Journal of Anthropological Archaeology 26 (2007) 283–298

www.elsevier.com/locate/jaa

Hunter-gatherer burials and the creation of persistent places in southeastern Australia

Judith Littleton ¤, Harry Allen

Department of Anthropology, University of Auckland, Private Mail Bag 92019, Auckland, New Zealand

Received 10 May 2006; revision received 9 November 2006Available online 23 January 2007

Abstract

One of the diYculties in interpreting hunter-gatherer mortuary practices is that many mortuary theories are derivedfrom sedentary societies and rely upon an excavated record. This paper is an analysis of both historical and archaeologicalevidence of Aboriginal burial practices in the Murray River region of southeastern Australia. The archaeological data reliesprimarily upon analysis of burials exposed through erosion rather than systematic excavations which limits the range ofburial characteristics that may be recorded and interpreted. The mortuary practices identiWed are highly patterned butregionally and locally variable. It is argued that the evidence demonstrates the persistence of place for Aboriginal people.The existence of persistent places is further related to a potentially Xuid but structured connection between people and land.© 2006 Elsevier Inc. All rights reserved.

Keywords: Mortuary practices; Cemeteries; Hunter-gatherer burials; Australia; Murray Valley; Time-perspectivism; Aboriginal burials

Introduction

Burials are one of the few tangible remains left byhunter-gatherers and a prominent feature of thearchaeological record of the Murray Valley, south-eastern Australia. Studies of human burials in theMurray Valley, and in Australia generally, are car-ried out to reconstruct demography (Webb, 1987),to study the incidence of disease (Webb, 1995), andto study pre-historic behaviour in terms of mortuarypractices and what these can tell us about pre-con-tact ideologies (Pardoe, 1988). To a certain extent,archaeologists have assumed that it is possible to

* Corresponding author. Fax: +64 9 373 7441.E-mail address: j.littleton@auckland.ac.nz (J. Littleton).

0278-4165/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.jaa.2006.11.004

read prehistoric behaviour directly from the archae-ological record and, second, to link the recorddirectly with ethnographic observations of Aborigi-nal mortuary customs and beliefs concerning deathand the afterlife (Meehan, 1971).

One interpretation of Aboriginal burial practicesin the Murray Valley has been that numbers of buri-als in close physical proximity represent cemeteries(Pardoe, 1988). This was further tied to the wellknown set of mortuary theories linking the existenceof cemeteries with corporate control over resourcesand territorialism, following Goldstein (1981), Saxe(Saxe and Gall, 1977) and others. In this formula-tion a cemetery is more than a number of burialslocated together, it is a piece of land set aside for aparticular purpose—a social practice guided by a setof beliefs (Goldstein, 1981). Within the context of

284 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

the general understanding of hunting and gatheringsociety, and Australian Aboriginal society in partic-ular, cemetery behaviour is regarded as unusual(Pardoe, 1988; Woodburn, 1982). In this examina-tion of Murray Valley burial practices it is arguedthat an association between cemeteries and socialcomplexity outside Australia has been projectedonto the Australian evidence (Lourandos, 1997, pp.232–235). A closer examination of the variation inmortuary behaviours demonstrates that what wereinterpreted as cemeteries could more usefully beconsidered persistent places (Schlanger, 1992) ratherthan a single identiWable site type.

Cemeteries and territorialism cannot be directlyidentiWed from the number of burials set near eachother as there are a number of intermediary pro-cesses which must be taken into account before theconclusion of a cemetery can be drawn. The chief ofthese are taphonomic—the processes of erosion,decay, preservation, redeposition and recoverywhich can, at any one point in time, reveal or hideburials from the archaeologist’s view.

Allied to these are aspects of contemporarybehaviour which may bring individual Aboriginalburials into sight with varying degrees of fre-quency. Some of these relate to natural processesof erosion and dune mobility, for example, over-grazing and the reduction of vegetation. Others,however, are fully humanly informed: the locationof settlements, roads, irrigation canals, sand min-ing ventures, even archaeological survey projectsand the presence of a researcher (Bonhomme,1990). Each and all of these processes introducebias into the archaeological record and aVect theprobability that numbers of burials will beobserved in some places and not in others. Theway around these problems is to constructresearch designs and measures that take these fac-tors into account before conclusions concerningAboriginal behaviour are drawn.

Neither is it then possible to work directly fromthe record as if behaviour is then a simple empiricalfact. In between observations of burials, proximity,frequency, density, age, sex, etc. and the inferences ofbehaviour a set of concepts should be inserted thatorganise and make sense of these observations. Inthe following, the question of cemetery behaviourwill be examined in two adjacent but contrastingenvironments, using spatial, temporal, and behav-ioural inferences as organizing principles.

Time here is used in a number of ways using Bai-ley’s (1983) distinction between short and long termprocesses and spans (Table 1). While an individualburial may be a relatively transitory event, the mem-ory of the burial and any visible monuments mayendure over a longer span (e.g., Draper cited inWandsnider, 1992, p. 264). The potential then existsfor this longer span to stimulate further activities butthat link needs to be derived from the archaeologicalrecord, rather than assumed on the basis of proximity.

Interpreting such burials therefore becomes amatter of trying to extract diVerent levels of infor-mation represented by the archaeological remains(e.g., Olivier, 1999). It is only once the compositionof the record is understood that interpretation ofmeaning can follow. Yet burials because they reXectboth a point in time and a set of long-term activitiesdo give a diVerent lens onto the past (Carr, 1995). Inthis paper, both the archaeological record and theethnohistoric record of burials will be analysed andthe argument made that, despite current site invisi-bility, burials formed a signiWcant part of the land-scape of Aboriginal people and that the creation insome places of mortuary landscapes helped to struc-ture particular uses of the land.

Background

The particular nature of the Murray River needsto be explained Wrst. The Murray Basin extends over

Table 1Concepts and measures of time involved in the interpretation of burials

Process Span

Short-term The recording of the burial c.1 dayThe duration of the disposal event < 1 generationThe burial in relation to the seasonal round < 1 generation (1 year?)The death in the life cycle of an individual 1 generation

Unknown The duration of the visible monument > 1 generation?Burial exposure Potentially cyclical

Long-term The time of burial in relation to landscape formation 1 generation+The time of burials in relation to each other Simultaneous to 2000+ years

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 285

300,000 km2 of inland southeastern Australia(Fig. 1; Brown and Stephenson, 1991). The basincomprises the southern half of the Murray-DarlingRiver systems. Within this area, the Murray River,and tributaries, drains from the Great DividingRange in the east, across the plains of central andwestern NSW and north Victoria into South Austra-lia where it turns southwards towards the SouthernOcean.

In its middle and lower reaches the area theserivers traverse is largely semi-arid: evaporationexceeds rainfall and the average variability in rain-fall is 33% meaning cycles of droughts and Xoods.The rivers supply an important source of extra-rainfall water originating from the winter snow-melt in the ranges.

The area discussed in this paper is the centralregion of the Murray River. In the east of the regionlies the Hay Plain traversed by the Lachlan, Mur-rumbidgee and Murray Rivers and their tributariesand anabranches. In the western half the single rivercourse of the Murray narrows and is conWned as ittravels in a narrow Xood plain surrounded by aeo-lian landforms. In this paper, these two areas will be

referred to as the Hay Plain and Murray corridorand collectively as the central Murray.

The accounts of early explorers suggest that thiswas an area of relatively high population densitypermitted by the predictable riverine resources(Bickford, 1966). Birdsell’s (1953) study of Aborigi-nal population density and distribution along withTindale’s (1974) reconstruction of tribal boundariessupported this early impression. Tribal areas weremuch smaller within the Murray Basin than in thesurrounding areas which were less well suppliedwith water. Hercus’ work on languages (Hercus,1969, 1971, 1978; Hercus and White, 1971) supportsthis notion of population diversity and Pardoe(1993a) has argued on the basis of skeletal non-met-rics that such linguistic diversity is matched by highlevels of biological variation between populations.More recently archaeological work to the south ofthis region has suggested marked economic intensiW-cation during the mid-late Holocene (although thishas been hotly disputed) (Lourandos, 1983, 1997).

Given these indications of cultural, linguistic andbiological diversity, Pardoe (2003) has proposed amodel of Holocene occupation within this area of

Fig. 1. Map of Southeastern Australia showing the location of the Hay Plain and Murray corridor and burial locations with more than 10burials.

N

kilometres

0 200

MURRAYCORRIDOR

HAYPLAIN

LOWER MURRAY

LOWERDARLING

MURRAYMOUTH

HEADWATERS

Roonka

Katarapko

LakeVictoria

LindsayIsland

Swanport

SnaggyBend

Tucker's Creek

Euston

Robinvale

Nap Nap

ToogimbiePoon Boon

Coobool

Thule CreekKow

Swamp

Lake Tandou D

arling

River

Lachlan River

Murrumbidgee River

River M urray

286 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

the basin based on evidence of burials, human biol-ogy, ecological patterning and historical documen-tation. The model suggests that the Central Murrayrepresented a predictable, stable and productive riv-erine environment which could support large anddense populations. Given this high population den-sity he argues that the area was occupied by highlyterritorial, endogamous groups with well-defendedboundaries. The biological outcome of this is anincreased genetic distance between neighbouringgroups and he suggests that this was reXected in amortuary practice that included cemeteries as visualsymbols of land ownership (Pardoe, 1988). Theseprocesses operate in contrast to surrounding groupsof the arid and semi-arid country who occupied thecountry at low density and formed exogamous pop-ulations with large inclusive social networks, themortuary corollary of this being individual isolatedburials. In light of the argument suggested abovethat burials have several diVerent relationships withtime, this paper will re-examine the mortuary evi-dence from the region suggesting that large burialsites need to be seen in two lights: as part of a rangeof mortuary behaviour and as features constructedboth over and in time.

Ethnohistoric accounts of burial

There are ethnohistoric accounts of burials in thisregion but their completeness is unknown. Infec-tious disease from the Europeans, particularlysmallpox, arrived in the area before any direct phys-ical contact between the colonists and Aboriginalpeople (Dowling, 1997; Mein, 1897) Aboriginalaccounts collected by Europeans after settlementdescribe disease (probably smallpox) as coming witha great wind from various directions (from thesouth, up the river, from the east).

The historic accounts, therefore, need to be inter-preted in the awareness that the Aboriginal peoplealive at the time are the survivors of serious epidem-ics and that the impact of infectious disease mayhave been quite uneven so that the distribution ofpeople in the landscape does not necessarily repre-sent the pre-contact pattern.

There are two types of historic records: text basedrecords of visible burial remains interpreted throughEuropean eyes and accounts of actual burials whichmay or may not follow pre-European norms.

Early accounts do note regional variation. Forexample, Mitchell (1839) noticed a diVerence bothin style and location of the graves. Near the Mur-

rumbidgee–Murray junction, graves wereobserved:

Several graves all enclosed in separate parterresof exactly the same remarkable double or tripleridges, as those formerly seen of the lower part ofthe Lachlan. There were three of these parterresall lying due east and west. On oneƒthe ashes ofa hut appeared over the grave. On another, whichcontained two graves, (one of a small child) logsof wood, mixed with long grass, were neatly piled,transversely; and in the third, which was soancient that the enclosing ridges were barely visi-ble, the grave had sunk into a grassy hollow. Iunderstood from the widow that such tombs weremade for men and boys only, and that the ashesover the most recent one where the remains of thehut, which had been burnt and abandoned, afterthe murder of the personƒhad been avenged(Mitchell 1839(II): 87–88).

Further west, he wrote:

We continued our journey, and soon found allthe usual features of the Darling; the hills of softred sand near the river, covered with the samekind of shrubs seen so much higher up. Thegraves no longer had any resemblance to those onthe Murrumbidgee and Murray but they wereprecisely similar to the places of interment we hadseen on the Darlingƒ (Mitchell 1839(II): 113).

The accounts of burial sites have several commonfeatures despite the observations of regional vari-ants. They are most commonly accounts of up tothree burials, visibly mounded, surrounded by aboundary of some type (generally ridges), and oftensurmounted with a hut, or other structures.

The regional variations apply to location of thegraves (high ground along the Darling and Murraycorridor but not necessarily on the Hay Plain) andto the form of the visible monument (hut vs wood vsWshing net and logs). These were visible monumentsbut such monuments decayed over time and in someinstances their decay was assisted for example byburning down the hut which in the case mentionedabove marked the Wnal stage of mourning, theavenging of the death.

The other visible markers of graves were scarredor carved trees particularly east of the RiverinePlain (Bonhomme, 1990) and artefacts associatedexplicitly with burials: widows caps of gyspum plas-ter, kopi eggs (spindle shaped gypsum plaster)placed near the grave These are particularly

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 287

associated with Barkindji people from the lowerDarling River (Matthews, 1911).

Accounts of actual burials are much less com-mon. Hobler, for instance, describes what he consid-ers a rudimentary burial on the MurrumbidgeeRiver:

the blacks do not always pay so much respect tothe dead as we supposed—I heard lately of theirturning the bones out of an old grave to makeroom for another body—and a few days ago theyopened an old grave at the end of the garden andtucked in the body of an old man so quickly thatwe were not aware of it till the stench found thefact—I gave them credit for acting the part ofScots Old Mortality as they observed to havepacked the bark upon the grave more carefullyand scraped away the weeds for a few yardsaround it, but it seems it were from lazinessonly—they made the sunken grave hold anothertenant. (Hobler, 1992: 7 March, 1847).

In contrast, Musgrave (1930) wrote an extensivedescription of a burial practice whereby the bodiesof children who had died were carried by theirmothers until an appropriate adult male had died atwhich time they were buried with him.

The ethnohistoric accounts point to the visibilityof at least some graves, but also at the potentiallyshort term nature of such visibility, i.e., visibility interms of a generation not over hundreds of years.Clustering of some graves is noted but again men-tion is only of small numbers. Aboriginal people areclearly aware of where burials occur and severalwriters mention both avoidance of burial locationsfor camping and avoidance of the names of the dead(Mereweather, 1859, p. 125). Regional variation isnoted in placement of the body, location of thegraves, and speciWc grave form (KreVt, 1866) butthere is an overall adherence to both primary burial(except for children) and rules that relate in varyingcircumstances to clan (Robinson cited in Clark,1990), age and sex (Littleton, 1998a). The diYculty isassessing the relationship, if any, between theseaccounts which incorporate short term events at thetime of death and the long term archaeological accu-mulation of burials in a cemetery which might spanhundreds and possibly thousands of years.

Archaeological work

There have been a number of amateur excava-tions of burial sites in this region starting with

Cunningham (1817). Murray Black excavated nineburial grounds in the Euston-Lake Bennanee-Rob-invale Area (Sunderland and Ray, 1959), collectingmore than 400 individuals. He excavated a furtherseries of sites in the Lake Victoria-Rufus River-Lindsay Island area, collecting 570 individuals. Themajority of these have now been repatriated andreburied.

In contrast there have been relatively few profes-sional excavations. Of sites with large numbers ofburial, only three have been excavated: a site nearRobinvale potentially aVected by sandmining wasexcavated in 1983 (Bowdler, 1983), the burial site asKow Swamp was excavated in the 1970s (Thorneand Macumber, 1972), and test excavations at LakeVictoria (Littleton et al., 1994).

There have, however, been major surveys of burialsites undertaken in consultation with Aboriginalcommunities documenting graves exposed througherosion. Pardoe has focused on the Murray corridor/lower Darling region (Pardoe, 1985, 1988, 1993a),Littleton on the Hay Plain (Littleton, 1997, 1998a,b,2002), and Bonhomme undertook a general survey ofNational Parks and Wildlife Service sites (Bonho-mme, 1990). It is the results from these surveys thatnow provide most of the data on site location andvariation over space while the excavations provideinformation about particular points in space.

Taphonomy

The results of both surveys and excavations needto be interpreted in the context of the conditionsunderlying either the need for their excavation(sandmining, water erosion) or the reason for theirinitial visibility.

Throughout western New South Wales Aborigi-nal burials are exposed through wind and water ero-sion. Developments, particularly sand quarrying,overstocking, and maintenance of higher water lev-els all play a major role in exposure but even prior tothis burials were coming to the surface. The relativeshallowness of burials and preferential burial insand bodies rather than heavier soils explain why somany are currently exposed in western NSW(Johnston and Littleton, 1993). Soil movementsmean that new burials are appearing all the time.This has been very clearly demonstrated at LakeVictoria, in particular at Gecko Island, Lake Victo-ria (Fig. 1), where a site initially subjected to watererosion and then exposed to wind for two monthswas documented in detail.

288 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

Gecko Island is part of a series of dunes lining thesouthern end of the lake near its inlet and outletcreeks. The western end of the island is a broad windand water scoured platform of grey sandy clay withsome carbonate, and numerous burials. This is theprincipal unit in which burials, mapped as in situ,visible pit or scatter, and artefacts are located. Aminimum of 111 individual burials (43 in situ, 59pits, 9 scatters) were recorded (Littleton et al., 1994).

The number of exposed burials, however, bearslittle relationship to the potential total number ofburials. As seen in Fig. 2a, below the 25.8 m contour

burial pits predominated, while above this contourintact burials were more common. On the higher,less eroded surface there is the potential for morein situ burials to lie beneath the current ground sur-face. Burial pits below this layer, however, representthe lowest level of burials with upper layers proba-bly already lost through erosion.

The burials on Gecko Island were recorded andmapped at two diVerent times: in April, 1994, whenWrst observed, and six weeks later in June after thesite had been visited many times and after severewindstorms in late May. The diVerence in terms of

Fig. 2. Site plan of Gecko Island showing the diVerence in the number and nature of burials recorded in (a) April and (b) June 1994. Dark-ened symbols indicate remains which in (a) disappeared before June and in (b) appeared in June only.

metres

0 20

N

metres

0 20

N

APRIL

JUNE

Burial pit

Scatter

In situ burial

25.2

5 m 25.50 m

25.50 m

25.75 m

25.75 m26.00 m

26.25 m26.50 m

Burial pit

Scatter

In situ burial

25.50 m25.75 m

26.00 m

26.25 m26.50 m

25.2

5 m 25.50 m

25.75 m

a

b

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 289

results is indicated in Fig. 2. Of the records made inApril twenty-Wve burials could not be relocated inJune. The most obvious reason for loss was wind-blown sand which had moved across the siteparticularly on the eastern side. In June Wfty newrecordings were made. The burial of a single adult inApril was cleared of approximately Wve more centime-tres of sand revealing the burial of Wve closely placedbodies. Several processes were clearly in operation:wind erosion and the removal of sand; the obscuring ofmaterial by drifts of windblown sand; and the decay ofpoorly preserved burials, both by exposure to the ele-ments and damage by human traYc.

What the recording on Gecko demonstrated wassimply the magnitude of diVerence a short period oftime such as two months could make to a localeindicating that burials need to be interpreted interms of their ephemeral visibility. In addition, whentrying to assess absolute numbers of burials it is nec-essary to take into account factors such as stratigra-phy, degree of preservation and patterns of erosion.This is as true for excavations as for survey data.

Apart from the cycle of processes occurring atone site, it has also become apparent that there arelonger-term cycles aVecting site visibility. During1993, a severe drought on the Hay Plain causedmany more burials to become exposed. At the heightof the drought the regional archaeologist was receiv-ing approximately one report of exposed burialsevery week (Johnston, 1993 pers. com). Around Hayanecdotal evidence is that comparable conditionshave only been seen in the late 1940s/early 1950s. Isit possible that many burials recorded in the inter-vening years were Wrst exposed in this earlierdrought? The question has implications for estima-tions of how many burials there are in an area. MoresigniWcantly, however, it conWrms that the burialrecord is not a static representation of the past but arecord of what is visible as moments of an erosioncycle. Hence, the information inherent in burialsneeds to be interpreted in terms of recurrent pattern-ing and relationships between particular site ele-ments, rather than as a total or proportional sample.

Findings

Given this limitation due to site taphonomy char-acteristics of burial sites need to be interpreted care-fully. As described historically, there are apparentdiVerences in burial practices across this regiondespite some persistent common features. The fol-lowing focuses upon a few characteristics of burial

sites from the region: location, numbers of burialsper locale, density, burial form, demographic repre-sentation, body position.

Location

The majority of burials in the central Murrayhave been identiWed in source-bordering dunes orlunettes near permanent lakes or rivers (Table 2).Pardoe (1993b) suggests that there is a distinctinverse relationship between areas of occupationdebris and large burial sites.

On the Hay plain there is a much greater degree ofvariability in burial location (Littleton, 1998b). Ratherthan being concentrated in on the major permanentwater sources, burials are distributed more widelyacross the landscape being associated at times withephemeral water courses and ancestral stream deposits(Fig. 1). A high proportion of the burial locales arewithin cultural deposits: 2% in middens and 56%within oven mounds. The burials appear to postdatethe living use of these locations and there is anunknown temporal gap between the two events. It is,however, the accumulation of a settlement depositwhich then creates a potential location for a burial.

In addition, unlike many of the locales with mul-tiple burials in the Murray corridor, on the HayPlain, multiple burials are not necessarily placedbeside or in easily recognisable locations (e.g., lakeinlets) or on high points of the landscape. Suchburial locations are less overt on the landscapealthough researchers have found that with experi-ence it is relatively easy to predict potential buriallocations which are often not the most highly visiblepoints but lower sand bodies (Littleton, 2002).

Numbers per site

Excavations at Lake Victoria (two sites), Robin-vale and Kow Swamp have demonstrated that

Table 2Distribution of burial locales (%) by landform

Hay Plain Murray corridor

Claypan 13.2 0Levee 3.6 0Mound 55.7 0Midden 1.8 0Riverbank 0.6 0Residual 1.8 0River Terrace 0 6.8Dune 21.6 93.2Total N 167 44

290 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

there may be a wide disparity between the numberof burials visible in a surface exposure and thenumber actually within the site (Littleton, 1998b).Hence data on the number of burials per locale istenuous, representing the minimum number possi-ble and partially depends upon inferences basedupon surface density and maximum size and depthof the deposit.

SigniWcantly, however, there are relatively fewindividual burials identiWed in the Murray corridor(Fig. 3). Even within lunettes more than one burial isfrequently found within a radius of less than 30 m.Twenty recorded locales (47.6%) contain between 2and 10 individuals with nearly a quarter of localescontaining more than 20 exposed burials (Littleton,1998b).

In contrast, as a proportion of all burial locales,in the Hay Plain 39.5% are individual graves notassociated with other burials, while a further 53%are burials of less than 10. Large numbers of graveswithin a small area are much rarer.

Taphonomy could create a potential bias in thisrecord. The Murray corridor burials being largely insand bodies could be more subject to erosion, expo-sure and decay than bodies in the heavier soils oroccupation deposits of the Hay Plain. This couldmean that the smaller number of burials on the HayPlain is due to a lack of erosion. However, these soilsand deposits have been subject to much more exten-sive sand-mining and cultivation which should haveexposed large accumulations of grave if they werepresent.

The diVerence in numbers buried per locale ties inwith the diVerence in burial location. In the Murraycorridor it appears that potential burial locations

are restricted to two particular landforms causingburials to accumulate even by chance, on the HayPlain the choice of potential burial places is muchwider.

Density

One way of distinguishing this is to look at thedensity of burials within the deposit (Table 3). Thedata are based upon exposed burials and are calcu-lated only for locales with three or more burials bycalculating the area from the diameter of the twomost distant graves. In the Hay Plain, locales withthree or more burials have an average density of 8.3burials per 100 square metres. Burial locations com-prise small numbers of individual interments andsingle inhumations, but at the same time, burials oflarger groups of people (up to 30 at least) occur intightly packed locations, and multiple clusters ofburials may coexist in one area (Fig. 4). Clustering,therefore, is a deliberate act not a random place-ment. In the Murray corridor in contrast, burial sitesare often larger, more restricted in site locations butalso locally less dense.

Table 3Density of burials in sites containing 3+ bodies

Hay Plain Murray Corridor

N 63 19Number of peopleAverage 6.6 23.4Standard deviation 5.4 27.5Density per 100 m2

Average 8.3 1.9Standard deviation 24.4 2.3

0

10

20

30

40

50

60

1 2 - 10 11-20 21-50 50+

Number of burials per site

% o

f Si

tes

Hay Plain

Murray Corridor

Fig. 3. Distribution of the number of burials per locale in the Murray Corridor and on the Hay Plain (%).

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 291

Burial form

Primary burial is the most common form of burialthroughout the central Murray region. However, theevidence for secondary burial is hard to identify. Onevisible sign of secondary processing is the bundling ofthe body where remains were exposed or carried priorto their Wnal interment. This is visible in the combina-tion of disarticulation with tight Xexure or stacking ofindividual elements. Only 3% of burials recorded inthe Murray corridor are bundles (Table 4). A higherpercentage have been found on the Hay Plain (9.0%)

In the case of children these bundles are placed withan adult burial. Such bundles are found within clus-ters of burials in both areas suggesting that the bodyhas been returned to a speciWc location or kept till aparticular death as in Musgrave’s (1930) descriptionof child burials.

A small proportion of individuals were cremated.In the Murray corridor 2.9% of all burials were cre-mations compared to 8.3% on the Hay Plain (Table4). Distinctively, of the 42 cremations on the HayPlain nine were children and all four of the adultsfor whom sex could be identiWed were female. Of the

Fig. 4. Map of one location (Toogimbie) on the Hay Plain showing a number of burial clusters.

CHANNELLED FLOOD PLAIN

SAND PLAIN

CEMETERY

STURT HIGHWAY

Oven Mound

1 burial

2 burials

3-10 burials

21+ burials

11-20 burials

Murrumbidgee R. N

0 1 2 3 km

292 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

22 cremations in the Murray corridor two werefemales (the remainder could not be sexed) and onlyone of the 22 was a child.

The type of cremation and its subsequent treatmentvaried. In some cases the body was burnt in situ andeither left uncovered or only lightly covered. In the HayPlain the cremated remains particularly of childrenwere interred with the primary burial of an adult sug-gesting that cremation was part of delaying a burial. Afurther small subset are instances where the remainswere burnt, then they were deliberately broken intoeven smaller fragments, bundled with ochre and thenburied. There are three identiWable cases of this prac-tice: two in the Murray corridor (Littleton et al., 1994;Pardoe, 1993b) and one on the Hay Plain (Littleton,2002). All involve a young woman’s remains. This isthe same treatment given to the burial of Mungo Idated to 24500 BP (Thorne, 1975) suggesting a longhistory to this burial form. What cannot be assumed isa continuity of the symbolism. Meehan (1971) docu-ments local variations in beliefs concerning death andthat most areas had multiple forms of burial, includingcremation. Hence there is no simple associationbetween burial form and ideology.

If cremation and bundle burials are a way ofdelaying burials then such practices are more com-mon on the Hay Plain than in the Murray corridorand occur despite the ubiquity of possible buriallocations in the landscape. Delay may be of particu-lar signiWcance when children are involved.

Demographic representation

Despite this diVerential treatment of childrenthere are few diVerences across the region in termsof age and sex representation. In both the Murraycorridor and Hay Plain males are more frequentlyidentiWed than females (Table 5). On the Hay Plainin particular males form a higher proportion of thesingle isolated burials, women being more fre-

Table 4Body position and burial type

N, number for which that characteristic is observable.

Hay Plain Murray Corridor

% N % N

Cremation 8.3 496 2.9 753Inhumation 91.7 496 97.1 753

Extended 19.4 144 53.2 297Flexed 71.5 144 43.8 297

Bundle burial 9.0 144 3.0 297

quently buried within groups. There is, however,also an inexplicable dearth of middle-aged femalesin clusters: young women and old women predomi-nate (Littleton, 1998a). This apparent division is notobserved in the Murray corridor.

Burials of subadults (less than 15 years) are morefrequently observed on the Hay plain than in theMurray corridor. In both areas, however, such buri-als are predominately found with other burials. Ofthe subadults however, the overall proportion isvery low (lower than expected if there is a demo-graphic cross-section) and the burials of very youngchildren in particular are underrepresented. Thismay be the result of taphonomic basis.

Body positioning

Body orientation and position are two aspects ofmortuary behaviour that do remain visible andaccording to the ethnohistorical evidence were partof the customary practices surrounding the dead(Robinson cited in Clark, 1990).

In the Murray corridor there is a fairly clear pref-erence for bodies to be laid in an extended position(Table 4) with the head pointing towards to south-west (Fig. 5). There are, however, some signiWcantdiVerences between sites. At Lake Victoria in theEast Nanyah burial location bodies are primarily inthis orientation and laid in extended positions, yetacross the creek most burials are oriented to thesouth and in a Xexed position suggesting highly pat-terned behaviour within particularly discrete loca-tions (Littleton et al., 1994).

This, however, is much clearer on the Hay Plainwhere, rather than there being one predominant ori-entation, direction tends to vary by cluster so that,while within a group of burials there is a fairly consis-tent orientation and position, orientations do nothold across groups in the same locale (Littleton,2002). In this area various Xexed positions rather thanextended positions predominate (Table 5) giving anoverall appearance of heterogeneity between sites buthomogeneity within sites.

Table 5Age and sex representation among observable burials

Murray Corridor Hay Plain

Sexed Adults 529 449Females % 40.9 40.1Males % 59.1 59.9Aged Individuals 753 506Subadults% 10.8 18.0

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 293

Dating

Very few burials have been radiocarbon dated andeven less frequently has more than one burial from asite been dated. Known dates are given in Table 6.The majority of graves date from the mid- to lateHolocene with a particular increase around 6000 BPand a fairly constant distribution after that date(Holdaway et al., 2005). Furthermore, however, theyindicate that a single site (such as Katarapko) mayencapsulate a large amount of calendrical time. Thisis also suggested by sites such as Nap Nap wherethere are marked diVerences in the layers in whichbodies are buried (some are very close to the surface,others in the dune core) and where this correspondsto the degree of mineralization of the bone indicativeof a large amount of geomorphological time. Yet,despite these indications that multiple burials mayaccumulate large amounts of time, there are someindications of relatively short time gaps betweenpaired burials (e.g., Hobler, 1992: 7 March, 1847).

Discussion

There are common features that identifyhunter-gatherer burials within this large region.The ethnohistoric record suggests that burialswere highly visible, at least for short periods of

time, were avoided for camping, and involved pat-terned placement of bodies. The archaeologicalrecord conWrms: a general avoidance of burialareas for occupation (although occupation is aprecondition for some burial locations on the HayPlain); that primary burial of all individuals pre-dominates but that there is some patterning by ageand sex; and that burial patterns tend to be forma-lised. In the Murray corridor it has been arguedthat there is a normative position and orientationbetween most sites (Pardoe, 1988), on the HayPlain such formalism is patterned in particularlocations.

Pardoe divided burials into cemeteries—groupsof burials deliberately contiguous and occurring athigh density within bounded locations—and iso-lated burials occurring at low density with little orno association between each burial event (Pardoe,1988). Cemeteries were interpreted as aYrmingmembership to a particular group and asserting forthose who bury their dead in that location theirancestral rights. In this sense isolated burials becomea null event in that they fail to symbolise belongingto a particular group.

The Hay Plain burials however complicate thispicture by demonstrating a range of burial practicesthat are, in some locations, very formalised and pat-terned, yet involve only small numbers of people

Fig. 5. Distribution of burial orientations (%) on the Hay Plain and in the Murray Corridor.

0

10

20

30

40

50

60

0 30 60 90 120 150 180 210 240 270 300 330

Orientation of burial (degrees)

% o

f bu

rial

sHay PlainMurray corridor

294 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

(Littleton, 2002). They serve to question the exis-tence of a simple boundary between cemetery andnon-cemetery burials since they have many of thecharacteristics of burials but lack the large numbersand possibly the time depth of cemeteries. EVectivelythey demonstrate that hunter-gatherer burials donot neatly fall into particular categories andattempts to create such categories tend to leave largenumbers undeWned, not simply because of the prob-lems of preservation, but also because the archaeo-logical record of burials is a long temporal series ofindividual acts. Pattern is imposed post hoc and wasnot necessarily obvious to the actors in the past.

The record is created out of several repeatedactions from which some patterns can be recreated.The majority of the burials are primary with thevery speciWc exceptions of children whose bodiesmay (but not always) be carried for long periods oftime or cremated (and then carried). Children’sbodies in this area, at least until early adolescence,are transportable and need to be transported. SarahMusgrave’s account suggests a particular set ofpractices explicitly for this age group (Musgrave,1930). The material Wnds match her account (Hope,1998; Witter et al., 1993). It suggests that burials of

children may be less tied to a particular place thanto the deaths of particular adults. Until early adoles-cence children held distinctly diVerent status toadults and that this particular practice occurredover a very wide area encompassing in historic timesat least, tribal groups not thought to be in close cul-tural contact.

For adults, however, burials seem to have beenmore tied to place. Analysis of the Hay Plain andMurray corridor burials suggests there are twounderlying characteristics of burial places: prefer-ence for particular locations and maintenance ofspeciWc locations over time. Preference is evidencedby the multiple use of highly speciWc locales overtime. Unfortunately few sites have a good sequenceof radiocarbon dates but on Katarapko Islandapproximately 2000 years separates two burials(Dowling, 1989); on Nap Nap burials vary consider-ably in which layer they are laid and later burials cutthrough earlier burials (Littleton, 1997). All of thesesuggest preference over time.

Isolated burials, while possibly more variable inlocation, also occur in particular landforms. This isnot simply a matter of easy digging—longitudinaldunes for instance are not used for burials in con-

Table 6Known dates of burials in the Central Murray

Location Date (BP) Material Lab# Source

Hay PlainCoobool Ck 14300 § 1000 U/Th LLO-416 Brown (1989)Narcurrie 11440 § 160 Bone Brown (1994)Kow Swamp 1 10070 § 250 Apatite ANU-403b Thorne (1975)Kow Swamp 9 9300 § 220 Apatite ANU-619b Thorne (1975)Kow Swamp 14 8700 § 220 Apatite ANU-1038 Thorne (1975)Barham K2/1 5400 § 90 Bone Daley (1986)Jeraly 5010 § 180 Apatite ANU-8974b Pardoe (1995)Bahram E4/1 4670 § 110 Bone Daley (1986)Kerrie East #1 4060 § 160 Characoal ANU-10213 This Study

Central MurrayMallee CliVs 6610 § 190 Apatite ANU-5787A Pardoe (1988)Keera Stn #28 5350 § 290 Collagen GaK-1431 Blackwood and Simpson (1973)Katarapko Island 4700 § 140 Bone ANU-6896 Dowling (1989)Warrananga 4140+60 Coolamon ANU-8152A Pardoe (1995)Lindsay Is. 3580 § 370 Collagen ANU-420d Gill (1973)Mallee CliVs 2430 § 260 Collagen ANU-5787B Pardoe (1988)Katarapko Is. 1560 § 200 Bone ANU-6895 Dowling (1989)Katarapko Is. 1070 § 200 Bone ANU-6894 Dowling (1989)Katarapko Is. 590 § 220 Bone ANU-6901 Dowling (1989)Katarapko Is. 350 § 200 Bone ANU-6899 Dowling (1989)Keera Stn #11 4400 § 220 Collagen GaK-1433 Blackwood and Simpson (1973)Keera Stn #10 4170 § 200 Collagen GaK-1430 Blackwood and Simpson (1973)Keera Stn #69 750 § 170 Collagen ANU-421 Blackwood and Simpson (1973)Wamba yadu 5690 § 280 Apatite ANU-8648 Pardoe (1995)David’s dune 7140 § 300 Apatite ANU-8647 Pardoe (1995)

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 295

trast to dunes resulting from nearly water bodies.From the frequency of use it appears that places inthe landscape were recognised as appropriate orinappropriate for burials. These places, however,may not always be reused despite conforming to thepattern of appropriate place.

Preference exists on two levels: Wrst, a genericpreference for particular landforms which are notnecessarily places distinct from occupation, and sec-ond, in the case of clustered burials, preference forparticular places on a landform or for particular iso-lated landforms.

Furthermore areas used for multiple burials aremaintained as burial locales. They were not subse-quently reused for occupation despite what musthave been in some instances prime locations, forexample, the foreshore dunes at Lake Victoria. Suchavoidance, however, does not necessarily indicate asymbolic setting aside of an area for current andfuture burials which is often implied by the termcemetery. The ethnographic record is clear that buri-als are marked by speciWc physical remains: huts,ridges, mounds of sticks and, around the lower Dar-ling, widows’ caps, kopi eggs. It is also apparent thatfollowing a burial people moved away. This mayhave been only a short distance but they did move.EVectively the burial structured subsequent actionsand helped to create a landscape of meaning. In aresource-rich zone such as Lake Victoria it is easilyto imagine how a place could become by a series ofburials, set aside for that activity. After time a groupmay start to assert rights or linkages to those burialsor that burial area but again the arrangement maybe post hoc rather than deliberately planned. Thismeans that burial areas may be created by a processof accumulation over time, and may in turn, bybecoming mortuary landscapes, structure humanactivity and contribute to the landscape of meaning.

Pardoe (1988) following Goldstein (1980), sug-gests that those buried within a particular area areburied in particular formal styles and that thesereXect common group membership and territorial-ity. There is no doubt that within burial clusters theform of burial may be highly repetitive These buri-als, however, occur over variable lengths of time: atKatarapko 2000 years or more. Despite the poten-tially large numbers involved along the MurrayRiver, they do not necessarily represent continuityand exclusive use by one group of a particular loca-tion. The numbers of individual burials at any onesite are too small. Even the highest estimates of totalburial numbers at Lake Victoria represent only one

burial per generation if use of the site was continu-ous (Littleton et al., 1994). Continual and exclusiveuse by one group of people for a short period of timeis unlikely given the few dates involved and strati-graphic positions within sites. It is more likely thatpeople are using multiple burial locations at onetime. This is particularly clear on the Hay Plainwhere the grouped burials are small in number.

In both areas of the Murray some burial clustersare highly formalised with groups of burials withdistinctly diVerent normative burial positions. Thisis apparent on the various foreshore dunes of LakeVictoria as well as on a smaller and much morecomplex scale at some locales on the Hay Plainwhere each individual cluster is distinctly diVerentfrom the next.

Consistent pattering like this is invisible to thosenot involved in the burial. There is no record of anyvisible signs on the surface such as head and footmarkers to indicate appropriate orientations. Laterburials which uncovered earlier ones would revealsuch evidence but there are few clear instances ofthis (Dowling, 1989). In most cases a speciWc burialposition and orientation is eVectively insider knowl-edge. A burial locale may be known to all but thespeciWcs of the burial known only to those whoshare those practices.

This suggests that there are groups sharing gen-eral locations (and ideas about what constitutes asuitable location). Unless there is a rapid succes-sion of one group over another for example, thegroup practising extended burials replaced by thegroup practicing Xexed burials, then what is occur-ring is almost the opposite of strict territorialitywith rigid boundaries. While there may have beenat the time of Wrst contact in the area an expressedideology of clan and territory (Clark, 1990), thelong term mortuary record argues for a Xexible,more Xuid and less bounded system where claimsto territory “existed at the level of the individualor individual family, the local group and thetribe,ƒ individuals could call on a variety of crite-ria to legitimise access to resources or claims toland (Allen, 1996, p. 664).” This is the long termmaterial record of a phenomenon noted by Mor-phy (1998, p. 138) that “the system linking people,ancestral beings, and places together has resultedin a very stable relationship between ancestral lawand place and a very Xuid relationship betweenactual groups of people and place.”

So rather than there being a strongly delineated“cemetery equals territory” system constant over

296 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

time, people were moving across the land. At timesthey are close to a burial ground at times they arenot. When they are not people are buried elsewhere.There may be diVerences in the importance of anindividual reXecting how far they are moved but attimes both male and female adults were buried inisolated locations.

Based on a survey of ethnographic accounts,Hofman (1986) demonstrates that many hunter-gatherer examples of common burial locationsindicate burial symbols and locations shared byunilineal descent groups but that these are not nec-essarily tied to corporate group control or toresources. In the central Murray the highly forma-lised clusters of burials are partial representations ofthe group producing the dead. These clusters aremore likely to signify a social grouping than a cor-porate land holding group.

The continuity of a particular descent group witha particular burial location cannot be assumed.Even if a group leaves an area and are eventuallyreplaced by others the landscape symbols willattract similar but new stories and designs (Morphy,1995). There are two diVerent time processes occur-ring. One is a persistence of place over long periodsof time whereby human activity is structured byexisting monuments. Two, over a shorter span in theHay Plain and possibly longer in the Murray corri-dor, is a persistence of custom within a group.

It cannot be hypothesised, however, that suchgroups have exclusive and long term rights to a par-ticular place. One, the record is too slight—there aresimply insuYcient burials to argue for long termsedentary and large populations, and second, loca-tions are used by more than one descent group (orsimilar sized organisation) either at the same time orin succession. What we may be observing in the cen-tral Murray is much more the result of a social orga-nisation that allows for Xexibility and a system thatpersists even when resources are relatively predict-able because, while they may be predictable in com-parison to arid Australia, they are still subject tohigh levels of variability.

Within this system burials may be understood interms of persistent places, cultural features which“structure the use and reuse of the larger landscape(Schlanger, 1992, p. 92).” Schlanger suggests thereare three ways in which persistent places may be cre-ated: Wrst, through the recognition of unique quali-ties whereby particular locales are suited to certainactivities; second, through remains that focus reoc-cupations once built, attracting reuse and reoccupa-

tion and structuring later activities; and third,through long term processes of occupation andrevisitation. Burials locations along the Murray Val-ley are persistent places in two of these senses. First,speciWc landforms are recognised as possible buriallocations for example, in the Murray corridorsource bordering dunes fulWll this possible function.However, as has been argued here, more speciWcallyonce a burial has occurred then its very existenceand physical marking can focus and restructure suc-cessive use serving as a marker to avoid reoccupa-tion but to site burials in that location. Hence thetwo characteristics of a place, its persistence and itsactive maintenance, create the current burial record.It may well be that such persistent places are thenused by speciWc actors to legitimate ancestral linksbut this will be only one of multiple ways in whichthe relationships between land use and land holdingwere negotiated.

Conclusion

Making use of a wider variety of spatial, temporaland behavioural inferences has allowed us to demon-strate that cemetery behaviour is not necessarilyassociated with increased territorialism as previouslyhypothesised. Neither is it necessary to go outsideour understanding of the range of hunter-gathererbehaviour to explain the existence of cemeteries. Onthe other hand, we are still far away from under-standing the speciWc historical and cultural contextsunderlying these burials. A major unknown in thisreconstruction is the absolute time span over whichburials were occurring. ConWrming our potentialreconstruction requires a more substantial datingeVort. Such projects, however, need to be undertakenin accordance with the desires of local traditionalcustodians. The analysis so far demonstrates that it isnecessary to see how the record of what is left byhunter-gatherers is created—what does the mix ofhuman action and taphonomy leave for us to read?In this reading it is very tempting to move directlyfrom physical remains to symbolic reconstructionsbased upon observations of very diVerent economicand social groupings when what is potentially beingobserved is a series of human actions which thenstructure future human behaviour.

Acknowledgments

The support and assistance of Aboriginalcommunities in southwestern New South Wales is

J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298 297

gratefully acknowledged. Thanks are also due tolandholders in the region, the Baulch and McDon-ald families, and Harvey Johnston and the NPWSoYce at Buronga. This work has been supported bythe Australian Research Council, the AustralianInstitute of Aboriginal and Torres Strait IslanderStudies and the University of Auckland. Finallythanks to Simon Holdaway and the two reviewersfor comments and to Seline McNamee for prepara-tion of the graphics.

References

Allen, H., 1996. Horde and hapu: the reiWcation of kinship andresidence in prehistoric Aboriginal and Maori settlementorganisation. In: Davidson, J., Irwin, G., Leach, B., Pawley, A.,Brown, D. (Eds.), Oceanic Culture History: Essays in Honourof Roger Green. New Zealand Journal of Archaeology Spe-cial Publication, Auckland, pp. 657–674.

Bailey, G., 1983. Concepts of time in quaternary prehistory.Annual Review of Anthropology 12, 165–192.

Bickford, A., 1966. The traditional economy of the Aborigines ofthe Murray Valley, BAHons Thesis, University of Sydney.

Birdsell, J., 1953. Some environmental and cultural factors inXu-encing the structuring of Australian Aboriginal Populations.American Naturalist 87, 171–207.

Blackwood, R., Simpson, K., 1973. Attitudes of Aboriginal skele-tons excavated in the Murray Valley Region between Milduraand Renmark, Australia. Memoirs of the National Museumof Victoria 34, 99–150.

Bonhomme, T., 1990. Report on Burials and Sandmining in theRiverine Plain. New South Wales National Parks and WildlifeService, Sydney.

Bowdler, S., 1983. Archaeological investigation of a threatenedAboriginal burial site near Robinvale, on the Murray River,Victoria. Report to Victorian Archaeological Survey, Mel-bourne.

Brown, P., 1989. The People of Coobook Creek. Research Schoolof Australian and PaciWc Studies, Canberra.

Brown, P., 1994. Skeletal remains. In: Horton, I. (Ed.), The Ency-clopaedia of Aboriginal Australia. Australian Studies Press,Canberra, pp. 90–91.

Brown, C., Stephenson, A., 1991. Geology of the Murray Basin,Southeastern Australia. Bulletin of Mineral Resources 235.Bureau of Mineral Resources, Canberra.

Carr, C., 1995. Mortuary practices: their social, philosophical-religious, circumstantial and physical determinants. Journal ofArchaeological Method and Theory 2, 105–200.

Clark, I., 1990. Aboriginal Languages and Clans: An HistoricalAtlas of Western and Central Victoria, 1800–1900. MonashPublications in Geography 37. Monash University, Melbourne.

Cunningham, A., 1817. Diary of a journey with Oxley in 1817.Mitchell Library, Sydney.

Daley, C., 1986. The Barham skeletons. MA thesis, University ofSydney.

Dowling, P., 1989. Violent Epidemics. MA thesis, AustralianNational University.

Dowling, P., 1997. A great deal of sickness. Ph.D. thesis, Austra-lian National University.

Gill, E., 1973. Geology and geomorphology of the Murray Riverregion between Mildura and Renmark, Australia. Memoirs ofthe National Museum of Victoria 34, 1–96.

Goldstein, L., 1980. Mississippian mortuary practices: a casestudy of two cemeteries in the lower Illinois Valley. ScientiWcPapers No. 4., Northwestern University Archaeological Pro-gram, Evanston, Illinois.

Goldstein, L., 1981. One-dimensional archaeology and multi-dimensional people: spatial organisation and mortuary analy-sis. In: Chapman, R., Kinnes, I., Randsborg, K. (Eds.), TheArchaeology of Death. Cambridge University Press, Cam-bridge, pp. 53–70.

Hercus, L., 1969. The Languages of Victoria: A late survey. Aus-tralian Institute of Aboriginal Studies, Canberra.

Hercus, L., 1971. Eaglehawk and crow: a Madimadi Version.Mankind 8, 137–140.

Hercus, L., 1978. A note on Narinari. Papers in Australian Lin-guistics A51, pp. 119–121.

Hercus, L., White, I., 1971. The last Madimadi Man. VictorianNaturalist 88 (1), 11–19.

Hobler, G., (1992 [1882]). The Diaries of Pioneer George Hobler.October 6, 1800–December 13, 1882, C&H ReproductionsInc., Sydney.

Hofman, J., 1986. Hunter-gatherer mortuary variability: towardan explanatory model. Ph.D. thesis, University of Tennessee,Tennessee.

Holdaway, S., Fanning, P., Littleton, J., 2005. Assessing thefrequency distribution of radiocarbon determinations from thelate Holocene of western New South Wales, Australia. Austral-asian Archaeometry Conference, 13th December, Canberra.

Hope, J., 1998. Making up stories: bias in interpretations of Aborig-inal burial practices in southern Australia. In: Casey, M., Don-lon, D., Hope, J., Welfare, S. (Eds.), RedeWning Archaeology:Feminist Perspectives. ANU Press, Canberra, pp. 239–245.

Johnston, H., Littleton, J., 1993. Report on the Burials Project,Western New South Wales. New South Wales National Parksand Wildlife Service, Sydney.

KreVt, G., 1866. On the manners and customs of the Aboriginesof the lower Murray and Darling. Transactions of the Philo-sophical Society of New South Wales, 357–374.

Littleton, J., 1997. Burials on the Hay Plain (Technical Report).Australian Institute of Aboriginal and Torres Strait IslanderStudies, Canberra.

Littleton, J., 1998a. Dying for a change: Women’s Health alongthe Murray. In: Casey, M., Donlon, D., Hope, J., Welfare, S.(Eds.), RedeWning Archaeology: Feminist Perspectives. ANUPress, Canberra, pp. 227–233.

Littleton, J., 1998b. East and west: burial practices along theMurray River. Archaeology in Oceania 34, 1–14.

Littleton, J., 2002. Mortuary behaviour on the Hay Plain: do cem-eteries exist? Archaeology in Oceania 37, 105–122.

Littleton, J., Johnston, H., Pardoe, C., 1994. The Lake VictoriaLakebed Survey. New South Wales National Parks and Wild-life Service, Sydney.

Lourandos, H., 1983. IntensiWcation: a late Pleistocene-Holocenearchaeological sequence from Southwestern Victoria. Archae-ology in Oceania 18, 73–80.

Lourandos, H., 1997. Continent of hunter-gatherers: new per-spectives in Australian prehistory. CUP, Cambridge.

Matthews, R., 1911. Some mourning customs of the AustralianAborigines. Australian Association for the Advancement ofScience 13, 445–449.

298 J. Littleton, H. Allen / Journal of Anthropological Archaeology 26 (2007) 283–298

Meehan, B., 1971. The form, distribution and antiquity of Austra-lian Aboriginal mortuary practices. MA Thesis, University ofSydney.

Mein, P., 1897. Letter to H.C. Russell. Mitchell Library, Sydney.6pp.

Mereweather, J.D.R., 1859. Diary of a working clergyman in Aus-tralia and Tasmania kept during the years 1850–1853. T.H.Boone, London.

Mitchell, T., 1839. Three expeditions into the interior of easternAustralia. T.H. Boone, London.

Morphy, H., 1995. Landscape and the reproduction of theancestral past. In: Hirsch, E., O’Hanlon, M. (Eds.), TheAnthropology of Landscape. Clarendon Press, Oxford, pp.184–209.

Morphy, H., 1998. Cultural adaptation. In: Harrison, G., Morphy,H. (Eds.), Human Adaptation. Berg, Oxford, pp. 99–150.

Musgrave, S., 1930. The Way Back. Cumberland Argus, Parram-atta.

Olivier, L., 1999. The Hochdorf ‘princely’ grave and the questionof the nature of archaeological funerary assemblages. In:Murray, T. (Ed.), Time and Archaeology. Routledge, London,pp. 109–135.

Pardoe, C., 1985. Variation in Mortuary patterning along theDarling River. Australian Institute of Aboriginal Studies,Canberra.

Pardoe, C., 1988. The cemetery as symbol. The distribution ofprehistoric Aboriginal burial grounds in southeastern Austra-lia. Archaeology in Oceania 16, 173–178.

Pardoe, C., 1993a. Ecology of river Murray peoples in the midHolocene. Paper presented at the Australasian Society forHuman Biology Annual Meeting, Adelaide.

Pardoe, C., 1993b. Wamba yadu, a later Holocene cemetery of thecentral Murray River. Archaeology in Oceania 28, 77–84.

Pardoe, C., 1995. Riverine, biological and cultural evolution insoutheastern Australia. Antiquity 69 (265), 696–714.

Pardoe, C., 2003. The Menindee Lakes: a regional archaeology.Archaeology in Oceania 57, 42–53.

Saxe, A., Gall, A., 1977. Ecological determinants of mortuarypractices: the Temuan of Malaysia. In: Wood, W. (Ed.), Cul-tural Ecological Perspectives in South-east Asia. Ohio Univer-sity Press,, Athens, Ohio, pp. 74–82.

Schlanger, S., 1992. Recognizing persistent places in Anasazi set-tlement systems. In: Rossignol, J., Wandsnider, L. (Eds.),Space, time and Archaeological Landscapes. Plenum Press,New York, pp. 91–112.

Sunderland, S., Ray, L., 1959. A note on the Murray Black collec-tion of Australian Aboriginal Skeletons. Proceedings of theRoyal Society of Victoria 71, 45–50.

Thorne, A., 1975. Kow Swamp and Lake Mungo. Ph.D. thesis,University of Sydney.

Thorne, A., Macumber, P., 1972. Discoveries of Late Pleistoceneman at Kow Swamp, Australia. Nature 238, 316–319.

Tindale, N., 1974. Aboriginal tribes of Australia. ANU Press,Canberra.

Wandsnider, L., 1992. The spatial dimension of time. In: Rossig-nol, J., Wandsnider, L. (Eds.), Space, time and ArchaeologicalLandscapes. Plenum Press, New York, pp. 257–282.

Webb, S., 1987. A palaeodemographic model of Late HoloceneCentral Murray Aboriginal Society, Australia. Human Evolu-tion 2, 385–406.

Webb, S., 1995. The Palaeopathology of Aboriginal Australians.Cambridge University Press, Cambridge.

Witter, D., Fullagar, R., Pardoe, C., 1993. The TerramungamineIncident: a double burial with grave goods near Dubbo, NewSouth Wales. Records of the Australian Museum 12 (Suppl.17), 77–89.

Woodburn, J., 1982. Social dimensions of death in four Africanhunting and gathering societies. In: Bloch, M. (Ed.), Deathand the Regeneration of Life. Cambridge University Press,Cambridge, pp. 187–211.