• 2020-2024close
• Open Access close
• Closed Access close
• Publications close
• Research software close
• Other research products close
• AU close
• Sydney eScholarship close
• Digital Humanities and Cultural Her... close

This thesis has used the abandoned village site of Mwanihuki located on Makira in the Solomon Islands as a case study to explore the archaeological evidence associated with the initial occupation of the region and the emergence of later inter and intra-island trade and exchange systems. The main aims of the thesis were to establish the timing of initial occupation, understand patterns of settlement and subsistence, and explore evidence for the rise of the southeast trade and exchange system and Mwanihuki’s place in that network. Intensive fieldwork was conducted on Mwanihuki and surroundings, while also analysing and incorporating the legacy material into this research. Radiocarbon evidence revealed phasing that placed the study area into two broad cultural periods. The first was an ephemeral use of Mwanihuki from c. 3000 BP, which was contemporaneous with the Lapita cultural tradition, though aceramic. The second phase of the site demonstrated intensive occupation from c.800BP, which included construction of burial structures, anthropogenic refuse mounds, and a rich material culture. It is argued that these latter items are evidence of shell valuable production and together with the dense concentrations of chert imported from Ulawa indicate that by c500BP Mwanihuki was a significant node in the emergent inter-island trade system. This material culture, along with Mwanihuki’s prominent headland location and strong island inter-visibility all contributed to the transfer of material culture and social and economic complexity. The abandonment of the site c400BP and the retreat to the mountainous interior and defended settlements detailed in oral history appears to be a consequence of an initial contact with Spanish explorers in 1595 AD and the rise of inter and intra island hostilities.

This thesis investigates the competition between different material-semiotic translations of pain in the period of neoliberalism. Moseley and Butler (2017) published a novel pain theory, which understands pain as emergent from a complex system, in which biological, psychological, and social elements interact. The "Pain Revolution" is a practical implementation of the theory, but this thesis uses the phrase to refer to a wider paradigm shift that is unrealised. This thesis synthesises actor-network theory, Bourdieusian field sociology and narrative theory, and describes a competition between different material-semiotic translations of pain. It demonstrates that Moseley and Butler’s theory has not become imbricated in the networks of major institutions and their translations of pain, nor become predominant in the pain field. To understand why, this thesis traces the coevolution, from the mid 19th century to 2020, of an “economy of responsibility” and a competition between different translations of pain. I establish that this economy of responsibility has been constituted through complex interactions between juridical, insurantial, and professional elements, and been coextensive with a network of body-mind dualism that evolved through liberal, welfare state, and neoliberal periods. We will find that the aforementioned institutions and their translations of pain align with and help to sustain a neoliberal version of an economy of responsibility and network of body-mind dualism. The Pain Revolution is shown to be incommensurable with the juridical, insurantial, and professional logics operating in these networks and so it has not become imbricated with them. I demonstrate that the pain field, rather than operating as a scientific subfield marked by closure and autonomy, has been open to the heteronomous logics of medical, legal and insurance fields. The Pain Revolution has not taken hold in the pain field because it does not fully align with these logics.

Ancient Chorasmia, located on the northern frontiers of Central Asia, south of the Aral Sea, was a polity barely known to the Classical world despite some scarce mentions in a few written sources. Chorasmian civilisation and material culture were initially recognised and defined by modern archaeological works carried out in the area since the early 1940s by the Soviet Archaeological-Ethnographic Expedition to Ancient Chorasmia led by S.P. Tolstov. While numerous archaeological sites have provided rich materials regarding multiple aspects of ancient Chorasmia, i.e., the construction of fortifications, ceramic typology and religious practices, the overall lack of a well-stratified sequence and solid chronological evidence has been a striking issue for most of the excavated sites. This is particularly problematic in the study of pottery which could easily be subject to arbitrary judgments without secure stratigraphy and dating. The excavations at Akchakhan-kala (1995-present) by the Karakalpak-Australian Archaeological Expedition have yielded well-stratified pottery sherds with chronological sequences spanning from the early 2nd BCE to 2nd Centuries CE. The site was identified as a royal seat closely linked with monumental and religious purpose, as attested at the Central Monument and Ceremonial Complex, where painted texts testifying to the presence of a “Chorasmian King” were initially found. It provides a precious chance to review the existing pottery typology and chronology of ancient Chorasmia in term of new evidence acquired from Akchakhan-kala, and cast new light on the periodization of the antique period of Chorasmia. Apart from reviews of archaeological sites and the ceramic chronology of ancient Chorasmia, the thesis provides a thorough analysis of the Akchakhan-kala assemblage, concerning fabrics, typology and chronology, presented by illustrations and catalogues.

Many societies across the world see birds as providers of information – be it environmental, cultural, or symbolic. In Central Australia, birds are seen by Aboriginal people as referents. One way in which central Australian Aboriginal people ‘know’ of monsters is through the visual, acoustic and sensory presence of birds: distinctive calls, fleeting movements, camouflaged sightings, scratched tracks and the sensation of being ‘watched’ are qualities displayed in uncannily similar ways by various species of birds and their monstrous counterparts. Whilst some birds warn of monsters and some accompany them, here I focus on a type of monster I call bird/monsters. They appear as ancestral beings in the songs and associated Dreaming narratives of Warlpiri people, who traditionally lived in the Tanami Desert and today live in towns fringing the Tanami as well as further afar. Bird/monsters are understood to be male figures that, at once, are both men and birds and exist amongst other ancestral beings which take on the form described by Rose (2011: 122) as “shape-shifters, sometimes walking as humans, sometimes travelling in the form of the being they would become.” Being birds and men simultaneously also distinguishes them from classical hybrid figures such as centaurs (part man, part horse) and werewolves (sometime person, sometimes wolf). What is clear is that like all monsters, bird/monsters defy easy categorisation (Cohen 1996). The two-part terminology I apply when describing them as bird/monsters reflects both this and their ability to move between different realms. Cohen suggests that “the ways in which [monsters] shift and refuse definition is what makes them so feared” (1996:6). The spiritual associations that birds have to Warlpiri people and their ever presence in their environment link them closely to the human realm, yet the immoral and culturally inappropriate acts of the monsters they embody continue to make this categorisation uneasy. As I show, despite potentially becoming more human-like these bird/monsters do not play by the rules of the human world, a factor which enhances their power to control and frighten. I begin by presenting portraits of four bird/monsters and explain how I understand them to be ‘monsters’. My main focus is on showing how the manifestations of bird/monsters as immoral, socially inept, violent and culturally defiant monsters highlight deep-seated social fears. The stories of these bird/monsters are passed on and made known to Warlpiri people through Dreaming narratives and songs, intimately linking them to fundamental and highly valued components of Warlpiri cultural heritage. I demonstrate how bird/monsters continue to have monstrous signification even when what is feared has changed. These bird/monsters continue to invoke fear in contemporary contexts marked by the widescale social changes associated with neo-colonialism and increased connections to a broader and more globalised world. Contemporary fears are concerned with loss—of connections to country, of traditional patterns of social organisation, of control over women’s sexuality, and of the gendered forms of sociality which have until recently typified Warlpiri life.

Dispersed habitation patterns have long been noted as a feature of pre-industrial Mainland Southeast Asian settlements and recent research at Greater Angkor in central Cambodia has demonstrated that, at its height, this settlement was a massive low-density complex of dispersed temples, habitation mounds, and water features spread amongst agricultural fields. However, little is known about how common large settlements with low occupation densities were across Mainland Southeast Asia more broadly. This study begins to address this lacuna by marshalling available data on settlement forms from sites across the Mainland over the course of the first and second millenniums CE. With a focus on settlement scale, density, and morphology, this work demonstrates that a shift towards low-density configurations was a common trajectory associated with the initial development of large-scale (>100ha) settlements across the Mainland, but that a new trend towards the formation of larger areas of relatively dense habitation occurred during the second half of the second millennium CE. These region-wide trends indicate the need to consider settlement form as a factor in the creation large-scale historical outcomes in the region and to incorporate the spatial patterning of settlements into broader understandings of historical change in Mainland Southeast Asia.

With over 200 peer-reviewed papers published over the last 20 years, 3D modelling is no longer a gimmick but an established and increasingly common analytical tool for stone artefact analysis. Laser and structured light scanning, photogrammetry, and CT scanning have all been used to model stone artefacts. These have been combined with a variety of different analytical approaches, from geometric morphometrics to custom reduction indices to digital elevation maps. 3D lithic analyses are increasingly global in scope and studies aim to address an ever-broadening breadth of research topics ranging from testing the functional efficiency of artefacts to assessing the cognitive capabilities of hominid populations. While the impact of the computational revolution on lithic analysis has been reviewed, the impact of 3D modelling on lithic analysis has yet to be comprehensively assessed. This paper presents a review of how 3D modelling in particular has impacted the field of stone artefact analysis. It combines a quantitative bibliometric analysis with a qualitative review to assess just how “revolutionary” 3D modelling has been for lithic analysis. It explores trends in the use of 3D modelling in stone artefact analysis, its impact on the wider lithic analysis field, and methodological, regional and theoretical gaps which future research projects could explore.

The field of Natural Language Understanding (NLU) concerns the mapping of textual utterances to representations of meaning upon which a machine can act. The standard representation is a dual level semantic frame; a sentence level intent and a token level annotation of the words semantically important to the fulfilment of the intent. The joint task of intent detection and slot labelling defines the framework of experimentation in NLU. There are two problems with the current state of the field. Firstly, the two tasks should inform each other, but the models in the literature rely on implicit sharing of information, or if it is explicit it is unidirectional from one task to the other. Further, the dominant data sets used are single turn, that is single standalone sentences. However the natural environment for NLU should be multi-turn conversations between the human and the machine, with the topic of discussion changing over time. In this thesis, we firstly design an architecture that incorporates an explicit, bi-directional exchange of information between the tasks during the learning process and outperforms the existing state of the art. We then adapt the annotation on two multi-turn conversational data sets used for dialogue state tracking to form NLU data sets. This offers an extra level to the semantic frame, that of a domain which changes during the conversation. We then pioneer NLU on this extended framework with the design of an explicit, tri-level model and show that the inclusion of an extra task improves performance of the dual level tasks. We design a multi-turn dual-level data set for toxicity detection in in-game chat. We are among the first to show that NLU architectures can be used to successfully analyse problems from other NLP fields when cast in the NLU framework.

In this thesis, we propose novel deep learning algorithms for the vision and language tasks, including 3D scene graph generation and dense video captioning. The dense video captioning task is to firstly detect multiple key events from the untrimmed video and then describe each key event using natural language. The 3D scene graph generation task is to segment the object on an indoor 3D scene and then predict the predicates between every two objects. The main contributions of this thesis are listed below. Firstly, we formulated the dense video captioning task as a new visual cue-aided sentence summarization problem and proposed a new division-and-summarization (DaS) framework for this task. In the division stage, we generate multiple sentence descriptions for describing diverse visual content for each event proposal. In the summarization stage, we propose a two-stage Long Short-Term Memory (LSTM) network equipped with a new hierarchical attention mechanism. Secondly, after having the DaS framework for dense video captioning, we further proposed a GCN (Graph Convolutional Network)-enhanced summarization module for sentence summarization. It utilizes the word relationship to refine the hidden representations of the generated sentence. Specifically, we treat the semantic words as the nodes in a GCN graph and learn their interactions via the tightly coupled GCN and LSTM networks. Finally, we proposed a newly designed position-aware two-branch framework for the 3D scene graph generation. In this two-branch framework, we additionally propose the position-aware branch, which contains explicit and detailed position information for relationship modeling. In addition, we propose a Switchable Multi-stage fusion Graph Transformer (SMGT) for progressive and effective two-branch fusion. In SMGT, multiple fusion modules with fusion module selection procedure are applied inside each layer of our graph transformer.

This thesis examines the issue of Aboriginal cultural heritage management and governance by Aboriginal people in the context of dispossession. Partnering with the Bogan River Wiradjuri Traditional Owners Corporation, an Aboriginal community in Parkes Shire, in the Central West of New South Wales, this research examines the role that the violent dispossession of Aboriginal people and their forced incarceration on missions and reserves, played in the current disenfranchisement of Aboriginal people from management of their own heritage. It demonstrates that the manner in which Aboriginal heritage is currently being managed is continuing that dispossession and the associated disconnection between cultural management and community expectations and aspirations. Research into Aboriginal governance has highlighted a number of reasons why Aboriginal people have become disempowered from management of their own heritage. The issues examined include outdated legislation, the role of the State in archaeological heritage management; the authority of archaeologists and the multiplicity of meanings contained in the word “heritage” created through the politicisation of heritage. Contributing to these issues are the poor educational outcomes for Aboriginal people which contributes to a sense of powerlessness within the Aboriginal community. This thesis will demonstrate that the impact of colonisation and post-colonial legislation and cultural heritage management processes have impacted on the ability of Aboriginal people to determine their own cultural heritage. This has included the educational disadvantage and ongoing failure of our education system for Aboriginal people which has contributed to the creation of barriers to Aboriginal led management. Quotes from historic sources have been used throughout this thesis. The reader should be aware that the language used in many of these quotes is racist, abhorrent and offensive. It has been used to demonstrate the attitudes of the times that led to or reflected discriminatory and racist policies and the oppression under which Aboriginal peoples suffered. The author does not support or agree with such language and I apologise for the offence that these quotes cause.