The world from Malarrak: depictions of South-east Asian and European subjects in rock art from the Wellington Range, Australia.

May, Sally K. ; Tacon, Paul S.C. ; Paterson, Alistair 等

Abstract: This paper investigates contact histories in northern Australia through an analysis of recent rock paintings. Around Australia Aboriginal artists have produced a unique record of their experiences of contact since the earliest encounters with South-east Asian and, later, European visitors and settlers. This rock art archive provides irreplaceable contemporary accounts of Aboriginal attitudes towards, and engagement with, foreigners on their shores. Since 2008 our team has been working to document contact period rock art in north-western and western Arnhem Land. This paper focuses on findings from a site complex known as Malarrak. It includes the most thorough analysis of contact rock art yet undertaken in this area and questions previous interpretations of subject matter and the relationship of particular paintings to historic events. Contact period rock art from Malarrak presents us with an illustrated history of international relationships in this isolated part of the world. It not only reflects the material changes brought about by outside cultural groups but also highlights the active role Aboriginal communities took in responding to these circumstances.

Introduction

Few changes would have been as dramatic and confronting as the early encounters between Indigenous groups and strangers arriving in their country after having crossed the sea. This research is concerned with the contact period and the rock art that documents this period of change from an Aboriginal perspective. We argue that Aboriginal artists have produced a unique record of their experiences of contact since the very earliest encounters (with groups such as Macassan/South-east Asian (1) fisherman, British explorers and Christian missionaries). Their art often illustrates experiences not otherwise understood from historical literature.

[FIGURE 1 OMITTED]

Surprisingly, the detailed study of contact rock art within Australia has been a recent development. Previous studies of contact rock art include work undertaken by Layton (1992), Frederick (1997, 1999), Clarke (1994; Clarke and Frederick 2006) and Roberts (2004). Since 2008 publications addressing contact rock art have also emerged from the Australian Research Council-funded project titled Picturing Change: 21st century perspectives on recent Australian rock art (May et al. 2010; Paterson 2012; Tacon et al. 2010; Tacon et al. 2012).

Outside Australia, contact rock art has been the focus of some important studies, particularly in South Africa (i.e. Ouzman 2003; Ouzman and Loubser 2000; Ouzman and Smith 2004; Smith and van Schalkwyk 2002), North America (i.e. Keyser and Klassen 2003; Klassen 1998; Klassen et al. 2000; Molyneaux 1989) and, more recently, Malaysia (Mokhtar and Tacon 2011).

This paper centres on research results from north-western Arnhem Land. More specifically, we explore recent Australian Aboriginal rock paintings of introduced subject matter at the rock shelter complex called Malarrak in north-western Arnhem Land. We examine the historical significance of these paintings, as well as their role in interpreting both South-east Asian and European contact histories.

Site overview

The Malarrak complex is located within the Wellington Range (Figure 1), the northern-most outlier of the sandstone Arnhem Land Plateau, and is bordered by the Arafura Sea to the north, the Cobourg Peninsula to the north-west and the King River to the east. The Wellington Range is home to extensive and diverse rock art, including many examples of paintings that reflect contact between local Aboriginal groups and visitors to their shores. This range covers a large geographical area and is associated with various clans. The Malarrak sites (Figure 2) are located on the Namunidjbuk estate, within the traditional country of Maung speakers, where Ronald Lamilami is the Senior Traditional Owner.

We define two overarching and overlapping phases in recent centuries during which local Aboriginal people experienced periods of cross-cultural contact: (1) Macassan (South-east Asian sailors and trepangers) and (2) European contact.

The commonly accepted date for the earliest Macassan visits is contested, as is the theory of pre-Macassan contact (for example, Berndt and Berndt 1954; Evans 1992:66; McIntosh 2004). European accounts, such as those of Matthew Flinders in 1801 (Flinders 1814), have led researchers to suggest that Macassan visits began between 1650 and 1750 (e.g. Macknight (1976), who later revised this to 1780 (Macknight 2011); Crawford (1969) suggests 1660). The interpretations of these early radiocarbon dates continue to be debated (Clarke 2000). However, recent Accelerator Mass Spectrometry (AMS) dating of a beeswax snake design directly over the top of a prau (perahu/ship) painting at the nearby Maung site of Djulirri suggests praus were present prior to at least 1664 AD, and possibly much earlier as the beeswax figure has a date range of 1517 AD to 1664 AD and a median age of 1577 AD (Tacon et al. 2010). This is further supported by findings from recent archaeological excavations and dating of human skeletal remains (of South-east Asian origin) buried near Anuru Bay, north-western Arnhem Land. For the Anuru Bay A site it is argued that one individual was buried before 1730 AD, while another could date to an earlier period (pre-1700s) (Theden-Ringl et al. 2011:45).

[FIGURE 2 OMITTED]

Macassan praus were home to ethnically diverse crews, with sailors from Sulawesi, Madura, Java, Borneo, Flores, Timor Roti and even New Guinea (e.g. see Earl 1846:240). While their main priority may have been to obtain trepang for trade with China, they were also part of wider regional trade patterns which, after 1500 AD, included Arab, Chinese, Portuguese and Spanish traders. From the seventeenth century Dutch, then British, interests dominated these trade networks and Macassan visits to Arnhem Land were largely over by the early twentieth century (Macknight 1976). Irregular Aboriginal contact with Europeans could have occurred after the sixteenth century and there are abundant historical accounts for the early nineteenth century when military outposts were established in the region. These historical records date to the early 1800s, when British outposts were established on the Cobourg Peninsula (Allen 2008). More regular contact characterised the twentieth century, when Aboriginal groups became enmeshed with the activities of buffalo and crocodile hunters, missionaries, traders, pearlers, explorers, scientists, soldiers and others.

Given that many diverse interactions have taken place, we may expect considerable archaeological evidence. This appears to be the case at Malarrak, a site complex that comprises many rock shelters, four of which were analysed for this study. While these are not all directly adjoining shelters, they are all within a one kilometre radius. The paintings here reveal elements of introduced material culture, including a South-east Asian prau, a knife in its sheath, European watercraft, smoking equipment, a building, firearms, horned animals and even a drinking mug. We are aware that a focus on only introduced subject matter is a limited way of understanding any site complex; however, it is an important stage in the overall comparative analysis of this and other sites in the region. It is also important to remember that other rock art was being produced in the contact period, including more traditional subject matter such as kangaroos and fish (May et al. 2010).

Methodology and results

During the dry season of 2008, the Malarrak complex (incorporating shelters WR011, WR012, WR013 and WR014) was recorded with the approval and assistance of Ronald Lamilami and his family, including two of his sons, Patrick and Leonard Lamilami. During the recording process the Lamilami family provided ethnographic information relating to the shelter and, where possible, the individual paintings.

The recording of Malarrak involved compiling a detailed inventory of the art. Each image was allocated a unique number, described in detail, and extensively photographed with and without scales. Descriptions included interpretation of subject matter, dimensions, technique, colour/s, form and style. All images within the shelter were recorded, not only the contact period images. In addition, information concerning the stylistic chronology of the main shelter, available from superimposition, was noted.

The main shelter (WR011) comprises one large panel that measures 31 metres long by 4.8 metres deep and up to 6.9 metres high (height to main drip line). We recorded 232 paintings and eight stencils at this shelter and identified 17 layers of rock art. The remaining three rock shelters contain at least (1) 33 paintings, (2) 62 paintings and two beeswax figures, and (3) 33 paintings and six stencils (May et al. 2010:61-2). In total, 34 paintings that clearly depict introduced subject matter were recorded at Malarrak (Table 1).

Macassan/South-east Asian imagery

At Malarrak a single Macassan prau is depicted, initially in white but with further detail later added using yellow ochre. The painting (Figure 3) measures 102 centimetres in width (from stern to bow) and 99 centimetres in height (from the base of the hull to the top of the main mast). The vessel is depicted with its bow oriented to the right. It appears to show the sails furled and the vessel has no crew depicted. However, it does have an overall shape, mast and decking typical of South-east Asian praus, features also highlighted in depictions of praus elsewhere in northern Australia.

In the same rock shelter as the prau is a depiction of a knife in its sheath (Figure 4). It is 33 centimetres by 119 centimetres and is painted using the X-ray technique with a solid white background and purple-red outlines. A rock painting is classified as X-ray if any internal features of the subject are illustrated. This X-ray depiction enables the viewer to see through the sheath to the details of the knife located within, as only the handle is visible outside the sheath. This knife has been identified as of Macassan origin due to its design features, which are typical of an Indonesian small sword-like object known as badik (Chaloupka 1996:136). The badik is a particular form of kris (keris), a dagger with a hilt (handle) set at an angle in the plane of the blade (Gardner 1992:8, 41). Most notably, the badik has a 'small, straight, usually single-edged blade, with a straight or concave edge' (Gardner 1992:41). The knife painted at Malarrak exhibits all these features.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

The interpretation of the third painting that could relate to Macassan contact is debatable (Figure 5). Chaloupka (1996:136) argues that 'It represents two monkeys in a tree and is, in all probability, the work of one of the many men from this western region who travelled with the Makassans to Sulawesi'. Although we acknowledge that the figure on a lower branch does resemble a macaque in many regards, the figure on the higher branch (holding what appears to be a curved object that has both a boomerang-like and badik-like shape) is clearly human-like.

As for the depiction of the monkey, it could also be argued that this is a stylised depiction of another animal such as a Northern Spotted Quoll (Dasyurus hallucatus). When shown a photograph of this painting, Jimmy Galareya Namarnyilk (an Aboriginal Elder in the region) immediately identified it as djabbo (Northern Spotted Quoll) and suggested that it is being hunted by the person above. Of course, Jimmy was primarily thinking about Australian fauna from his region when looking at the photograph but the painted image does have a bushy tail more like a quoll than a monkey. In this case, there is not enough information to confirm or deny this is a Macassan-related artwork.

Another controversial painting (subject to varying interpretation) is the rare depiction of a building with a peaked roof (Figure 6). Internal elements suggest vertical supports and possibly decorative or design elements. These are at both ground level and upper stories.

[FIGURE 5 OMITTED]

This rock painting has been argued to represent a South-east Asian building or, more specifically, a Macassan smoke house (Chaloupka 1993, 1996). Chaloupka (1996:136), for instance, states that:

 Reports from European observers, and an
 outfitter's contract located by Macknight
 (1976:20) in South Sulawesi record that
 Makassans brought with them bamboo and
 prefabricated wall panels, in a form of kajang
 and ataps, mats of woven cane and palm
 leaf from which they constructed their living
 quarters and smokehouses for curing trepang.

This painting also resembles some Macassan buildings shown in an 1845 sketch from Victoria, Port Essington, by HS Melville (Macknight 1976:Fig. 11). Certainly, the building painted does not appear to be consistent with any British structures at Port Essington (described by Allen 2008). However, it could represent a structure at later settlements such as the Oenpelli or Goulburn Island Missions. Indeed, it is just as likely to represent a house or church made from sheets of bark at the Goulburn Island and Oenpelli Mission stations. This argument is strengthened by its proximity and similarity in colour and style to two ships--one immediately to the right (also seen in Figure 6) and one below the house painting. In this case, context is important for interpretation. As detailed rock art evidence for Macassan interactions in northern Australia is minimal in relation to that of European interactions, we need to guard against any desire to automatically see evidence for Macassan contact in ambiguous rock art. As Smith and van Schalkwyk (2002:236) warn:

 All of us who, following in the footsteps of
 the Abbe [Henri Breuill, seek to use rock
 art as a window into another culture face
 the same danger; that we will see in the art
 a mirrored reflection of our own prejudices
 and preconceptions. Worse still, we may then
 pass these on to others through our writings.

[FIGURE 6 OMITTED]

European imagery

Like the Macassan visits, European settlement in the north had far-reaching effects on local populations and this is reflected in the region's rock art. At Malarrak, many of the contact paintings depict objects of European origin--predominantly watercraft. We identified 17 European watercraft painted within the Malarrak complex (May et al. 2010:61). Figure 7 highlights one of these ship paintings, in this case a yellow painted sailing vessel. Distinguishable features identify each of these painted ships as of European origin and from different time periods. The vessels include luggers and steamers dating from the nineteenth and twentieth centuries.

The Wellington Range is becoming known for its diversity and density of watercraft paintings (May et al. 2010; Tagon et al. 2010; Tacon et al. 2012). A nearby rock shelter, Diulirri, is home to at least 20 depictions of European sailing vessels and Malarrak has a similar number at 17. This is, of course, only a minimum number as many others may be hidden beneath other paintings or they may have simply faded away over time.

[FIGURE 7 OMIITED]

As well as the European watercraft, there is one painting that could be argued to depict a non-Indigenous person, most likely of British origin. This person is depicted in the hands-on-hips posture that characterises Indigenous people's visual interpretation of Europeans--not just in Australia but also in many parts of the globe (e.g. Mokhtar and Tacon 2011). In South Africa, Ouzman (2003:14) argues:

 Human ethology--the study of universal
 human gestures--suggests that the hands-on-hips
 or 'teapot' posture is a 'possessive-aggressive'
 posture that is directed at
 showing exclusive ownership ... Alternatively,
 the human figures could have been painted
 to show their hands in their pockets--a
 posture that carries similarly sinister overtures
 of hiding one's actions/intentions.

There are also depictions of introduced animals at Malarrak. For example, Figure 8 reveals a person grasping in one hand the tail of, what is most likely, a goat and holding an axe in the other hand. Goats were an important source of milk and meat in most early settlements in western and northwestern Arnhem Land. The Malarrak figures have characteristic goat horns, tails and body proportions and differ from depictions found elsewhere in western Arnhem Land that have been identified or interpreted as Banteng cattle and buffalo. This, almost humorous, scene of an axe-wielding person hunting a disproportionately large goat is reminiscent of fishing scenes in older Arnhem Land rock art whereby the fish are many times larger than the boat.

[FIGURE 8 OMITTED]

Other introduced items painted at Malarrak include weapons and tobacco-smoking paraphernalia. A unique depiction is of a painting of a recently discharged pistol, with smoke rising from the barrel, but there are also two rifles. One rifle appears to be a Martini-Henry rifle popular from the 1870s and shows the bullet in the breech, again representative of the X-ray convention of rock art in this region. This weapon was popular with many European visitors to north-western Arnhem Land in the late 1800s. For instance, Customs collector Alfred Searcy, who travelled the Arnhem coast, boasted, 'I always used a Martini-Henry carbine when after big game' (Searcy 1907:63). Interestingly, no weapons are depicted being held by human figures. This is the same for two pipes and one tobacco pouch, which are standalone images.

Finally, one of the simplest and most evocative contact images at Malarrak is a simple drinking vessel or mug (Figure 9). The mug, measuring 16 centimetres by 14 centimetres, has been painted with a mixture of red and white ochres and is painted in profile so that the handle is clearly visible to the right. Clearly this image could relate to almost any early or later settlement but was valued highly enough to warrant a very prominent position in the main shelter at Malarrak.

Discussion

Malarrak provides a strange cacophony of contact period rock art. The images communicate interesting individual and collective stories of the contact encounters that were taking place. At Malarrak and other Wellington Range sites with contact subject matter, we are able to broadly distinguish between images related to encounters with Macassans/South-east Asians, the nineteenth-century British and others of the twentieth century.

Historical documents suggest that from at least the mid-seventeenth century Macassans made seasonal visits to the region to harvest trepang and trade with Aboriginal groups for goods such as turtle shell, iron wood, pearls and pearl shells, in return providing food, tobacco, alcohol, cloth, axes and knives (Clarke 2000; Macknight 1976). These seasonal visits, which did not officially cease until 1906, provided artists with new subjects to paint, and images of knives and praus have been identified in many places throughout Arnhem Land (Chaloupka 1993:191-2; Macknight 1976:84; Roberts 2004). During their extended period of contact, the Macassans developed close social, as well as economic, ties with the local Aboriginal groups. Yet, does the rock art reflect this extensive and ongoing relationship?

[FIGURE 9 OMITTED]

Importantly, the detail of the prau painting at Malarrak shows familiarity with Macassan fleets. It is depicted with recognisably distinctive features, such as a tripod mast and deck structures, and a flat bottom, a representation of the waterline. Based on the accuracy of the painting, it is suggested that the artist had an intimate knowledge of these watercraft and the way in which they sailed, indicative of direct experiences with the Macassan fleets along the coast, several kilometres north of the rock shelter. This intimate understanding of ships is not always evident in Aboriginal rock paintings or engravings of ships in Australia, and even within north-western Arnhem Land the standard and detail vary enormously depending on the artistic skills of the artist and their personal experiences of the ships (Tacon et al. 2012). Aboriginal people not only observed praus in their waters but also went on board and sometimes travelled on praus to Macassar and back (Earl 1846:239-40; Lamilami 1974:70; Macknight 1976:85), so it was possible for them to become very familiar with their features. It also is interesting to note that when other details were added in yellow to the white original, a European ship was painted next to it, perhaps reflecting changing times with the arrival of new (European) sailing vessels.

Contact rock art is known to have been produced both at and away from geographical centres of cultural contact. This follows from the understanding that the influences of cultural contact extend far beyond the isolated context of such encounters. For instance, Earl (1846:248) notes that Aboriginal people from south of Cobourg who visited Port Essington:

 spoke of a white people who dwelt in the
 country to the south, and who built houses of
 stone ... I have no doubt that they alluded to
 our colonists in South Australia, or in New
 South Wales. Scraps of news pass so rapidly
 from one tribe to another, that an event of
 any importance is known over a large extent
 of country in the course of a very few months.

Thus it cannot be assumed that contact art occurs only within the location where the cross-cultural encounters took place (Frederick 1999:140): information and objects were incorporated into Aboriginal social networks as, similarly, goods collected from Indigenous lands and information about these events were removed to Macassar and other Asian ports, and from there into South-east Asian networks. Therefore, the images in Arnhem Land should be considered as historical accounts of activities with local, regional and international dimensions, especially as South-east Asian trade articulated with global networks of trade and communication.

The depiction of a Macassan prau, in a rock shelter directly inland from coastal sites such as Anuru Bay (Macknight 1976; Theden-Ringl et al. 2011), indicates that influence on Aboriginal communities extended away from the immediate contact zone. In Australia and elsewhere, Aboriginal depictions of watercraft in rock art and other media indicate subjects of significance to artists and communities (e.g. Clarke and Frederick 2006; Roberts 2004). The image of a prau is the primary depiction in known rock art in which Aboriginal artists have represented Macassan/Aboriginal cultural interaction (Tacon et al. 2010). Such paintings support oral histories of the movement of people and close connections between Aboriginal and Macassan groups, something that traditional archaeology is struggling to achieve.

The painting of the knife in its sheath is likely evidence of its importance in trading relationships between the two cultural groups. Knives were a valued trade item during cross-cultural encounters across Australia (e.g. Layton 1992). As Mitchell (2000:182) notes, 'one of the most visible consequences of culture contact with outsiders ... was the adoption of foreign material culture as trade goods within indigenous societies'. Importantly, and as mentioned briefly earlier, the knife is illustrated using the traditional X-ray manner of depiction, with the blade shown within its sheath. The use of this traditional technique indicates the continuation of artistic conventions that may demonstrate something of what Frederick (1999:134) argues as 'the measures Indigenous Australians took towards securing their own cultural survival in a transforming world'.

The presence of both the prau and the knife paintings within the rock shelter, along with other depictions of praus at nearby sites, indicates that the Macassans had an influence not only on the art but also on the material culture of the Arnhem Land region. Such material influences do not always survive in the archaeological record. Yet it should also be noted that the majority of introduced subject matter at Malarrak relates to contact with Europeans during the nineteenth and twentieth centuries.

Contact with Europeans in northern Australia is argued to date from the arrival of the Dutch, who made landfalls at Cape York in 1606 (Veth et al. 2008). From the twentieth century onwards contact with Europeans in north-western Arnhem Land was regular, as Aboriginal groups became associated with the activities of hunters, missionaries, traders, explorers, scientists, settlers and military personnel. Historical accounts of these settlements and activities exist. However, these rarely indicate the views of the Aboriginal communities and their perspectives on contact. Rock art of this period is a direct source that illustrates Aboriginal responses to changes during this period.

Painted within Malarrak are at least 17 depictions of European watercraft. These depictions are more than 'pier head' views of the sea by Aboriginal community members. Rather, they 'represent skilled mariners recording aspects of foreign maritime traditions' (Burningham 1994:145). Boats were central to cross-cultural encounters (e.g. Acheson and Delgado 2004) and are a large component of contact rock art traditions in Australia (Bigourdan and McCarthy 2007). The prevalence of watercraft in the rock art record at Malarrak is a clear indicator of the interaction between local Aboriginal groups and European intruders.

The types of vessels most commonly represented are, like elsewhere in the region, 'twin-masted sailing ships, being either ketches or schooners, typical of the European-style trading or fishing vessels that operated along the northern coastlines and rivers from around the 1870s to 1930s' (Roberts 2004:26). Introduced European vessels at Malarrak, including single-masted (n = 8) and twin-masted (n = 4) sailing vessels typical of coastal and riverine transport, are the most common depictions, while three-masted larger sailing ships (n = 2) are rarer, as are steam ships (n = 3). Most vessels show rigging and tend to have their sails furled; those with sails set are single- and twin-masted vessels. The internal aspects of the vessels are regularly shown, as are hull details beneath the waterline. Most do not have any crew or cargo depicted, in contrast to many of the watercraft painted at nearby Djulirri (Tacon et al. 2010).

In at least two instances watercraft depictions were augmented over time. In the main shelter, the Macassan prau, originally painted in white, has yellow pigment added. At a different Malarrak shelter, a European sailing/steam vessel appears to have been through several reinterpretations with the addition and incorporation of elements such as a funnel and two crew members (Figure 10).

[FIGURE 10 OMITTED]

Although contact with the Macassans occurred over a longer period of time, European watercraft are depicted in much higher numbers (n = 17 versus n = 1). Of course, this may in part reflect the greater visibility and survival of more recent painted images. European contact saw irreversible changes to Aboriginal societies, introducing new technologies and material culture, new foods and ways of living. Aboriginal representations of introduced watercraft represent a clear preoccupation with sailing vessels and point to the significant role such vessels had in redefining Aboriginal life at the time.

The depiction of other European material culture is also significant. The second-most common introduced subject matter at Malarrak is horned animals. Other items of note are firearms and smoking-related items. All of these animals/objects were more than just observed--they were objects that were used by Aboriginal people and valued. The painting of a drinking mug at Malarrak is unusual and rare. To find such a simple item painted in a rock shelter means it had importance to the artist but its exact story has been lost. It is possible that the mug is of missionary origin, relating to the period in the early to mid-1900s when missionary societies (such as the Church Missionary Society in Oenpelli/Gunbalanya) began establishing themselves in the area. The significance of such an item could not be inferred from historical sources. Only its depiction within this rock shelter indicates any attachment to the object that is out of the ordinary.

Despite the presence of introduced material culture, artists of the time continued to depict more traditional subject matter in conventional ways during the contact period. In some instances, traditional themes are found superimposed over paintings of introduced subject matter such as sailing vessels. In fact, besides the mug, the most recent paintings at all of the sites recorded depict classic Arnhem Land subjects such as fish and macropods (and, at Djulirri, an emu; see May et al. 2010:61). According to May et al. (2010:35), 'it is as though the local artists were noting and commenting upon the introduced aspects of the visiting cultures and then simply returning to their more usual artistic activities'. Also significant is that paintings of introduced material culture are painted using traditional artistic protocols including, but not restricted to, the X-ray convention, choosing to highlight distinguishing features of the object at the loss of others.

Conclusion

Our aim in this paper has been to explore recent Australian Aboriginal rock paintings of introduced subject matter and to question the historical significance of these paintings, as well as their role in interpreting contact histories. The presence of paintings depicting introduced material culture alongside traditional subject matter indicates that the rock art of the contact period is a record of both change and continuity. The art not only reflects the material changes brought by outsiders but also highlights the active role Aboriginal people took in responding to these circumstances. This record illustrates some of the cultural significance of introduced material culture that cannot otherwise be found in historical literature or other forms of archaeological research. The study of introduced subject matter in rock art offers an opportunity to study the influences of cross-cultural contact not just as historical events of the past but as ongoing relationships and social circumstances generated from encounters, providing essential and fascinating links between prehistory, the recent past and contemporary times.

Acknowledgments

We thank Ronald Lamilami and his family h)r supporting this research and for their generosity and hospitality in the field. Picturing Change was funded by Australian Research Council Discovery Grant DP0877463 and we would like to acknowledge Dr June Ross as a fellow Chief Investigator on this project. Thanks also to the following people for their assistance in the field: Janet and Phil Davill, Wayne Brennan, Ines Domingo Sanz, Melissa Marshall, Kirsten Brett, Michelle Langley, Megan Berry and Daryl Wesley. Susie Davies is thanked for help with archival research. Thanks to Injalak Arts and Crafts and the Northern Land Council, as well as the Australian National University, Griffith University and The University of Western Australia.

REFERENCES

Acheson, Steven and James P Delgado 2004 'Ships for the taking: Culture contact and the maritime fur trade on the northwest coast of North America' in Tim Murray (ed.), The Archaeology of Contact in Settler Societies, Cambridge University Press, Cambridge, pp. 48-77.

Allen, Jim 2008 Port Essington: The historical archaeology of a north Australian nineteenth century military outpost, Sydney University Press, Sydney.

Berndt, Ronald M and Catherine H Bern& 1954 Arnhem Land: Its history and its people, FW Cheshire, Melbourne.

Bigourdan, Nicolas and Michael McCarthy 2007 'Aboriginal watercraft depictions in Western Australia: On land, and underwater?', Bulletin of the Australasian Institute for Maritime Archaeology 31:1-10.

Burningham, Nick 1994 'Aboriginal nautical art: A record of the Macassans and the pearling industry in northern Australia', The Great Circle 16(2):139-51.

Chaloupka, George 1993 Journey in time: The 50,000 year story of the Australian Aboriginal rock art of Arnhem Land, Reed Books, Chatswood, NSW.

--1996 'Praus in Marege: Makassan subjects in Aboriginal rock art in East Arnhem Land, Northern Territory, Australia', Anthropologie 34:131-42.

Clarke, Anne 1994 Winds of Change: An archaeology of contact in the Groote Eylandt archipelago, Northern Australia, unpublished doctoral thesis, Australian National University, Canberra.

--2000 'The "Moorman's trowsers": Macassan and Aboriginal interactions and the changing fabric of indigenous social life' in Sue O'Connor and Peter Veth (eds), East of Wallace's Line: Modern Quaternary Research in Southeast Asia 16, AA Balkema, Rotterdam, pp.315-35.

--and Ursula Frederick 2006 'Closing the distance: Interpreting cross-cultural engagements through indigenous rock art' in Ian Lilley (ed.), Archaeology of Oceania: Australia and the Pacific islands, Blackwell Publishing, Oxford, pp.116-33.

Crawford, Ian 1969 Late prehistoric changes in Aboriginal culture in Kimberley, Western Australia, unpublished doctoral thesis, University of London, London.

Earl, George Windsor 1846 'On the Aboriginal tribes of the northern coast of Australia', Journal of the Royal Geographical Society of London 16:239-51.

Evans, Nicholas 1992 'Macassan loanwords in Top End languages', Australian Journal of Linguistics 12(1):45-91.

Flinders, Matthew 1814 A Voyage to Terra Australis ... in the Years 1801, 1802, and 1803 ... (2 volumes and atlas), GW Nicol, London.

Frederick, Ursula 1997 Drawing in differences: Changing social contexts of rock art production in Watarrka (Kings Canyon) National Park, Central Australia, unpublished Master's thesis, Australian National University, Canberra.

--1999 'At the centre of it all: Constructing contact through the rock art of Watarrka National Park, Central Australia', Archaeology in Oceania 34(3): 132-43.

Gardner, Gerald B 1992 Keris and Other Malay Weapons, Progressive Publishing Company, Singapore.

Keyser, James D and Michael Klassen 2003 'Every detail counts: More additions to the Plains biographic rock art lexicon', Plains Anthropologist 48 (184):7-20.

Klassen, Michael 1998 'Icon and narrative in transition: Contact-period rock-art at Writing-On-Stone, Southern Alberta, Canada' in Christopher Chippindale and Paul SC Tacon (eds), The Archaeology of Rock-art, Cambridge University Press, Cambridge, pp.42-72.

--, James D Keyser and Lawrence Loendorf 2000 'Bird Rattle's petroglyphs at Writing-On-Stone: Continuity in the biographic rock art tradition', Plains Anthropologist 45(172): 189-201.

Lamilami, Lazarus 1974 Lamilami Speaks, the Cry Went Up: A story of people of Goulburn Islands, North Australia, Ure Smith, Sydney.

Layton, Robert 1992 Australian Rock Art: A new synthesis, Cambridge University Press, Cambridge.

McIntosh, Ian 2004 'Personal names and the negotiation of change: Reconsidering Arnhem Land's adjustment movement', Anthropological Forum 14(2):141-62.

Macknight, Campbell 1976 The Voyage to Marege, Melbourne University Press, Melbourne.

--2011 'The view from Marege: Australian knowledge of Makassar and the impact of the trepan industry across two centuries', Aboriginal History 35:121-43.

May, Sally K, Paul SC Tacon, Meg Travers and Daryl Guse 2010 'Painting history: Indigenous observations and depictions of the "other" in far north Australia', Australian Archaeology 71:57-65.

Mitchell, Scott 2000 'Guns or barter? Indigenous exchange networks and the mediation of conflict in post-contact western Arnhem Land' in Robin Torrence and Annie Clarke (eds), The Archaeology of Difference: Negotiating cross-cultural engagement in Oceania, Routledge, London, pp.182-214.

Mokhtar, Saidin and Paul SC Tacon 2011 'The recent rock drawings of the Lenggong Valley, Perak, Malaysia', Antiquity 85 (328):459-75.

Molyneaux, Brian 1989 'Post-contact Micmac rock art' in Howard Morphy (ed.), Animals into Art, Unwin Hyman, London, pp.193-214.

Ouzman, Sven 2003 'Indigenous images of a colonial exotic: lmaginings from Bushmen southern Africa', Before Farming 1(6):239-56.

--and Ben Smith 2004 'Taking stock: Identifying Khoekhoen herder rock art in southern Africa', Current Anthropology 45(4):499-526.

--and Johannes Loubser 2000 'Art of the apocalypse: Southern Africa's Bushmen left the agony of their end time on rock walls', Discovering Archaeology 2(5):38-44.

Paterson, Alistair 2012 'Rock art as historical sources in colonial contexts' in Maxine Oland, Siobhan M Hart and Liam Frink (eds), Decolonising Indigenous Histories: Exploring prehistoric/colonial transitions in archaeology, University of Arizona Press, Tucson, pp.66-85.

Roberts, David Andrew 2004 'Nautical themes in the Aboriginal rock paintings of Mount Borradaile, western Arnhem Land', The G teat Circle 26(1):19-5.

Searcy, Alfred 1907 In Australian Tropics, Kegan Paul, Trench, Trubner, London.

Smith, Ben W and Johnny A van Schalkwyk 2002 'The white camel of the Makgabeng', Journal of African History 43:235-54.

Tacon, Paul SC, Sally K May, Stewart Fallon, Meg Travers, Daryl Guse and Ronald Lamilami 2010 'A minimum age for early depictions of Macassan praus in the rock art of Arnhem Land, Northern Territory', Australian Archaeology 71:1-10.

--, June Ross, Alistair Paterson and SalIy K May 2012 'Picturing change and changing pictures: Contact period rock art of Australia' in Josephine McDonald and Peter Veth (eds), A Companion to Rock Art, Blackwell Publishing, Oxford, pp.420-36.

Theden-Ringl, Fenja, Jack Fenner, Daryl Wesley and Ronald Lamilami 2011 'Buried on foreign shores: Isotope analysis of the origin of human remains recovered from a Macassan site in Arnhem Land', Australian Archaeology 73:41-8.

Veth, Peter, Peter Sutton and Margo Neale (eds) 2008 Strangers on the Shore: Early coastal contacts in Australia, National Museum of Australia Press, Canberra.

NOTE

(1.) Following convention, we use the term 'Macassan' to refer to those people from South-east Asia visiting northern Australia as part of the trepang industry. As Macknight (2011:128-9) notes, it appears anthropologists Ronald and Catherine Berndt replaced 'Malay', prevalent in historic documents, with 'Macassan' to describe South-east Asian visitors. An alternative spelling, 'Makassan', is also widely used; however, for this paper we have chosen to simply use the original 'Macassan' spelling.

Sally K May

The Australian National University

Paul SC Tacon

Griffith University

Alistair Paterson

The University of Western Australia

Meg Travers

University of New England

Sally K May is a Senior Lecturer in the School of Archaeology and Anthropology and convenor of the Graduate Program in Liberal Arts (Cultural and Environmental Heritage) and the ANU Rock Art Research Centre. Sally's PhD was in interdisciplinary cross-cultural research and focused on the role of art and community art centres in remote Indigenous Australian communities. Her research interests include Indigenous archaeology, rock art studies, the archaeology of art, contact archaeology and the anthropology of collecting. Sally has been working in western Arnhem Land since 2000. She is the author of Collecting Cultures: Myth, politics and collaboration in the 1948 Arnhem Land Expedition (2009) and has edited two books, Archaeologies of Art: Time, place and identity (2008, with I Domingo Sanz and D Fiore) and Macassan History and Heritage (forthcoming, with M Clark). Sally has also produced numerous journal articles and book chapters relating to her research areas.

<sally.may@anu.edu.au>

Paul SC Tacon FAHA FSA is Chair in Rock Art Research and Professor of Anthropology and Archaeology in the School of Humanities, Griffith University, Queensland. He also directs Griffith University's Place, Evolution and Rock Art Heritage Unit (PERAHU). Professor Tacon has conducted archaeological and ethnographic fieldwork since 1980 and has more than 80 months of field experience in remote parts of Australia, Cambodia, Canada, China, India, Malaysia, Myanmar, southern Africa, Thailand and the United States. He co-edited The Archaeology of Rock-art with Dr Christopher Chippindale (1998 and republished four times) and has published more than 190 academic and popular papers on prehistoric art, body art, material culture, colour, cultural evolution, identity and contemporary Indigenous issues. Much of his current research is related to better situating Australian archaeology and contact history in a South-east Asian regional context and to more fully involving Indigenous peoples in archaeological research. Professor Tacon leads the Protect Australia's Spirit campaign devoted to raising awareness about and threats to Australia's unique rock art heritage and the establishment of an Australian national rock art heritage strategy.

<p.tacon@griffith.edu.au>

Alistair Paterson, Professor and Discipline Chair of Archaeology, The University of Western Australia.

Professor Alistair Paterson is an archaeologist in the School of Social Sciences at the University of Western Australia. His research and teaching covers culture contact, historical archaeology, colonisation and methods. Much of this is located in Western Australia, including regional studies of Australia's north-west, islands and coasts in colonial settings, and early colonial race relations. He is the author of A Millennium of Cultural Contact (Left Coast Press, 2011) and The Lost Legions: Culture Contact in Central Australia (Alta Mira, 2008) and co-editor of Archaeology in Practice: A student guide to archaeological analyses (Wiley Blackwell, 2013).

<alistair.paterson@uwa.edu.au>

Meg Travers is a PhD candidate from the University of New England. She completed her undergraduate studies and a Master of Liberal Arts (Cultural and Environmental Heritage) at the Australian National University. She has been involved in rock art research since 2008, and has worked in both western Arnhem Land and the northwest Kimberley. Her current research looks at the form and pace of stylistic change in the rock art assemblage of the north-west Kimberley, specifically the Wararrajai Gwion (formally known as the Clothes Peg Figure) Period, in order to examine past shifts in cultural, technological, social and economic activities in the region.

<mtravers@myune.edu.au>

Table 1: General interpretation of the paintings of
introduced subject matter at Malarrak

Introduced subject matter Minimum number

Sailing vessel--European 17
Horned animal 4
Firearm 3
Smoking pipe 2
Unidentified 2
Building 1
Coffee Mug 1
Knife (badik) 1
Human (with hands on hips) 1
Sailing vessel--prau 1
Tobacco pouch 1