Symbolic language in Torres Strait, NE Australia: images from rock art, portable objects and human scars.

Brady, Liam M.

The Torres Strait is often in the research literature--unsurprisingly since it is not only a key area for early settlement but one where ancient and modern practice resonate. Rock art is a prominent archaeological source for the region--but not the only one. In this study the author shows how rock art interconnects with imagery on portable artefacts and scarification--scarring patterns on skin--to define cultural zones of the last few centuries in territories occupied by both horticulturalists and hunter-gatherers.

Keywords: Torres Strait, rock art, hunter-gatherers, horticulturalists, portable objects, scarification, cultural zones

Introduction

Studies that focus on images depicted on rock walls can yield noteworthy insights into social processes such as interregional interaction and the construction of boundaries (e.g. Bradley 1997; David & Chant 1995; Lee & Hyder 1991; Wilson 2002). However, studies that focus only on images produced on rock overlook the valuable contribution comparative designs recorded on portable art, and other forms of decorative media, can provide to the research process. In Western Europe, similarities in style recorded from parietal and portable art have been used to establish relative chronologies (e.g. Barandiaran 1973; Begouen & Clottes 1991; Leroi-Gourhan 1965; Omnes 1982). Elsewhere, studies from North America, Asia and Oceania have used this comparative method to discuss a range of topics including shamanic symbolism, gender and motif transference, identification of historical events, and differing contexts of production (e.g. Devlet 2001; Parsons 1987; Tacon 1993; Wilson 1999).

In ah Australian context there have been some attempts to explore the connections between similar images documented on different media. For example, Harte (1992) combined rock art and material culture to examine links between rock paintings and rainforest shields in north Queensland, while Tacon et al. (2003) incorporated designs recorded in rock art into a wider study looking at similar designs found on wood, stone and earth to document cross-cultural interaction between Europeans and Aboriginal Australians in south-east Australia. In 2000, research conducted by Best (2003) sought to identify spatial patterning of style conventions recorded on Aboriginal objects from Queensland. Although her study utilised rock art motifs as a secondary source, she provided a glimpse of the types of regional linkages that can be established between rock art and material culture in terms of spatial patterning of design conventions. While rock art research in Australia has attracted much attention over the years, studies that utilise multiple media to examine broad-scale interregional interaction are relatively few.

This paper reports results from a unique rock art research project carried out in the Torres Strait islands in tropical far north-eastern Queensland. The aim of the project was to identify interaction between islands and between regions, using the spatial distribution of images recorded on rock walls, portable objects, and in scarification designs documented on the Indigenous inhabitants of the islands (Torres Strait Islanders).

The islands are squeezed into a 150km expanse of water between two cultural realms: hunter-gatherers from Australia and horticulturalists from Melanesia. Torres Strait Islanders have long been scrutinised by anthropologists because of their intermediary position between two distinctly different cultural groups. Of key concern has been whether the islands represented a bridge or barrier to cultural, linguistic and genetic diffusion (e.g. Haddon 1935; Walker 1972). Although recent archaeological and linguistic research is beginning to expand our understanding of the nature and antiquity of occupation in the islands as well as shed light on the extent of possible Austronesian influences in the Torres Strait islands (e.g. Carter 2004; David et al. 2004b; McNiven et al. 2006) little research has been undertaken to investigate how Islander visual heritage can actas an indicator of interaction across this cultural divide.

Problems with Torres Strait ethnography and archaeology

The ethnography of the Torres Strait islands and the Australian and Papua New Guinea mainlands has been documented by various researchers since the late 1800s (e.g. Haddon 1904; Lawrence 1994; McNiven 1998). Yet recent work has turned up a number of contradictions between the ethnographic and archaeological evidence. For example, the source of stone axes (Kiwai 'type' axes) and stone-headed clubs (gabagaba) collected from various locations across Torres Strait and the southern Papuan Lowlands has been generally assigned to the Papuan mainland. However, recent geological sourcing of the stone used in the manufacture of these objects has revealed a likely stone source in the Western Torres Strait islands, indicating the presence of extensive exchange networks operating between Torres Strait and the Papuan mainland (McNiven 1998; McNiven et al. 2004a). Additionally, Alfred Haddon's (1890: 303) earlier assertion that 'there never has been anypottery' in Torres Strait was proved incorrect with the recent discovery of pottery sherds excavated from both the Eastern and Western islands (Carter 2004; McNiven et al. 2006). Research into the pottery sherds from the Eastern islands revealed a New Guinea source of the pottery tempers and 'the most likely place of manufacture' (Carter 2004: 307), thus providing evidence for another exchange item with considerable antiquity (2000 cal BE Carter 2004: 307). The artistic evidence, discussed below, provides further evidence for interactive relationships between people living in Torres Strait and neighbouring regions.

Artistic practice and interregional interaction

Islander artistic practices were documented during Haddon's extensive research in the late 1800s. He collected a large sample of material culture objects, drawings by Islanders on paper, and photographs of various aspects of Islander life--all of which constituted the bulk of his data on Torres Strait arts and crafts. As Haddon (1912: 342) noted in his introduction to Islander decorative art: 'the decorative art of the Islanders is so closely connected with that of the natives of the adjacent mainland of New Guinea (Daudai) and of the islands at the mouth of the Fly River, especially Kiwai, that it is impossible to study them apart ... it must be remembered that many decorated objects have been imported from New Guinea to Torres Straits and possibly a few have passed in the reverse direction.'

Interregional interaction in the Torres Strait region was (and still is) reflected in social alliances between island groups, oral traditions, totemic systems, marriage and a complex exchange system. However, there were some notable impediments to social interaction between some groups:

Linguistics: the islands were separated into two broad socio-linguistic blocs; the Eastern Islanders of Erub, Ugar, Mer, Dauar and Waier spoke a Papuan language (Meriam Mir) while the Central and Western Islanders spoke an Australian Aboriginal language (Kala Lagaw Ya) (Shnukal 1998);

Marriages: marriage data collected in 1898 reveals marriages frequently occurred between Western Islanders, and Western and Central Islanders, however; marriages among Eastern Islanders were mainly confined to islands within their own group (Rivers 1904);

Totemic dan systems: a totemic clan system separated Western and Central Islanders from Eastern Islanders. In the Western and Central islands, each clan was identified by a major totem, and each clan had its own recognised territory on the island (Haddon & Rivers 1904). Additionally, Haddon and Rivers (1904: 161) noted that Islanders visiting another island would be looked after by individuals of the same torero and in cases where there were no totemic relatives, they would stay with individuals whose totems were associated with theirs. Seventeen of the 36 totems documented by Haddon and Rivers in 1898 can be identified on more than one island, revealing a widespread distribution of markers associated with social affiliation. Clearly toreros were significant elements in linking people across the Western and Central islands. Conversely, in the Eastern islands, people were not totemically linked to the Western and Central Islanders due to the lapsing of the totemic system in that particular region (Haddon 1908: 255). Consequently, a mechanism to enable social interaction between the two major island groups was lacking. Significantly, no totemic linkages were established between Islanders, and mainland New Guineans of Australian Aboriginal groups.

Exchange practices: exchange practices involved the movement of specific items between islands and mainlands. Haddon's (1904: 293-5) tripartite division of primary exchange relationships in the Western islands consisted of:

* intra-insular exchange of food and material culture items between islands;

* exchange with New Guinea largely conducted by 'middlemen' at Saibai (north-Western islands) who procured items for and from other Western Islanders;

* exchange with Cape York which was carried out mainly via the Kaurareg of the south-Western islands.

In the Eastern islands exchange was carried out mainly with south-western Papua New Guinea, and minimal exchange with the Western islands was conducted through Central Islander intermediaries. A key institution that linked Fly river (south-western Papua New Guinea) and Torres Strait Islander communities was the extensive traffic in canoes that operated through a complex chain of Islander and Papuan intermediaries (Haddon 1908: 188).

Taking account of the boundaries or divisions recorded between Eastern, Western and Central Islanders at the time of European contact, we can begin to assess their role in the exchange of design conventions and design features within the Torres Strait region artistic system.

Torres Strait art sources

Rock art sites were systematically recorded from eight Western Islands, one Central Island, and one at the tip of Cape York between 2000 and 2004 (Figure 1). A total of 1281 paintings were documented from 56 sites, consisting of 983 'determinate' images, (those that could be formally identified) and 298 indeterminate images (unidentifiable because of heavy damage of deterioration). Images were largely executed in monochrome red (97.4 per cent) although yellow, white, blue, black, mulberry and pink paintings also occurred in monochrome and bichrome. Documented motifs include anthropomorphs, zoomorphs, material culture objects, faces/masks, anda large quantity of geometric and abstract images. The harsh coastal regime in Torres Strait has caused many images to suffer damage from water and salt. As a result, each image was digitally recorded and subjected to systematic computer enhancement (see Brady 2005 for details) along with all the rock surfaces containing any trace of pigment. The dating of Torres Strait rock paintings has only been established through relative dating techniques thus far. Using oral traditions (e.g. McNiven et al. 2004b; Sharp 1992), comparisons with earlier observations of images, and excavation data (in situ powder recorded from a dig below a panel of rock art, see David et al. 2004a for details) a relative age of a few hundred years has been accorded to most Torres Strait rock paintings.

Collections of Torres Strait material culture are found in museums around the world, although the largest collections were assembled by Haddon in 1888 and 1898 and donated to Cambridge University (e.g. Moore 1984). South-western Papua New Guinea material culture was largely collected by Gunnar Landtman (e.g. Landtman 1933) in the early 1900s, while very few objects have been collected from Cape York. Decorated images occurring on material culture are generally found on wooden objects (e.g. bamboo tobacco pipes, drums), although other materials including turtle-shell and stone also feature a range of motifs. Given the somewhat short lifespan of wooden material culture objects in a tropical climate it is likely that images on wood and images on rock walls are of similar age.

Pencil drawings produced on paper by Islanders, and images scarred on the human body (rendered on the skin through scarring technique), were also recorded by Haddon and other observers in the late 1800s. Although scarification images relate to practices extant a century or more ago, both forms of art provide a date for when individual design conventions were still being produced.

[FIGURE 1 OMITTED]

Distinctive motif forms

A large quantity of distinctive motif forms has been identified in the Western Torres Strait rock painting assemblage that reveal correlations with scarification designs and designs recorded on decorated objects (detailed in Brady 2005).

Fish headdress

Fish headdresses are one of the most distinctive and elaborate elements of Torres Strait ceremonial paraphernalia (examples are given in Figure 2). The body of the headdress was usually constructed of turtle-shell (in some cases with intricate carvings) and decorated with cassowary feathers and rattles. Two rock paintings depicting fish headdresses have been documented from Dauan and Mua in the top Western, and mid-Western islands, respectively. The former image, originally documented by McNiven et al. (2000), features the headdress surmounted on an anthropomorph, while the latter is displayed independently. Each headdress features a left-facing open mouth, nose mount, and a triangular design extending upwards from the middle of the body.

McNiven et al. (2004b) drew attention to similarities between the fish headdress rock painting from Dauan with a drawing by a Mabuyag Islander (Joani) and examples of turtleshell headdresses collected from Torres Strait and deposited in museums. Several other objects from across the Torres Strait region can also be included for comparative purposes. In the Eastern Islands, Haddon collected a spinning top from Mer with a drawing of a fish headdress and several tobacco pipes from Western and Central Torres Strait islands which feature incised fish headdresses. Haddon also collected numerous drawings on paper produced by Western and Eastern Islanders which depicted fish headdresses. In addition, Haddon produced colour sketches of a fish headdress used in the Sawfish Dance he witnessed on Waibene in 1888. Landtman (1927: 339) also noted that in the horiomu ceremony (connected with the cult of the dead) enacted by coastal Kiwai-speaking people and some top Western Islander groups, a similar headdress (karara) was employed. Although the basic structure of the design form (e.g. fish body with open mouth, nose mount and surmounted central design) remains consistent between rock paintings and decorated material culture, considerable design embellishments are visible on the material culture correlates (e.g. a line extending from the tip to tail of the fish; tassles; fins; and other linear designs extending from the body). This suggests that the depiction of the fish headdress asa rock painting occurs in a simplified form, although the overall design is indicative of an artistic expression that occurs across the Torres Strait islands and into coastal south-western Papua New Guinea.

Snake

Only one painting of a coiled snake (a common totemic animal species from the Western and Central islands, see Haddon & Rivers 1904) has been documented from Torres Strait rock art (Mua in the mid-Western islands) (Figure 3). However, the infilled coiled snake design convention can be traced across the Torres Strait region on two additional forms of media: decorated material culture objects andas a scarification design on Islanders. Two coiled snakes are depicted on a bamboo tobacco pipe collected in 1870 from Cape York (one of the rare instances where material culture correlates have been recorded from mainland Australia); Haddon (1946: 68) had previously suggested that the 'owner belonged to tabu, snake, totem clan. 'A second tobacco pipe collected from an unknown Torres Strait island during the HMS Challenger expedition from the mid-1870s also features two incised coiled snakes facing away from each other. Furthermore, during Haddon's research at Mabuyag he documented the coiled snake design form scarred on the lower back of Patagam (a Mabuyag Islander) as a totemic marker, while Gizu of Mabuyag also drew the snake totem for Haddon using the same design convention. Consequently, the coiled snake design convention is located on multiple media, and spatially restricted to the Torres Strait islands and Cape York where it is typically associated with the tabu (snake) totem.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

4-pointed star

A unique rock painting of an anthropomorph from Dauan features a distinctive design--an outlined 4-pointed star with a circle in the middle. It is drawn on the belly of the figure identified by the Dauanalgal asa dogai--an evil and mischievous female spiritual being that is greatly feared by Islanders (see e.g. Haddon 1904; Lawrie 1970) (Figure 4). Correlates with this design form were first identified by McNiven et al. (2004b: 232-6) on decorated material culture objects from Torres Strait, the Papuan Gulf and south-eastern West Papua. In many instances, stylistic variations of the 4-pointed star with a circle in the middle have been recorded as a design drawn on the belly of sculptured anthropomorphs, and anthropomorphs carved on wooden boards, bark belts and bowls from areas of the Sepik river region, the northern coastal region of Papua New Guinea and across the Papuan Gulf (details in Brady 2005). Yet the 4-pointed star design is not restricted to the belly of anthropomorphs. It has been recorded with stylistic embellishments on many wooden objects collected from Papua New Guinea and the Torres Strait region. Objects include: a headdress collected by Haddon from Mer in the Eastern islands; a bamboo tobacco pipe collected in 1870 at Cape York (see Figure 3); a wooden comb made by Waria of Mabuyag and collected by Haddon in the late 1800s (see Haddon 1912: 362); and carved wooden boards and bark belts collected from across the Papuan Gulf region and Kiwai island at the mouth of the Fly river in south-western Papua New Guinea (see e.g. Lewis 1931; see also Brady 2005 for further examples).

McNiven et al. (2004b: 234) also drew a link between this design form and gari ritual paraphernalia of the Marind-Anim in south-east West Papua. They suggested that the 4-pointed star design may have an astrological association given that the star on the gari represents the morning star, while some stars and constellations in Torres Strait are identified as dogai variants.

Thus, while this distinctive design form is restricted to only one occurrence in Torres Strait rock art, it otherwise has a widespread geographic distribution, extending from the north coast and Sepik river regions of Papua New Guinea southwards to Torres Strait (as lar south as Cape York, and eastwards to Mer), eastwards through the Fly river estuary (Kiwai island) and into the Papuan Gulf region, and westward into West Papua (Figures 1 and 5). Although the design form is only recorded in a single rock painting from Dauan, an extensive geographical relationship spreading into Papua New Guinea does appear to exist between design form and placement (on the belly of an anthropomorph) found on both fixed and portable media. Several other distinctive motifs (e.g. masks, see McNiven et al. 2006) and design conventions (e.g. concentric circle eyes, and interlocked teeth, see Brady 2005) with limited occurrences in Torres Strait rock art also display extensive material culture links with coastal southern Papua New Guinea.

Triangle variants

A distinctive form of triangle variant--here identified asa 'Hooked Triangle'--has been recorded in the rock art from three mid-Western islands (Zurath, Badu and Pulu). The design form consists of an infilled triangle with two small curved lines emanating from either side of the tip of the triangle, and has been recorded individually or laterally linked (Figure 6). Variations of the Hooked Triangle are common on decorated objects. The design form is recorded as an engraving on the upper jaw of warup drums--a popular collectors' item--obtained from several Torres Strait islands including Saibai, Tudu, Nagi, Muralag, and on gama drums collected from Mer, and from Kiwai island (e.g. Haddon 1912). The Hooked Triangle has also been recorded on bamboo tobacco pipes collected from Mer and Kiwai island, and wooden combs collected from Kiwai island. The distinctive Hooked Triangle design is also featured on a pottery sherd collected from Kalumpang in Sulawesi (Indonesia) in 1951 (Glover 1986: 74). This similarity represents one of the first director indirect artistic links beyond the Torres Strait region and into Southeast Asia. In addition to portable material culture objects, a variation of the Hooked Triangle was recorded by Haddon as a scarification design on a girl from Mer. This design expression--identified as a kip sor koima, back koima, or kip user is located on a girl's back and is considered a sign of mourning (Haddon 1912: 22).

Consequently, there exists a broad-scale distribution of the Hooked Triangle design form across the Torres Strait region and beyond, and across a range of media. Although the form of the Hooked Triangle in Torres Strait rock art is relatively consistent (red infilled triangle with two curved lines), its appearance on both portable material culture objects and people reveals a much more varied and elaborate appearance.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Discussion

The majority of images examined occur in various forms and are recorded on various media across the horticultural regions of Torres Strait and Papua New Guinea. Only the snake and 4-pointed star motifs are recorded at Cape York--on the same portable object (bamboo tobacco pipe) which Haddon (1946: 68) has already indicated most likely belonged to a tabu or snake totem clan member (most probably from Torres Strait). The Hooked Triangle also occurs to the west into Indonesia (where there are few links with Torres Strait) on a single pottery sherd. The distribution of shared imagery largely correlates with the north-south gradient in the occurrence of horticulture in the region. Instances of shared imagery are most common in the northern Torres Strait-Papua New Guinea region where horticulture is well-established, but decline sharply south of Muralag in the south-Western islands. This decline mirrors the decrease on the emphasis on horticultural practice--Islanders in the south-Western group engaged in only intermittent or sporadic horticulture while Aboriginal groups at Cape York relied exclusively on foraging, fishing and hunting' (Harris 1995: 852). The lack of shared imagery found in Cape York rock art suggests that Cape York Aboriginal groups, although close exchange and intermarriage partners with south-Western Islanders, remained separate from the Torres Strait-Papua New Guinea artistic system. The location of the divide or separation between Cape York, and Torres Strait-Papua New Guinea is also similar to the spatial distribution of canoe rock paintings where images of canoes from Cape York were markedly different from those recorded in Torres Strait (details in Brady 2005). As a result, the imagery examined here indicates that artistic behaviour, involving distinctive design forms, correlates with the cultural divide between horticulturalists and hunter-gatherers.

[FIGURE 6 OMITTED]

Portable and fixed images will have had different meanings in their region. A rock painting executed on a granite boulder ora sandstone rockshelter represents a fixed symbolic marker that reflects social interaction or communication, and establishes a connection or relationship to a place. The distribution of portable artefacts and the symbolic images or designs they bear, also represents social interaction and exchange. Portable artefact distribution represents the spatial extent of design forms shared between rock art and material culture. Designs recorded on portable artefacts have been documented from collection locales that cross social and linguistic boundaries. This observation suggests that the interactive sphere involving exchange of distinctive imagery produced on portable objects circulating between Papua New Guinea and Torres Strait is much more extensive than previously anticipated, although Cape York remains on the periphery. The symbolism associated with the design on the portable artefact lies in its ability to reference social interaction through a shared material expression (David 2002). Distinctive designs are symbolic since they reflect a connection directly or indirectly--with other areas where that design has also been encountered. While the original meaning associated with the individual distinctive designs may be lost or forgotten, the spatial distribution of portable artefacts tells us that the region is a socially engaged environment where social interaction is reflected in the spread of distinctive design forms.

While the geographical or spatial extent of shared artistic language can be visually identified in the Torres Strait region and beyond, the temporal nature and origins of distinctive design forms is slightly more difficult to characterise. As noted above, there is a relative temporal similarity between rock paintings and wooden material culture objects. However, this similarity provides little indication of whether the circulation of images in the region has changed over time of if designs on one medium predate the other. Obtaining a temporal framework for production of designs remains a future research objective to fine-tune our understanding of artistic systems from the region. Additionally ascribing origins to the individual designs is problematic--it is very possible that different motifs had different origins and have since spread through the agency of travelling objects of people. There is no evidence at this time to suggest that the occurrence of distinctive designs in either rock-art or material culture also marks their origin.

As imagery that remains fixed in place, rock paintings are a convenient anchor point from which to conduct analyses of spatial patterning. Imagery that is locked in place offers considerably more information regarding the context and social dimensions associated with design forms. In contrast, designs documented on portable artefacts tend to lack context and details surrounding the manufacture and history of exchange of the object. However, as the results discussed here suggest, when designs recorded on different forms of media are studied together in the context of spatial distribution, they can provide a more comprehensive perspective on interregional interaction.

Acknowledgements

Thanks are extended to the Islander and Aboriginal communities from Torres Strait and Cape York for their invitation to work with them in documenting their cultural heritage and permission to reproduce images found in this paper. Funding for various aspects of this project was provided by the Australian Institute of Aboriginal and Torres Strait Islander Studies (G2002-6625, G2003-6758), ARC Discovery Grant DP0344070, and the Environmental Protection Agency's (Queensland) Cultural Heritage Incentive Program. Thanks to Bruno David, Tim Denham, Amanda Kearney, Ian McNiven and Lynette Russell for providing comments on an earlier draft, and to my referees and Martin Carver for their helpful suggestions.

Received: 21 December 2006; Accepted; 23 April 2007; Revised 21 December 2007

References

BARANDIARAN, I. 1973. Art mueble del paleolitico cantabrico (Monografias Arquelogicas 13). Zaragoza: Universidad de Zaragoza.

BEGOUEN, R. & J. CLOTTES. 1991. Portable and wall art in the Volp Caves, Montesquieu-Avantes (Ariege). Proceedings of the Prehistoric Society 57(1): 65-79.

BEST, A. 2003. Regional variation in the material culture of hunter gatherers: social and ecological approaches to ethnographic objects from Queensland, Australia (British Archacological Reports International Series 1149). Oxford: Archaeopress.

BRADLEY, R. 1997. Rock art and the prehistory of Atlantic Europe: signing the land. London & New York: Routledge.

BRADY, L.M. 2005. Painting patterns: Torres Strait region rock-art. Unpublished PhD dissertation, Monash University.

CARTER, M. 2004. North of the Cape and south of the Fly: the archaeology of settlement and subsistence on the Murray Islands, Eastern Torres Strait. Unpublished PhD dissertation, James Cook University.

David, B. 2002. Landscapes, rock-art and the dreaming: an archaeology of preunderstanding. London: Leicester University Press.

DAVID, B. & D. CHANT. 1995. Rock art and regionalisation in North Queensland prehistory. Memoirs of the Queensland Museum 37(2): 357-528.

DAVID, B., I.J. McNIVEN, L. MANAS, J. MANAS, S. SAVAGE, J. CROUCH, G. NELIMAN & L. BRADY. 2004a. Goba of Mua: archaeology working with oral tradition. Antiquity 78: 158-72.

DAVID, B., I.J. McNIVEN, R. MITCHELL, M. ORR, S. HABERLE, L. BRADY & J. CROUCH. 2004b. Badu 15 and the Papuan-Austronesian settlement of Torres Strait. Archaeology in Oceania 39(2): 65-78.

DEVLET, E. 2001. Rock art and the material culture of Siberian and Central Asian shamanism, in N. Price (ed.) The Archaeology of Shamanism: 43-55. New York & London: Routledge.

GLOVER, I. 1986. Archaeology in Eastern Timor, 1966-1967 (Terra Australis 11). Canberra: Australian National University, Research School of Pacific Studies.

HADDON, A.C. 1890. The ethnography of the Western Tribe of Torres Straits. Journal of the Anthropological Institute 19: 297-440.

--1893. The secular and ceremonial dances of Torres Strait. Internationales Archiv fur Ethnologie 6: 131-62.

--(ed.) 1904. Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 5. Sociology, magic and religion of the Western Islanders. Cambridge: Cambridge University Press.

--(ed.) 1908. Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 6. Sociology, magic and religion of the Eastern Islanders. Cambridge: Cambridge University Press.

--(ed.) 1912. Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 4. Arts and crafts. Cambridge: Cambridge University Press.

--1935. Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 1. General ethnography. Cambridge: Cambridge University Press.

--1946. Smoking and tobacco pipes in New Guinea. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences 232(586): 1-278.

HADDON, A.C. & W.H.R. RIVERS. 1904. Totemism, in A. Haddon (ed.) Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 5. Sociology, magic and religion of the Western Islanders: 153-93. Cambridge: Cambridge University Press.

HARRIS, D. 1995. Early agriculture in New Guinea and the Torres Strait divide. Antiquity 69: 848-54.

HATTE, L. 1992. 'Boring coastal stuff': rock images of the Townsville district, north Queensland, in J. McDonald & I. Haskovec (ed.) State of the art: regional rock art studies in Australia and Melanesia (Occasional AURA Publication 6): 71-5. Melbourne: Australian Rock Art Research Association.

LANDTMAN, G. 1927. The Kiwai Papuans of British New Guinea: a nature born instance of Rousseau's ideal community. London: MacMillan.

--1933. Ethnographical collection from the Kiwai District of British New Guinea in the National Museum of Finland, Helsingfors (Helsinki): a descriptive survey of the material culture of the Kiwai people. Helsinki: Commission of the Antell Collection.

LAWRENCE, D. 1994. Customary exchange across Torres Strait. Memoirs of the Queensland Museum 34(2): 241-446.

LAWRIE, M. 1970. Myths and legends of Torres Strait. St. Lucia: University of Queensland Press.

LEE, G. & W.D. HYDER. 1991. Prehistoric rock art as an indicator of cultural interaction and tribal boundaries in South-central California. Journal of California and Great Basin Anthropology 13(1): 15-28.

LEROI-GOURHAN, A. 1965. Prehistoire de l'art occidental. Paris: Mazenod.

LEWIS, A.B. 1931. Carved and painted designs from New Guinea (Anthropology Design Series 5). Chicago (IL): Field Museum of Natural History.

McNIVEN, I.J. 1998. Enmity and amity: reconsidering stone-headed club (gabagaba) procurement and trade in Torres Strait. Oceania 69(2): 94-115.

McNIVEN, I.J., B. DAVID &J. BRAYER. 2000. Digital enhancement of Torres Strait rock art. Antiquity 74: 759-60.

McNIVEN, I.J., F. VON GNIELINSKI & M. QUINNELL 2004a. Torres Strait and the origin of large stone axes from Kiwai Island, Fly River estuary (Papua New Guinea). Memoirs of the Queensland Museum, Cultural Heritage Series 3(1):271-90.

McNIVEN, I.J., B. DAVID, L. BRADY & J. BRAYER. 2004b. Kabadul Kula rock-art site, Dauan Island, Torres Strait. Memoirs of the Queensland Museum, Cultural Heritage Series 3(1): 227-56.

McNIVEN, I.J., W.R. DICKINSON, B. DAVID, M. WIESLER, F. VON GNIELINSKI, M. CARTER & U. ZOPPI. 2006. Mask Cave: red-slipped pottery and the Australian-Papuan settlement of Zenadh Kes (Torres Strait). Archaeology in Oceania 41: 49-81.

MOORE, D.R. 1984. The Torres Strait collections of A. C Haddon: a descriptive catalogue. London: British Museum Publications.

OMNES, J. 1982. La Grotte ornee de Labastide (Hautes-Pyrenees). Lourdes: author.

PARSONS, M.L. 1987. Plains Indian portable art as a key to two Texas historic rock art sites. Plains Anthropologist 32 (117): 257-74.

RIVERS, W.H.R. 1904. The regulation of marriage, in A. Haddon (ed.) Reports of the Cambridge Anthropological Expedition to Torres Straits. Vol 5. Sociology, magic and religion of the Western Islanders: 233-47. Cambridge: Cambridge University Press.

SHARP, N. 1992. Footprints along the Cape York sandbeaches. Canberra: Aboriginal Studies Press.

SHNUKAL, A. 1998. At the Australian-Papuan linguistic boundary: Sidney Ray's classification of Torres Strait languages, in A. Herle & S. Rouse (ed.) Cambridge and the Torres Strait: centenary essays on the 1898 Anthropological Expedition: 181-200. Cambridge: Cambridge University Press.

TACON, P.S.C. 1993. Stylistic relationships between the Wakeham Bay petroglyphs of the Canadian Arctic and Dorset portable art, in M. Lorblanchet & P.G. Bahn (ed.) Rock art studies: the post-stylistic era or where do we go from here? (Oxbow Monograph 35): 151-62. Oxford: Oxbow.

TACON, P.S.C., B. SOUTH & S.B. HOOPER. 2003. Depicting cross-cultural interaction: figurative designs in wood, earth and stone from south-east Australia. Archaeology in Oceania 38: 89-101.

WALKER, D. (ed.) 1972. Bridge and barrier: the natural and cultural history of Torres Strait. Canberra: Australian National University.

WILSON, M. 1999. Bringing the art inside: a preliminary analysis of black linear rock-art from limestone caves in Erromango, Vanuatu. Oceania 70(1): 87-108.

--2002. Picturing Pacific prehistory: the rock-art of Vanuatu in a western Pacific context. Unpublished PhD dissertation, Australian National University.

Liam M. Brady, Centre for Australian Indigenous Studies, Monash University, PO Box 55, Clayton, Victoria, Australia 3800 (Email: Liam. Brady@arts. monash.edu.au)