campbell, J. 1875 TheEarly Settlement of Queensland and Other Articles. Queensland: Ipswich Observer. Campbell, V. 1969 AField Study of Shell Middens of the lower Macleay Valley, with Reference to their Potential and Possible Methods oflnvestigation. Unpublished B.A (Hons) thesis, Department of Prehistory and Archaeology, University ofNew England. 1972 Some radiocarbon dates from Aboriginal shell middens in the Lower Macleay River Valley, New South Wales. Mankind 8(4):283-286. Casteel, R.W. 1970 Core and column sampling. American Antiquity 35(4):465-467. 1972 Some biases inthe recovery of archaeological fa una! remains. Proceedings of the Prehistoric Society 38:382-3 88. 1976a Comparison of column and whole unit samples fo r recovering fishremains. World Archaeology 8(2):192-198. 123 1976b Fish Remains in Archaeology and Palaeo-Environmental Studies. London: Academic Press. Catterall, C.P. and I.R Poiner 1984 Distribution, abundance and resilience to gathering of macro-molluscs in the seagrass bedsofNorth Stradbroke Island. In R.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradhroke: Ne w Information on North Stradbroke Island and Surrounding Areas, I974-1984, pp.291-299. Brisbane: Boolarong Publications. 1987 The potential impact ofhuman gathering on shellfish populations, with reference to some NEAustralian intertidal flats. Oikos 50(1):1 14-122. Chappell, J. 1982 Sea levels and sediments: Some fe atures of the context of coastal archaeological sites in the tropics. Archaeology in Oceania 17(2):69-78. Claassen, C. 1991 Normative thinking and shell-bearing sites. In M.B. Schiffer (ed.), Archaeological Method and Theory 3:249-298. Tucson: University of Arizona Press. Coaldrake, J.E. 1961 The Ecosystems of the Coastal Lowlands ("wallum ") of Southern Queensland. C.S.I.R.O. Bulletin 283. Melbourne: Commonwealth Scientific and Industrial Research Organisation. Coleman, J. 1978 The Analysis of Vertebrate Fauna! Remains from Four Shell Middens in the Lower Macleay River District. Unpublished B.A. (Hons) thesis, Department of Prehistory and Archaeology, University ofNew England. Colley, S.M. 1990 The analysis and interpretation of archaeological fish remains. In M.B. Schiffer (ed.), Archaeological Me thod and Theory 2:207-253. Tuscon: University of Arizona Press. Collier, S. and K.A Hobson 1987 The importance of marine protein in the diet of coastal Australian Aborigines. Current Anthropology 28(4):559-564. Connah, G. 1975 Current research at the Department ofPrehistory and Archaeology, University of New England. Australian Archaeology 3:28-3 1. 1976 Archaeology at the University ofNew England, 1975-6. Australian Archaeology 5:1-6. 124 Covacevich, J. and G.J. Ingram 1975 The reptiles of Stradbroke Island. Proceedings of the Royal Society of Queensland 86:55-60. Crooks, J. 1982 Report ofThree Excavations at Pumicestone Passage, Moreton Bay, Southeast Queensland. Unpublished B.A (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Curtis, J. 1838 Shipwreck of the Stirling Castle. London: George Virtue. David, 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. David, B. and D. Chant 1995 Rock art and regionalisation in north Queensland prehistory. Memoirs of the Queensland Museum 37(2):357-528. Dodson, J.R., V.M. McRae, K. Molloy, F. Roberts and J.D. Smith 1993 Late Holocene human impact on two coastal environments in New South Wales, Australia: A comparison of Aboriginal and European impacts. Vegetation History and Archaeology 2(2):89-1 00. Dortch, C.E. 1991 Rottnest and Garden Island prehistory and the archaeological potential of the adjacent continental shelf, Western Australia. Australian Archaeology 33 :38-43. Dortch, C.E. and I.M. Godfrey 1990 Aboriginal sites in a submerged landscape at Lake Jasper, southwestern Australia. Australian Archaeology 31:28-33. Dortch, C.E., G.W. Kendrick and K. Morse 1984 Aboriginal mollusc exploitation in southwestern Australia. Archaeology in Oceania 19(3):81-104. Dowling, R.M. 1986 The mangrove vegetation of Moreton Bay, Queensland. Botany Bulletin 6. Queensland Department ofPrimary Industries. Draper, N. 1978 A Model of Aboriginal Subsistence and Settlement in the Moreton Bay Region of Southeast Queensland. Unpublished B.A (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 125 1980 Food resources ofthe Moreton Bay Aborigines. InP .K. Lauer (ed. ), Occasional Papers in Anthropology 10:124-147. St. Lucia: Anthropology Museum, University of Queensland. Dunbar, J.S., S.D. Webb and M. Faught 1992 Inundated prehistoric sites inApalache Bay, Florida, and the search fo r the clovis shoreline. In L.L. Johnson. (ed.), Paleoshore/ines and Prehistory: An In vestigation of Me thod, pp. 117-146. Boca Raton: CRC Press. Dunnell, R.C. and W.S. Dancey 1983 The siteless survey: A regional scale data collection strategy. In M.B. Schiffer (ed.), Advances in ArchaeologicalMe thod and Theory 6:267-287. New York: Academic Press. Durbidge, E. and J. Covacevich 1981 North Stradbroke Island. Amity Point: Stradbroke Island Management Organisation. Dwyer, B. and N. Buchanan 1986 The Rescue of E/iza Fraser. Noosaville: Noosa Graphics. Dwyer, P.D., M. Hockings and J. Willmer 1979a Mammals of Cooloola and Beewah. Proceedings of the Royal Society of Queensland 90:65-84. Dwyer, P.D., J. Kikkawa and G.J. Ingram 1979b Habitat relations of vertebrates in subtropical heathlands in coastal southeastern Queensland. In R.L. Specht (ed.), Heathlands and Related Shrublands: Descriptive Studies, pp.281-299. Ecosystems of the World 9A. Amsterdam: Elsevier Scientific Publishing Company. Edmonds, V.J. 1986 Subtropical Rainforest: Archaeological Evidence fo r Aboriginal Use in the Border Ranges, Southeast Queensland. Unpublished M.Litt. thesis, Department of Archaeology and Palaeoanthropology, University ofNew England. Ellway, C.P. and E.J. Hegerl 1972 Fishes of the Tweed River estuary. Op erculum 2(1): 16-23. Erlandson, J.M. 1988 The role of shellfish in prehistoric economies: A protein perspective. American Antiquity 53(1):102-109. 126 1991 Shellfish and seeds as optimal resources: Early Holocene subsistence on the Santa Barbara coast. In J.M. Erlandson and R.H. Colten (eds), Hunter-Gatherers of Early Holocene Coastal California, pp.89-100. Perspectives in California Archaeology Vol. I. Los Angeles: Institute of Archaeology, University of California. Fairholme, J.K.E. 1856 The blacks of Moreton Bay and the porpoises. Proceedings of the London Zoological Society24:353-354. Fitch, J.E. 1969 Fish remains, primarily otoliths, from a Ventura, California Chumash village site (VEN-3). Memoirs of the Southern California Academy of Sciences 8:56-70. Flinders, M. 1814 A Voyage to Terra Australis. 2 vols and atlas. London: G. and W. Nicol. Flood, J. 1980 The Moth Hunters. Canberra: Australian Institute of Aboriginal Studies. 1995 Archaeology of the Dreamtime: The Story of Prehistoric Australia and its People. 3rd ed. Sydney: Angus and Robertson. Flood, J., B. David, J. Magee and B. English 1987 Birrigai: A Pleistocene site in the south-eastern highlands. Archaeology in Oceania 22(1):9-26. Flood, P.G. 1981 Carbon-14 dates from the coastal plains of Deception Bay, southeastern Queensland. Queensland GovernmentMi ning Journal 82: 19-23. 1984 A review ofHolocene sea level data, southeastern Queensland. In R.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke: Ne w Information on North Stradbroke Island and Surrounding Areas, I974-1984, pp.127-131. Brisbane: Boolarong Publications. Foley, R. 1981a A model of regional archaeological structure. Proceedings of the Prehistoric Society 47:1-17. 1981b Off- Site Archaeology andHuman Adaptation in Eastern Africa: An Analysis of Regional Artefa ct Density in the Amboeli, Southern Kenya. Oxford: British Archaeological Reports, International Series 97. Frankel, D. 1988 Characterising change in prehistoric sequences: A view from Australia. Archaeology in Oceania 23(2):41-48. 127 Frankland, K. 1990 Booral: A Preliminary Investigation of an Archaeological Site inthe Great Sandy Strait Region, Southeast Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Gaughwin, D. and R. Fullagar 1995 Victorian offshore islands in a mainland coastal economy. Australian Archaeology 40:38-50. Gillespie, R. and H.A. Polach 1979 The suitability of marine shells for radiocarbon dating of Australian prehistory. In R. Berger and H. Suess (eds), Radiocarbon Dating: Proceedings of the 9th In ternational Conference on Radiocarbon Dating, pp.404-421. Los Angeles: University of California Press. Gillieson, D.S. 1982 AboriginalSites of theAlbert Valley, South East Queensland. Cultural Resource Management Monograph Series 3. Brisbane: Archaeology Branch, Department of Community Services. Gillieson, D.S. and J. Hall 1982 Bevelling bungwall hashers: A use-wear study from southeast Queensland. Australian Archaeology 14:43-61. Godfrey, M.C.S. 1989 Shell midden chronology in southwestern Victoria: Reflections of change in prehistoric population and subsistence. Archaeology in Oceania 24(2):65-69. Godwin, L. 1983 Archaeological site surveys on the eastern margin of the New England tablelands. Australian Archaeology 17:38-47. 1990 Inside Information: Settlement and Alliance in the Late Holocene of northeastern New South Wales. Unpublished PhD thesis, Department of Archaeology and Palaeoanthropology, University ofNew England. Gordon, E.A. 1993 Screen size and differential fa unal recovery: A Hawaiian example. Journal of Field Archaeology 20(4):453-460. Grayson, D.K. 1979 On the quantification of vertebrate archaeofaunas. In M.B. Schiffer (ed.), Advances in Archaeological Me thod and Theory 2:199-23 8. New York: Academic Press. 1984 Quantitative Zooarchaeology: Top ics in the Analysis of Archaeological Faunas. Orlando: Academic Press. 128 Gregor, R.J. 1846 Reply to a circular letter. Ne w South Wa les Legislative Council Vo tes and Proceedings 2:549. Haglund, L. 1974 The Sandstone Point shell midden: A preliminary report. Appendix 5. In J. Stockton, Report of an Archaeological Survey in the Vicinity of Bribie Island, South-East Queensland, pp. 118-131. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 1976 Th e Broadbeach Aboriginal Burial Ground: An Archaeological Analysis. St. Lucia: University of Queensland Press. Ha:glund-Calley, L. 1968 An Aboriginal burial ground at Broadbeach, Queensland: Excavation report. Ma nkind 6(12):676-680. Haglund-Calley, L. and M. C. Quinnell 1973 A shell midden at Cascade Gardens, Broadbeach, southeast Queensland. Memoirs of the Queensland Museum 16(3):399-409. Hall, J. 1980a Archaeology at the University of Queensland: A brief outline. Australian Archaeology 10:79-85. 1980b Minner Dint: A recent Aboriginal midden on Moreton Island, southeastern Queensland. InP .K. Lauer (ed.), Occasional Papers in Anthropology 10:94-112. St. Lucia: Anthropology Museum, University of Queensland. 1982 Sitting on the crop of the bay: An historical and archaeological sketch of Aboriginal settlement and subsistence in Moreton Bay, southeast Queensland. In S. Bowdler (ed.), Coastal Archaeology in Eastern Australia: Proceedings of the I980 Va lla Conference on Australian Prehistory, pp. 79-95. Occasional Papers in Prehistory No. 11. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. 1984a Exploratory excavation at Toulkerrie midden (LB:B75), Moreton Island, S.E. Queensland. QueenslandArchaeo logical Research 1 :61-84. 1984b Fishing with dolphins?: Affirmingtraditional a Aboriginal fishing story inMoreton Bay, SE. Queensland. InR.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke : Ne w Information on North Stradbroke Island and Surrounding Areas, I974-I984, pp.16-22. Brisbane: Boolarong Publications. 1986 Exploratory excavation at Bushrangers Cave (Site LA:Al l), a 6000-year-old campsite in southeast Queensland: Preliminary results. Australian Archaeology 22:88-103. I29 I987 A short prehistory of the Moreton Region. In R. Fisher (ed.), Brisbane: Aboriginal, Alien, Ethnic, pp.I4-22. Brisbane History Group Papers 5. St. Lucia: Department of History, University of Queensland. I990 20 000 years of human impact on the Brisbane River and environs. In P. Davie, E. Stock and D.L. Choy (eds), The Brisbane River: A Source-Book for the Future, pp.I75-I82. Brisbane: Australian Littoral Society and the Queensland Museum. I995 The Saint-Smith Midden Site at Beachmere, S.E. Queensland: A Report on Archaeological Fieldwork I994. Unpublished report to the Cultural Heritage Section, Queensland Department of Environment and Heritage. St. Lucia: Department of Anthropology and Sociology, University of Queensland. in press The impact of sea-level rise on the archaeology of South East Queensland. In J. Hall and I. McNiven (eds), Australian Coastal Archaeology: Current Research and Future Directions. Hall, J. and G. Bowen I989 An excavation of a midden complex at the Toulkerrie Oysterman's lease, Moreton Island, SE Queensland. Queensland Archaeological Research 6:3-27. Hall, J., S. Higgins and R. Fullagar I989 Plant residues on stone tools: A case fo r fe rnroot processing in S.E. Queensland via starch grain analysis. In W. Beck, L. Head and A. Clarke (eds), Plants in Australian Archaeology, pp.l36-I60. Tempus I. St Lucia: Anthropology Museum, University of Queensland. Hall, J. and P. Hiscock I988 The Moreton Regional Archaeological Project (MRAP) - Stage IT: An outline of objectives and methods. Queensland Archaeological Research 5:4-24. Hall, J. and I. Lilley I987 Excavation at the New Brisbane Airport site (LB :C69): Evidence for early rnid Holocene coastal occupation in Moreton Bay, SE Queensland. Queensland Archaeological Research 4:54-79. Hall, J. and RP. Robins I984 A working model of Moreton Island prehistory: MRAP Stage I. Queensland Archaeological Research I :85-94. Hall, J., I. Waiters and A. Nolan I987 The Sandstone Point Archaeological Project: Stage I. Unpublished Report to the Minister, Department of Community Services, Brisbane. St. Lucia: Department of Anthropology and Sociology, University of Queensland. 130 Hardley, RG. 1975 Some ofthe Factors that Influenced the Coastal, Riverine and Insular Habitats of the Aborigines of South-East Queensland and of Northern N.S.W. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Harris, P. T. 1995 Marine geology and sedimentology of the Australian continental shelf. In L.P. Zann and P. Kailola (eds), The State of the Marine Environment Report for Australia Te chnical Annex 1, The Marine En vironment, pp. 11-23 . Townsville: Great Barrier Reef Marine Park Authority. Harris, P.T., C.B. Pattiaratchi, A.R. Cole and J.B. Keene ' 1992 Evolution of subtidal sandbanks in Moreton Bay, eastern Australia. Marine Geology 103:225-247. Hayden, B., B. Chisholm and H.P. Schwarz 1987 Fishing and fo raging: Marine resources in the Upper Paleolithic ofFrance. In 0. Soffer (ed.), The Pleistocene Old Wo rld: Regional Perspectives, pp.279-291. New York: Plenum Press. Head, J., R Jones and J. Alien 1983 Calculation of the 'Marine Reservoir Effect' from the dating of shell-charcoal paired amples from an Aboriginal midden on Great Glennie Island, Bass Strait. Australian Archaeology 17:99-113. Head, L. 1983 Environment as artefact: A geographic perspective on the Holocene occupation of southwestern Victoria. Archaeology in Oceania 18(2):73-80. 1986 Palaeoecological contributions to Australian prehistory. Archaeology in Oceania 21(2):121-129. 1987 The Holocene prehistory of a coastal wetland system: Discovery Bay, southeastern Australia. Human Ecology 15(4):435-462. 1994 Both ends ofthe candle? Discerning human impact on the vegetation. Australian Archaeology 39:82-86. Heather, A. 1983 Running Rings Around Moreton: Towards an Understanding ofthe Location of Earth Circle Sites in the Moreton Region, South-East Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 131 Heinsohn, G.E. 1991 Dugongs. In R Strahan (ed.), The Australian Museum Complete Book of AustralianMa mmals, pp.474-476. North Ryde: Cornstalk Publishing. Heke� H., W.T. Ward, M. Jones and D.E. Searle 1979 Geological development ofnorthern Moreton Bay. In A Bailey and N.C. Stevens (eds), NorthernMo reton BaySyrfl posium: The Proceedings of a Symp osium held atthe Abel SmithLecture Theatre, Un iversity of Queensland, September 23-24, 1978, pp.7-18. Brisbane: Royal Society of Queensland. Hendy, C.H., T.A. Rafter and N.W.G. Macintosh 1952 The fo rmation of carbonate nodules in the soils of the Darling Downs, Queensland, Australia and the dating of the Talgai cranium. In Proceedings of the • 8th In ternational Conference on Radio Carbon Dating, pp.D106-126. Wellington. Higgins, S 1988 Starch Grain Differentiation of Archaeological Residues: A Feasibility Study. Unpublished B.A. (Hons), Department of Anthropology and Sociology, University of Queensland. Hiscock, P. 1988 Developing a relative dating system fo r the Moreton Region: An assessment of prospects for a technological approach. Queensland Archaeological Research 5:113-132. 1994 Technological responses to risk in Holocene Australia. Jo urnal of Wo rld Prehistory 8{3):267-292. Hiscock, P. and I. Waiters 1994 Archaeology at the Northern Territory University: Update May 1993. Australian Archaeology 38:51-55. Hitchcock, G. 1994 Complex Labour Organisation and Intergroup Gatherings in Aboriginal Australia. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Hobson, K.A. and S. Collier 1984 Marine and terrestrial protein m Australian Aboriginal diets. Current Anthropology 25(2):238-240. 1986 On Australian Aboriginal diets: Reply. Current Anthropology 27{1):54. Holdaway, S. and N. Porch 1995 Cyclical patterns in the Pleistocene human occupation of southwest Tasmania. Archaeology in Oceania 30{2):74-82. 132 Horsfall, N. 1987 Living inRainf orest: The Prehistoric Occupation ofNorth Queensland's Humid Tropics. Unpublished PhD thesis, Department of Behavioural Science, James Cook University. Hoyt, W.H., J.C. Kraft and M.J. Chrzastowski 1990 Prospecting for submerged archaeological sites on the continental shelf; Southern mid-Atlantic Bight of North America. In N.P. Lasca and J. Donahue (eds), Archaeological Geology of North America, pp.147-160. Geological Society of America, Centennial Special Volume 4. Boulder, Colarado: Geological Society of America. Hudson, J. ' 1993 The impacts of domestic dogs on bone info rager camps; or, the dog-gone bones. In J. Hu dson (ed.), From Bones to Behavior, pp.301-323. Carbondale: Southern lllinois University. Hughes, P.J. and R.J. Lampert 1982 Prehistoric population change in southern coastal New South Wales. In S. Bowdler (ed.), Coastal Archaeology in Eastern Australia: Proceedings of the 1980 Va lla Conference on Australian Prehistory, pp.16-28. Occasional Papers in Prehistory No. 11. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. Hughes, P.J. and M.E. Sullivan 1981 Aboriginal burning and late Holocene geomorphic events in easternNSW. Search 12(8):277-278. Hutchings, P. and P. Saenger 1987 Ecology of Mang roves. St Lucia: University of Queensland Press. Ingram, G.J. and C.J. Corben 1975 The frog fa una ofNorth Stradbroke Island with comments on the 'acid' frogs of the wallum. Proceedings of the Royal Society of Queensland 86:49-54. Jablonski, D. 1980 Apparent versus real biotic effects of transgressions and regressions. Paleobiology 6:397-407. Jones, M.R. 1992 Moreton Bay and the sand islands. In W. Willmott and N. Stevens (eds), Rocks and Landscapes of Brisbane and Ip swich: Geology and Excursions in the Brisbane, Ip swich, and Pine Rivers Districts, pp.27-3 1. Brisbane: Geological · Society of Australia, Queensland Division. 133 Jones, M.R and A W. Stephens 1981 Quaternarygeological framework and resource potential ofMoreton Bay. InG. Hoffman (ed.), 1981 Field Conference, Brisbane-Ipswich Area, pp.17-23. Brisbane: Geological Society of Australia, Queensland Division. Jones, R 1977 The Tasmanian paradox. In RV.S. Wright (ed.), Stone Tools as Cultural Markers: Change, Evolution and Complexity, pp.189-204. Prehistory and Material Culture Series No. 12. Canberra: Australian Institute of Aboriginal Studies. 1980 Different strokes fo r different folks: Sites, scale and strategy. In I. Johnson (ed.), Holier than Thou: Proceedings of the 1978 Kiola Conference on Australian Prehistory, pp.151-118. Canberra: Research School of Pacific Studies, Department of Prehistory, Australian National University. 1984 Hunters and history: A case study from Western Tasmania. In C. Schrire (ed.), Past andPresent in Hunter-Gatherer Societies, pp.27-65. New York: Academic Press. Keast, A (ed.) 1981 Ecological Biogeography of Australia. Boston: W. Junk. Kelly, M. 1982 A Practical Reference Source to Radiocarbon Dates Obtained from Archaeological Sites in Queensland. Cultural Resource Management Monograph Series 4. Brisbane: Archaeology Branch, Department of Community Services. Kent, S. 1981 The dog - Archaeologist's best friend or worst enemy: The spatial distribution of faunal remains. Jo urnal of Field Archaeology 8(3):367-372. 1987 Understanding the use of space: An ethnoarchaeological approach. In S. Kent (ed.), Me thod and Theoryfo r Activity AreaResearch: An Ethnoarchaeological Approach, pp. 1-60. New York: Columbia University Press. 1993 Variability in faunal assemblages: The influence of hunting skill, sharing, dogs, and mode of cooking on fa unal remains at a sedentary Kalahari community. Journal of Anthropological Archaeology 12(3):323-385. Kikkawa, J. 1975 A place for land vertebrates. Proceedings of the Royal Society of Queensland 86:47-48. Kirkman, H. 1975 A description of the seagrass communities of Stradbroke Island. Proceedings of the Royal Society of Queensland 86:129-131. 134 Lampert, RJ. 1981 The Great Kartan My stery. Terra Australis 5. Canberra: Department of Prehistory, Research School of PacificStudies, Australian National University. Lampert, RJ. and P.J. Hughes 1974 Sea level change and Aboriginal coastal adaptations in southern New South Wales. Archaeology and Physical Anthropology in Oceania 9(3):226-235. Lanyon, J.M., C.J. Limpus and H. Marsh 1989 Dugongs and turtles: Grazers in the seagrass system. In A.W.D. Larkum, A.J. McComb and S.A. Shepherd (eds), Biology of Seagrasses: A Tr eatise on the Biology ofSeagrasseswith Sp ecial Reference to the Australian Region, pp.610- 634. Aquatic Plant Studies 2. Amsterdam: Elsevier. Lauer, P.K. 1977 Report ofa preliminary ethnohistorical and archaeological survey ofFraser Island. In P.K. Lauer (ed.), Occasional Papers in Anthropology 8:1-38. St. Lucia: Anthropology Museum, University of Queensland. 1979 The museum's role in fieldwork: The Fraser Island study. In P.K. Lauer (ed.), OccasionalPapers in Anthropology 9:31-72. St. Lucia: Anthropology Museum, University of Queensland. Lilley, I.A. 1978 Prehistoric Subsistence and Settlement in the Sub-Coastal Zone ofthe Moreton Region, Southeast Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 1984 Late Holocene subsistence and settlement in subcoastal southeast Queensland. Queensland Archaeological Research 1:8-32. Lilley, I.A. and H.J. Ha11 1988 An archaeological assessment of northeastern Moreton Island. University of Queensland Archaeological Services Unit Report 126. St Lucia: Department of Anthropology and Sociology, University of Queensland. Lilley, I. and S. Ulm inpress The Gooreng Gooreng Cultural Heritage Project: Some proposed directions and preliminary results of the archaeological program. Australian Archaeology 41 (1995) Little, E.A. 1993 Radiocarbon age calibration at archaeological sites of coastal Massachusetts and vicinity. Jo urnal of Archaeological Science 20:457-471. 135 Lourandos, H. 1980a Change or stability?: Hydraulics, hunter-gatherers and population in temperate Australia. World Archaeology 11(3):245-264. 1980b Forces of Change: Aboriginal Technology and Population in South-Western Victoria. Unpublished PhD thesis, Department of Anthropology, University of " Sydney. 1983 Intensification: A late Pleistocene-Holocene archaeological sequence from southwestern Victoria. Archaeology in Oceania 18(2):81-94. 1985 Intensification and Australian prehistory. In T.D. Price and J.A. Brown (eds), Prehistoric Hunter-Gatherers: The Emergence of Cultural Complexity, pp.385- ' 423. Orlando: Academic Press. 1987 Pleistocene Australia: Peopling a continent. In 0. Soffer (ed.), The Pleistocene Old World: RegionalPer spectives, pp.147-165. New York: Plenum Press. 1988 Palaeopolitics: Resource intensification in Aboriginal Australia and Papua New Guinea. InT. Ingold, D. Riches and J. Woodburn (eds), Hunters and Gatherers 1: History, Evolution andSo cial Change, pp.148-160. Vol. 1. New York: Berg. 1993 Hunter-gatherer cultural dynamics: Long- and short-term trends in Australian prehistory. Jo urnal of Archaeological Research 1(1):67-88. Lourandos, H. and A. Ross 1994 The great 'intensification debate': Its history and place in Australian archaeology. Australian Archaeology 39:54-63. Luebbers, R.A. 1978 Meals and Menus: A Study ofChange in Prehistoric Coastal Settlements in South Australia. Unpublished PhD thesis, Research School of Pacific Studies, Australian National University. 1994 Harvesting, trading and fragmentation patterns in Donax deltoides as indices to Aboriginal settlement strategies inthe Coorong, South Australia. Paper presented at the Australian Coastal Archaeology Symposium: Current Research and Future Directions, Maroochydore, 24-26 September. Lyman, R.L. 1994a Vertebrate Taphonomy. Cambridge: Cambridge University Press. 1994b Quantitative units and terminology in zooarchaeology. American Antiquity 59(1):36-71. 136 Lyon, P. 1970 Differential bone destruction: An ethnographic example. American Antiqutiy 35(2):213-215. Martin,J.H.D. 1975 A list of mammals fromStradbroke Island. Proceedings of the Royal Society of Queensland 86:73-76. Mather, P. (ed.) 1976 The Na tional Estate in the Moreton and Wide Bay-Bumett Regions, Southeast Queensland and Recommendations on its Ma nagement: A Report to the Co ordinator General. Rev. ed. Brisbane: Queensland Museum. Mathew, J. 1910 Two Representative Tr ibes of Queensland. London: T. Fisher and Unwin. Maxwell, W.G.H. 1970 The sedimentary framework ofMoreton Bay, Queensland. Australian Journalof Ma rine and Freshwater Research 21 :71-88. McBryde, I. 1982 Coast and Estuary: Archaeolog ical In vestigations on the North Coast of Ne w South Wales at Wo mbah and Schnapper Point. Canberra: Australian Institute of Aboriginal Studies. McCarthy, T.D. 1943a Two pebble industry sites ofHoabinhian type on the north coast ofNew South Wales. Records of the Australian Museum 21 (1 ):21-25. 1943b Trimmed pebble implements of Kartan type from ancient kitchen middens at Clybucca, New South Wales. Records of the Australian Museum 21(3): 164-167. McDonald, J. 1992 The Archaeology of the Angophora Reserve Rock Shelter (or helping the police with their enquiries). Environmental Heritage Monograph Series No. 1. Hurstville: New South Wales National Parks and Wildlife Service. 1994 Sydney Basin Rock Art: Original Style in the Context of its Prehistory. Unpublished PhD thesis, Department ofArchaeology and Anthropology, Faculty of Arts, Australian National University. Mclnnes, P. 1988 Sea-level Stabilization Arguments: Assessing their Explanatory Potential. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 137 McNiven, I.J. 1984 Initiating Archaeological Research in the Cooloola Region, Southeast Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 1985 An archaeological survey of the Cooloola region, S.E. Queensland. Queensland Archaeological Research 2:4-37. 1988 Brooyar Rockshelter: A late Holocene seasonal hunting camp from southeast Queensland. Queensland Archaeological Research 5:133-160. 1989 Aboriginal shell middens at the mouth of the Maroochy River, southeast Queensland. Queensland Archaeological Research 6:28-52. 1990a Prehistoric Aboriginal Settlement and Subsistence in the Cooloola Region, Coastal Southeast Queensland. Unpublished PhD thesis, Department of Anthropology and Sociology, Un iversity of Queensland. 1990b Blowout taphonomy: Non-cultural associations between fauna! and stone artefact assemblages along the Cooloola coast, southeast Queensland. Australian Archaeology 31:67-74. 1991a Teewah Beach: New evidence fo r Holocene coastal occupation in coastal southeast Queensland. Australian Archaeology 33:14-27. 1991b Settlement and subsistence activities along Tin Can Bay, southeast Queensland. QueenslandArchae ological Research 8:85-107. 199lc The Double Island Point Aboriginal burials, coastal southeast Queensland. Australian Archaeology 32:10-16. 1992a Shell middens and mobility: The use of off-site fa una! remains, Queensland, Australia. Jo urnal of Field Archaeology 19(4) :495-508. 1992b Bevel-edged tools from coastal southeast Queensland. Antiquity 66:701-709. 1992c Ethnohistorical reconstructions of Aboriginal lifeways along the Cooloola coast, southeast Queensland. Proceedings of the Royal Society of Queensland 102:5-24. 1992d Sandblow sites in the Great Sandy Region, coastal southeast Queensland: Implications for models oflate Holocene rainforest exploitation and settlement restructuring. Queensland Archaeological Research 9: 1-16. 1993 Corroboree Beach, Fraser Island: Archaeological Survey and Management Recommendations. Vol. 1. Unpublished report to the Queensland Department of Environment and Heritage. 138 inpress Fissioningand regionalisation: The social dimensions of changes inAboriginal use of the Great Sandy Region, coastal southeast Queensland. In J. Hall and I. McNiven (eds), Australian Coastal Archaeology Symposium: Current Research and Future Directions. McNiven, 1., K. Frankland and J. Hall n.d. Booral Shell Midden-Fishtrap Site Complex, Southeast Queensland: Preliminary Report on Recent Excavations. Unpublished Manuscript, Department of Anthropology and Sociology, University of Queensland. Meehan, B. 1977 Man does not live by calories alone: The role of shellfish in a coastal cuisine. In J. Alien, J. Golson and R. Jones (eds), Sunda and Sahul: Prehistoric Studies in • Southeast Asia, Me lanesia andAustralia, pp.493-53 1. London: Academic Press. 1982 Shell Bed to ShellMidd en. Canberra: Australian Institute of Aboriginal Studies. Meltzer, D.J. and J.l. Mead 1985 Dating late Pleistocene extinctions: Theoretical issues, analytical bias, and substantive results. In J.I. Mead and D.J. Meltzer (eds), Environments and Extinctions: Man in Late Glacial North America, pp.145-173. Orono: Centre for the Study ofEarly Man, University of Maine at Orono. Morse, K. 1988 Mandu Mandu Creek rockshelter: Pleistocene human coastal occupation ofNorth West Cape, WesternAustralia. Archaeology in Oceania 23(3):81-88. 1993a New radiocarbon dates from North West Cape, Western Australia: A preliminary report. In M.A. Smith, M. Spriggs and B. Fankhauser (eds), Sahul in Review: Pleistocene Archaeology in Australia, New Guinea and Is land Melanesia, pp.155-163. Occasional Papers in Prehistory No. 24. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. 1993b Who can see the sea? Prehistoric Aboriginal occupation of the Cape Range peninsula. InW.F. Humphreys (ed.), The Biogeography of CapeRange, Western Australia, pp.227-242. Records of the Western Australian Museum No. 45. Perth: Western Australian Museum. Morwood, M.J. 1986 The archaeology of art: Excavationsat Gatton and Maidenwell Rockshelters, S.E. Queensland. QueenslandArchaeological Research 3:88-132. 1987 The archaeology of social complexity in south-east Queensland. Proceedings of the Prehistoric Society 53:337-350. 139 Moseley, M.E. and R.A. Feldman 1988 Fishing, farming, and the fo undations of Andean civilisation. InG. Bailey and J. Parkington (eds), The Archaeology of Prehistoric Coastlines, pp.125-134. Cambridge: Cambridge University Press. Mulvaney, D.J. 1975 The Prehistory of Australia. Rev. ed. Harmondsworth: Penguin Books. Murphy, A.M.C. 1991 Into a Kingdom of Bones: A Metrical Analysis of the Pectoral Girdles and Innominates of the Broadbeach Australian Aborigines and the New Zealand Polynesians. Unpublished PhD thesis, Department of Antatomy, University of Queensland. Nanson, G.C. 1986 Episodes of vertical accretion and catastrophic stripping: A model of disequilibrium flood-plain development. Geological Society of America Bulletin 97:1467-1475. Neal, R. 1984 Rescue Archaeology Near Blue Lake, North Stradbroke Island. Unpublished report to Associated Minerals Consolidated Ltd and the Archaeology Branch, Department of Community Services. St Lucia: Department of Anthropology and Sociology, University of Queensland. 1989 Results ofNational Estate funding of radiocarbon dates from archaeological sites on North Stradbroke Island. Unpublished report to the Cultural Heritage Section, Queensland Department of Community Services and Ethnic Affairs. Neal, R. and E. Stock 1986 Pleistocene occupation in the south-east Queensland coastal region. Nature 323:618-621. Nicholson, A. and S. Cane 1991 Archaeology on the Anxious Coast. Australian Archaeology 33:3-13. 1994 Pre-European coastal settlement and use of the sea. Australian Archaeology 39:108-117. Nolan, A. 1986 Sandstone Point: Temporal and Spatial Patterns of Aboriginal Site Use at a Midden Complex, South-East Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 140 Novello, M. 1989 A Taphonomic Analysis ofthe Faunal Remains from Platypus Rockshelter (Site KB :A70), S.E. Queensland. Unpublished B.A (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Oakley, K.P., B.G. Campbell and T.I. Molleson 1975 Catalogue of Fossil Hominids Part Ill: Americas, Asia, Australasia. London: British Museum (Natural History). O'Connor, S 1989 New radiocarbon dates from Koolan Island, west Kimberley, W.A. Australian Archaeology 29:30-39. ' 1990 30,000 Years inthe Kimberley. Unpublished PhD thesis, Centre for Archaeology, University ofWesternAustralia. 1994 A 6700 BP date for island use in the West Kimberley, Western Australia: New evidence from High Clifl'y Island. Australian Archaeology 3 9: I 02-1 07. O'Connor, S. and M. Sullivan 1994a Distinguishing middens and cheniers: A case study from the southernKimberley, Western Australia. Archaeology in Oceania 29(1):16-28. 1994b Coastal archaeology in Australia; Developments and new directions. Australian Archaeology 39:87-96. O'Connor, S. and P. Veth 1993 Where the desert meets the sea: A preliminary report of the archaeology of the southern Kimberley coast. Australian Archaeology 37:25-34. O'Neil, D.H. 1993 Excavation sample size: A cautionary tale. American Antiquity 58(3):523-529. Osawa, R. 1984 A preliminary report on road mortality and fo od preference of the swamp wallaby, Wa l/abia bicolor, on North Stradbroke Island. In R.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke: New Information on North Stradbroke Island and Surrounding Areas, 1974-1984, pp.333-341. Brisbane: Boolarong Publications. Osborn, A.J. 1977 Strandloopers, mermaids, and other fa iry tales: Ecological determinants of marine resource utilization - The Peruvian case. In L.R. Binford (ed.), For Theory · Building in Archaeology, pp. I 57-205. New York: Academic Press. 1980 Comments on 'Maritime hunter-gatherers: Ecology and prehistory' by D.R. Yesner. Current Anthropology 21(6):740-741. 141 Parmalee, P.W. and W.E. Klippel 1974 Freshwater mussels as a prehistoric fo od source. American Antiquity 39(3):421- 434. Pate, F.D. 1995 Stable carbon isotope assessment ofhunter-gatherer mobility inprehistoric South Australia. Jo urnal of Archaeologi cal Science 22:81-87. Pate, F.D. and M.J. Schoeninger 1993 Stable carbon isotope ratios in bone collagen as indicators of marine and terrestrial dietary composition in southeastern South Australia: A preliminary report. In B.L. Fankhauser and J.R. Bird (eds), Archaeometry: Current Australasian Research, pp.38-44. Occasional Papers in Prehistory No. 22. • Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. Petrie, C.C. 1904 Tom Petrie 's Reminiscences of Early Queensland. Brisbane: Watson, Ferguson and Co. Pollock, B.R. 1980 Surprises in Queensland angling study. Australian Fisheries 3 9(4): 17-19. Proser, I. 1987 The history of Holocene alluviation in the southeastern highlands of Australia. Search 18(4):201-202. Queensland Department of Environment and Conservation 1989 Moreton Bay Strategic Plan: Background Information. Report to the Queensland Department of Environment and Conservation. Quinn, R.H. and J.P. Beumer 1984 Wallum Creek - A study of the regeneration of mangroves. In R.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke: Ne w Information on North Stradbroke Island and Surrounding Areas, 1974-1984, pp.238-259. Brisbane: Boolarong Publications. Reitz, E.J. 1982 Availability and use of fish in coastal Georgia and Florida. Southeastern Archaeology 1(1):65-88. Richardson, N. 1979 The Sources of Raw Materials and their Use in the Manufacture of Prehistoric Stone Artefacts on Moreton Island. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 142 1984 An archaeological investigation of sandmining lease SML 93 1 on North Stradbroke Island. InR.J. Coleman, J. Covacevich and P. Davie (eds), Focus on Stradbroke: New Information on North Stradbroke Island and Surrounding Areas, 1974-1984, pp.23-32. Brisbane: Boolarong Publications. Richter, J. 1994 A Pound ofBungwall and Other Measures. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Rick, J. 1987 Dates as data: An examination ofthe Peruvian pre-cerarnic radiocarbon record. American Antiquity 52(1):55-73. Rcrbertson, G.R. 1994 An Application of Scanning Electron Microscopy and Image Analysis to the Differentiation of Starch Grains in Archaeological Plant Residues. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Robins, R.P. 1983 This Widow Land: An Evaluation of Public Archaeology in Queensland using Moreton Island as a Case Study. Unpublished M.A. thesis, Department of Anthropology and Sociology, University of Queensland. 1984a The results of a preliminary archaeological survey ofMoreton Island. Queensland Archaeological Research 1:33-50. 1984b The excavation ofthree archaeological sites on Moreton Island: First Ridge and the Little Sandhills. Queensland Archaeological Research I :51-60. Ross, A. 1981 Holocene environments and prehistoric site patterning in the Victorian Mallee. Archaeology in Oceania 16(3):145-154. 1984 If there were Water: Prehistoric Settlement Patterns in the Victorian Mallee. Unpublished PhD thesis, School ofEarth Sciences, Macquarie University. 1985 Archaeological evidence for population change in the middle to late Holocene in southeastern Australia. Archaeology in Oceania 20(3):81-89. Ross, A. and S. Coghill n.d. Stewardship: Aboriginal spirituality and the Bible compared. Unpublished manuscript. Rowland, M.J. 1980 The Keppel Islands -Preliminary investigations. Australian Archaeology 11:1-17. 143 1983 Aborigines and environment in Holocene Australia: Changing paradigms. Australian Aboriginal Studies 2:62-77. 1989 Population increase, intensification or a result of preservation?: Explaining site distribution patterns on the coast of Queensland. Australian Aboriginal Studies 2:32-41. Russell, H.S. 1888 Genesis of Queensland. Sydney: Turner and Henderson. Satterthwait, L.D. 1986 Aboriginal Australian net hunting. Ma nkind 16(1):31-48. ' 1987 Socioeconomic implications of Australian Aboriginal net hunting. Man (N.S.) 22:613-636. Shaffer, B.S. 1992 Quarter-inch screening: Understanding biases in recovery of vertebrate fauna! remains. American Antiquity 57(2): 129-136. Shaffer, B.S. and J.L.J. Sanchez 1994 Comparison of 1/8"- and 1/4"- mesh recovery of controlled samples of small-to medium-sized mammals. American Antiquity 59(3):525-530. Shine, R., C.P. Ellway and E.J. Hegerl 1973 A biological survey of the Tallebudgera Creek estuary. Op erculum 3(5-6):59-83. Smith, A.D. 1992 An Archaeological Site Location and Subsistence-Settlement Analysis ofBribie Island, Southeast Queensland. Unpublished B.A.(Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Smith, J. 1994 Beaudesert Shire Archaeological Project. Unpublished paper presented to the Department of Anthropology and Sociology, University of Queensland, 2 September 1994. Smith, M.A. 1986 The antiquity of seedgrinding in central Australia. Archaeology in Oceania 21(1):29-39. 1993 Biogeography, human ecology and prehistory inthe sandridge deserts. Australian Archaeology 37:35-50. 144 Smith, M.A and N.D. Sharp 1993 Pleistocene sites inAustralia, New Guinea and Island Melanesia: Geographic and temporal structure of the archaeological record. InM.A Smith, M. Spriggs and B. Fankhauser (eds), Sahul in Review: Pleistocene Archaeology in Australia, New Guinea andIsl and Me lanesia, pp.37-59. Occasional Papers in Prehistory No. 24. Canberra: Department ofPrehistory, Research School ofPacific Studies, Australian National University. ·· Specht, R.L. 1992 Ecosystem dynamics in coastal dunes of eastern Australia. In E. van der Maarel (ed.), Dry Coastal Ecosystems, pp.512-534. Amsterdam: Elsevier. Spriggs, M. and A. Anderson ' 1993 Late colonization of East Polynesia. Antiquity 67:200-217. Steele, J.C. 1972 Th e Explorers of Moreton Bay District 1770-1830. St. Lucia: University of Queensland Press. Stephens, A W. 1992 Geological evolution and earth resources ofMoreton Bay. In O.N. Crimp (ed.), Moreton Bay inBalan ce, pp.3-24. Brisbane: Australian Littoral Society. Stockton, J. 1974 Report of an Archaeological Survey in the Vicinity ofBribie Island, South-East Queensland. Unpublished B.A (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. Stright, M.J. 1990 Archaeological sites on the North American continental shelf. In N. Lasca and J. Donahue (eds), Archaeological Geology of North America, pp.439-465. Geological Society ofAmerica, Centennial Special Volume 4. Boulder, Colorado: Geological Society of America. Stuiver, M. and T.F. Braziunas 1993 Modeling atmospheric 14C influences and 14C ages of marine samples to 10,000 BC. Radiocarbon 35(1):137-189. Stuiver, M., G.W. Pearson and T. Braziunas 1986 Radiocarbon age calibration of marine samples back to 9000 ea! yr BP. Radiocarbon 28(2B):980-1021. Stuiver, M. and P.J. Reimer 14 1993 Extended C database and revised CALIB radiocarbon calibration program. Radiocarbon 35(1):215-230. 145 Sullivan, M.E. 1982a Aboriginal Shell Middens in the Coastal Landscape of New South Wales. UnpublishedPhD thesis, Department of Archaeology and Anthropology, Faculty of Arts, Australian National University. 1982b Exploitation ofoffshore islands along the New South Wales coastline. Australian ·· Archaeology 15:8-19. Sullivan, S. 1983 Making a discovery: The finding and reporting of Aboriginal sites. In G. Connah (ed.), Australian Field Archaeology: A Guide to Te chniques, pp. 1-9. Canberra: Australian Institute of Aboriginal Studies. Tejan-Kella, M.S., D.J. Chittleborough, R.W. Fitzpatrick, C.H. Thompson, J.R. Prescott and J.T. Hutton 1990 Thermoluminescence dating of coastal sand dunes at Cooloola and North Stradbroke Island, Australia. Australian Jo urnal of Soil Research 28:465-481. Thomas, D.H. 1969 Great Basin hunting patterns: A quantitative method for treating fa una! remains. American Antiquity 3 4( 4): 3 92-401. Thomson, AK. (ed.) 1967 The Collected Works of Thomas We lsby. 2 Vols. Brisbane: Jacaranda Press. Thomson, J.M. 1975 A place fo r fishes. Proceedings of the Royal Society of Queensland 86:25-27. Twyford, K. 1994 Investigations into the Diet of Dingoes on Fraser Island: Interim Report. Unpublished report prepared for the Natural Resource Mangement (Great Sandy) Division, Central Coast Region, Queensland Department of Environment and Heritage. 1995 Investigations into the dietary ecology of dingoes on Fraser Island: Third Interim Report. Unpublished report prepared for the Natural Resource Mangement (Great Sandy) Division, Central Coast Region, Queensland Department of Environment and Heritage. Ulm, S., B. Barker, A Border, J. Hall, I. Lilley, I. McNiven, R. Neal and M. Rowland m press Pre-European coastal settlement and use of the sea: A view from Queensland. Australian Archaeology 41 (1995) Vanderwal, R.L. 1978 Adaptive technology in southeast Tasmania. Australian Archaeology 8: 107-126. 146 Veth, P. 1993a Islands in the Interior: The Dynamics of Prehistoric Adaptationswithin the Arid Zone of Australia. International Monographs inPrehistory, Archaeological Series 3. AnnArbor: International Monographs in Prehistory. 1993b The Aboriginal occupation of the Montebello Islands, northwest Australia. Australian Aboriginal Studies 2':39-50. Walker, P.L. 1986 Porotic hyperostosis in a marine dependent California Indian population. American Journal of Physical Anthropology 69:345-354. Walker, P.L. and M.J. DeNiro ' 1986 Stable nitrogen and carbon isotope ratios inbone collagen as indices of prehistoric dietary dependence on marine and terrestrial resources in southern California. American Jo urnal of Physical Anthropology 71 :51-61. Waiters, I.N. 1979 Information Loss in Zooarchaeology: A Study of Vertebrate Remains fromtwo Middens, Moreton Island, Queensland. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. 1984 Gone to the dogs: A study of bone attrition at a central Australian campsite. Ma nkind 14(5):398-400. 1985 Bone loss: One explicit quantitative guess. Current Anthropology 26(5):642-643. 1987 Another Kettle of Fish: The Prehistoric Moreton Bay Fishery. Unpublished PhD thesis, Department of Anthropology and Sociology, University of Queensland. 1989 Intensified fishery production at Moreton Bay, southeast Queensland, in the late Holocene. Antiquity 63:215-224. 1992a Farmers and their fires, fishers and their fish: Production and productivity in pre European south-east Queensland. Dialectical Anthropology 17: 167-182. 1992b Seasonality of fishing in south-east Queensland. Queensland Archaeological Research 9:29-34. 1992c Antiquity of marine fishing insouth-east Queensland. Queensland Archaeological Research 9:35-37. Waiters, 1., P. Lauer, A. Nolan, G. Dillon and M. Aird 1987 Hope Island: Salvage excavation of a Kombumerri site. Queensland Archaeological Research 4:80-95. 147 Ward, G.K. 1994 On the use of radiometric determinations to 'date' archaeological events. Austalian Aboriginal Studies 2: I 06-109. Ward, W.T. 1986 Coastal evolution during the last million, the last 10,000, and the last 200 years insoutheast Queensland. In A V. 'Arakel (ed.), Proceedings of a Short Course on Geological and Environmental Aspects of Coastal Ma nagement Programs, pp.l72-176. Brisbane: Queensland Institute of Technology. Ward, W.T. and J.L.F. Hacker 1981 Sedimentation and soil development at Brisbane Airport. Unpublished C.S.I.R.O. report. 1982 Sedimentation and soil development at Brisbane Airport. Sedimentology Newsletter 11:54. Australasian Sedimentologists Group, Geological Society of Australia. Ward, W.T., A.W. Stephens and N. Mclntyre 1977 Brisbane's north coast and Fraser Island from the air. In R.W. Day (ed.), 1977 FieldCon ference, LadyEll iottJsland-F raser Jsland-Gayndah-Biggenden, pp. l4- 30. Brisbane: Geological Society of Australia, Queensland Branch. Waselkov, G.A. 1987 Shellfish gathering and shell midden archaeology. In M.B. Schiffer (ed.), Advances in Archaeological Me thod and Theory 10:93-210. New York: Academic Press. Watkins, G. 1891 Notes on the Aboriginals ofStradbroke and Moreton Islands. Proceedings of the Royal Society of Queensland 8:40-5 1. Welsby, T. 1917 Recollections ofthe natives ofMoreton Bay. Historical Society of Queensland Jo urnal i:II0-129. Whalley, P. 1987 An Introduction to the Aboriginal Social History of Moreton Bay, South-East Queensland from 1799 to 1880. Unpublished B.A. (Hons) thesis, Department of Anthropology and Sociology, University of Queensland. White, J.P. 1994 Australia: The different continent. InG. Burenhult (ed.), People of the Stone Age: Hunter-Gatherers and Early Farmers, pp.206-225. lllustrated History of Humankind Vol. 2. St Lucia: University of Queensland Press. 148 White, J.P. and J.F. O'Connell 1982 A Prehistory of Australia, New Guinea and Sahul. North Ryde: Academic Press. Williams, E. 1987 Complex hunter-gatherers: A view fromAustralia. Antiquity 61:310-321. 1988 Complex Hunter-Gatherers: A Late Holocene Example from Temperate Australia. Oxford: British Archaeological Reports, International Series 423. Wi llmot, W. 1992 Rocks andLandscapes of the Gold Coast Hinterland: Geology andExcursions in the Gold Coast and Albert and Beaudesert Shires. Brisbane: Geological Society of Australia, Queensland Division. Winterbotham, L.P. 1957 Some native customs and beliefs ofthe Jinibara and neighbouring tribes on the Brisbane and Stanley Rivers, Queensland. Unpublished Manuscript, Fryer Library, University of Queensland. Wood, P .A and H. King 1974 AHolocene molluscan fauna from Maroochydore, Queensland. Memoirs of the QueenslandMuseum 17(1):69-71. Woodroffe, C.D., J. Chappell and B.G. Thorn 1988 Shell middens in the context of estuarine development, South Alligator River, NorthernTerr itory.Archaeology in Oceania 23(3):95-103. Young, P.C. and H. Kirkman 1975 The seagrass communities ofMoreton Bay, Queensland. Aquatic Botany 1:191- 202. Young, R.W., Nanson, G. C. and E.A Bryant 1986 Alluvial chronology for coastal New South Wales: Climatic controls or random erosional events. Search 17(10-12):270-272.