This thesis sets out to explore, describe, quantify, and visualise uncertainty in a cultural informatics context, with a focus on archaeological reconstructions. For quite some time, archaeologists and heritage experts have been criticising the often toorealistic appearance of three-dimensional reconstructions. They have been highlighting one of the unique features of archaeology: the information we have on our heritage will always be incomplete. This incompleteness should be reflected in digitised reconstructions of the past. This criticism is the driving force behind this thesis. The research examines archaeological theory and inferential process and provides insight into computer visualisation. It describes how these two areas, of archaeology and computer graphics, have formed a useful, but often tumultuous, relationship through the years. By examining the uncertainty background of disciplines such as GIS, medicine, and law, the thesis postulates that archaeological visualisation, in order to mature, must move towards archaeological knowledge visualisation. Three sequential areas are proposed through this thesis for the initial exploration of archaeological uncertainty: identification, quantification and modelling. The main contributions of the thesis lie in those three areas. Firstly, through the innovative design, distribution, and analysis of a questionnaire, the thesis identifies the importance of uncertainty in archaeological interpretation and discovers potential preferences among different evidence types. Secondly, the thesis uniquely analyses and evaluates, in relation to archaeological uncertainty, three different belief quantification models. The varying ways that these mathematical models work, are also evaluated through simulated experiments. Comparison of results indicates significant convergence between the models. Thirdly, a novel approach to archaeological uncertainty and evidence conflict visualisation is presented, influenced by information visualisation schemes. Lastly, suggestions for future semantic extensions to this research are presented through the design and development of new plugins to a search engine.

Analytical survey of the above ground evidence has been undertaken on twelve areas of prehistoric fields in southern Britain. In all cases at least two phases were noted, one directly overlying the other; in ten of these areas the earlier phase comprised an extensive rectilinear grid and the later smaller areas of aggregated fields. The earlier field systems could be externally bounded and left little land unenclosed for open grazing and timber production, movement was only allowed along high ridges. It is suggested that the earliest of these fields date to the beginning of the 2nd millennium, on both sides of the Channel, where they were regarded as symbolic of status within a period of visible ostentatious possessions. The majority were created in the middle centuries of that period, possibly as a reaction to perceived land pressure. No settlements could be identified as coeval with these fields. The later fields represent a major contraction of enclosed land and their design is more suited to stock, rather than arable, production. Larger areas around the fields were marked by linear ditch systems or by cross ridge dykes. Settlements were frequently, and presumably deliberately, placed over the boundaries of the earlier fields, possibly in an act of incorporation; these settlements tend to date to the two centuries on either side of 1000BC, and it is likely, though not certain, that the later fields were contemporary with these settlements. The production of stock as evidence of wealth led to feasting, as exemplified by midden sites, and to a raiding culture within which aggression is more likely, but warfare not proven. The point is made that, with no structure visible at excavation across lynchets, analytical survey is the best method of recording phase differences. Also, given the lack of below ground evidence these sites, though widespread, are a diminishing resource and protection of the best examples is highly desirable.

Deteriorating climate in the period leading up to the Last Glacial Maximum (LGM) 20,000 years ago caused global sea levels to fall to a lowstand of 125m below modern levels. This resulted in the recession of the Sicilian palaeoshoreline by up to 150km and the emergence of vast tracts of coastal lowland. Following climate amelioration and deglaciation, rising sea levels inundated these formerly exposed areas. The earliest indication of a modern human presence on Sicily comes from Fontana Nuova, in the southeast of the island. The timing of this occupation, on the basis of cross-dating of Aurignacian lithic typology, is about 37,000 BP – a time when sea levels around the Sicilian coastline were some 40-80m lower than at present. The oldest scientifically-dated human remains come from Addaura Caprara, dated to 15,643–15,177 cal BP. Some archaeologists interpret the evidence as suggesting a brief, solitary visit to the island at around 37,000 BP followed by a gap of some 20,000 years before the establishment of a permanent presence during the 16th millennium BP. Others dismiss the veracity of the evidence from Fontana Nuova and hold that Sicily was never settled until some considerable time after the LGM. Until late 20th-century studies demonstrated the attractions of coastal ecotones, absence of evidence was sometimes interpreted as a rejection of coastal landscapes by Upper Palaeolithic and Mesolithic hunter-gatherers. My thesis will argue that Sicily was not uninhabited for 20,000 years or more but that any evidence for human coastal presence throughout prehistory is potentially submerged. I consider the attractions of coastlands for early modern humans. I discuss the physical background to glaciation and deglaciation resulting in sea-level change. By combining data on absolute sea-level change with evidence for terrestrial displacement resulting from tectonic forces, I have determined relative sea-level change affecting the coastlines of Sicily from 37,000 BP until the Iron Age (ending c750 BC). The results have been combined with digital bathymetric data within an ESRI ArcMap GIS program to produce a series of maps at archaeologically-significant dates. The very areas that are now recognised as being attractive to modern humans will be shown to be submerged today. With reference to modern scientific techniques and their application by specialists in a variety of locations, I shall demonstrate that the successful recovery of submerged archaeological evidence is achievable. I shall also consider a number of phenomena revealed by my field observations that have conspired to conceal or destroy the coastal archaeological record, the absence of such evidence being used illegitimately to support claims for an unpopulated island. These phenomena include natural catastrophes such as earthquakes and related tsunamis, volcanic activity and landslides. Additionally, coastline modification resulting from river estuary migration, and anthropogenic impacts will be considered.

n/a

No / Recent research by Andrew Reynolds has interrogated the archaeological record for evidence of Anglo-Saxon execution cemeteries (Reynolds 2009). This paper will discuss how osteological evidence can aid our interpretation of Anglo-Saxon capital punishment and give insight into the type of evidence that might aid in the identification of corporal punishment from skeletal populations. The importance of correctly interpreting skeletal trauma is essential, but this can be supported by scrutinising the palaeodemographic profile of execution populations, burial position, an understanding the decomposition process and the significance of post-depositional disturbance of burials. It will lay down a framework for the successful identification of corporal and capital p unishments, with reference to Anglo-Saxon documentary sources. / Full text of the author's final draft is unavailable due to copyright restrictions.