2.3.1 Stylistic Analysis
According to Renfrew and Bahn (2004, 427), ―style is any distinctive and therefore recognisable way in which an act is performed or an artefact made‖. To be more precise, objects made during a certain period, at a specific location, by a particular group of people, typically reflect individualistic, recognisable style (Renfrew & Bahn 2004, 124). In addition, artefact variability within a specific artefact category is explained by Heinze (2015, 94) as follows:
Artifactual variability within a given category, e.g. the different forms and styles of a dagger, is a result of this relationship between form (including material possibilities) and cultural conceptualizations of what daggers looks like (style) and what one can do with it (function). Style and function are, however, not two different dimensions of an object that coexist next to each other, with style adding an aesthetic dimension to the basic form of an object defined by its functional dimension. Rather, style is to be considered a function in itself as it allows for the sending of messages, is communicative and part of one‘s symbolic capital, with cultural concepts at the heart of every judgment on what is beautiful, desirable, valuable, and allowed.
Thus stated, stylistic analysis goes above and beyond the mere determination of style for artistic or aesthetic value. It is a qualitative endeavour in its own right, which holds the possibility of elucidating information about the object‘s chronology or typological sequence, and therefore authenticity. In the words of Renfrew and Bahn (2004, 126), ―different types of artefact change in style (decoration and shape) at different rates, and therefore vary in the chronological distinctions that they indicate‖. Therefore, if an object‘s stylistic attributes can be assigned (with relative certainty) to a specific time period, we can confirm its authenticity with a fair amount of certainty.
Since stylistic analysis is highly comparative, in the sense that unknown objects have to be compared to objects that have been positively curated, the analysis performed within the context of this thesis will rely heavily on existing published works, as well as online archival sources, museum databases and auction catalogues (as mentioned before). Because of the visual nature of this type of analysis, the thesis will provide a comprehensive series of in-text colour images, with the exception of a few grey scale images obtained from publications. The provision of colour images is of utmost importance, as colour profiles, variations and
30
similarities form part of the datasets of this thesis. In addition, certain textural phenomena, shading patterns and other physical visual phenomena are often lost when converted into grey scale.
2.3.2 Surface Investigation: SLR Photography and Digital Microscopy
High resolution photographs were taken using a Canon EOS 400D digital SLR Camera, which offers 10MP at a maximum resolution of 3888 x 2592 DPI. The camera features an EFS 18–55 mm wide-angle lens with a 0.25 mm close-focusing distance, with the latter specifications ideal for object photography. Although the camera comes with a built-in flash, basic guidelines for museum collections photography stipulate that flash photography should be avoided. Thus stated, objects were placed within a home-made light-box that utilises three individual, moveable light sources that ensure uniform illumination (ambient light). To avoid the unwanted formation of shadows, the moveable light sources could be adjusted to accommodate individual objects and the unique shadows they cast.
Digital microscopy was performed using a handheld Celestron Digital Microscope Pro. The 5 MP CMOS digital camera captures high resolution images through a 5-Element IR cut high- quality glass lens. The microscope has a 20x–200x power capability and connects to any computer using a USB port21. An adjustable LED light source offers the user the opportunity to regulate the direct light intensity to either reduce glare or highlight certain features.
Digital image manipulation, of images produced by both microscopy and photography, were kept to a minimum to ensure that colour variations and textures were faithfully represented. Image resizing (with aspect ratios maintained), the removal of unwanted backgrounds and the addition of digital photo scales represent the limit of image manipulation.
2.3.3 3D Visual Analysis: Microfocus X-ray Computed Tomography
Through the generous cooperation of the DNMCH, three-dimensional radiography was performed at NECSA‘s mixed radiation (Microfocus X-ray Radiography and Tomography) facility (MIXRAD). The Nikon XTH 225/320 LC22 dual source industrial system has a voltage setting of between 30 and 225kV, with beam currents ranging from 0 to 1mA and a maximum power output of 30W (Hoffman & de Beer 2012, 2).
21 System specifications can be found at http://www.celestron.com/browse-shop/microscopes/digital-
microscopes/handheld-digital-microscope-pro [accessed on 29 August 2015].
22 Product information can be found at https://www.nikonmetrology.com/en-gb/product/xt-h-225 [Accessed
31
Although the process of 3D nuclear imaging is quite complex, what follows is a brief explanation, from a layman‘s perspective, of how MXCT works (Fig. 2.3). A radiation source (A) emits a microfocus, stable X-ray beam (225kV with a 3µm focal spot size) (B) directed at a target/object (C). While the beam source (A) and detector (E) remain stationary, the object (C) is placed on a rotating base to ensure that tomograms are captured along a 360-degree rotation. X-rays are then attenuated (a reduction in flux intensity through a medium) based upon the object‘s material composition, with different materials exhibiting variable attenuation rates based upon their thickness, density and chemical composition23. These attenuated beams (D) exit the object and are captured by an x-ray sensitive detector (E) that captures the raw data and transfers it to a data capturing and processing computer (F).
Figure 2.3: Example of a MXCT beam-line setup. A radiation source (A), microfocus x-ray beam
(B), object (C), attenuated x-ray beams (D), x-ray sensitive detector (E), computer (F).
Since hundreds of individual tomograms are captured as the object rotates, any form of movement must be avoided during the 30-minute X-ray session. To ensure stability, objects are placed on pedestals made of polystyrene or florist foam, which appear invisible (radiotransparent) on tomograms due to their almost non-existent attenuation values. Using specialised image rendering software (VG-Studio Max), a three-dimensional model of the object is compiled from hundreds of individual tomograms. Once the digital reconstruction is complete, the composition is saved as a ―project‖, and the end-user is able to view objects
23 See https://en.wikipedia.org/wiki/Attenuation [Accessed 09/09/2018] and Escudier and Atkins (2019, 12). It
is also noted that the absorption of radiation by a target material follows an exponential decay function. Exponenetial decay is where one observes a ―decrease in a quantity at a rate that is proportional to its value‖ (Schaschke 2014: 204). Thus said, for some materials (depending on the thickness), it will not help to increase the x-ray intensity, as no amount of energy will be able to pass through the material. This fact should be kept in mind during the examination of the Egyptian bronze collection, as it explains why x-rays are limited to providing lower resolution images compared to neutron tomography.
32
using a simplified version of the compilation software, known as myVGL24. The user is then able to rotate the model and pan through the object from any direction, allowing for virtual interaction and non-destructive evaluation.
The material composition and sample size of the objects being scanned dictate the parameters entered into the system. Depending on the individual material‘s attenuation, the kV settings of the system have to be adjusted in order to ensure satisfactory beam penetration. For example, as bronze is known for its high attenuation of X-rays, the voltage was set at 205kV when examining the collection of Egyptian bronzes. As high density materials such as copper and lead (the main components of bronze alloy) offer high levels of resistance to the penetration of X-rays (De Beer 2005), copper filters (which act as monochromating agents by filtering out lower energy x-ray photons) were placed at the beam source in order to reduce possible beam scattering and possible post-reconstruction beam hardening. These measures were unnecessary when it came to the processing of the other objects (helmet, mask, gauntlet and dagger), as their compositions – mainly iron, steel and silver – presented less challenging attenuation conditions.
Beam scattering is an unwanted – yet often unavoidable – occurrence in nuclear imaging. Along broad scientific terms, scattering is defined as,
…a general physical process where some forms of radiation, such as light, sound, or moving particles, are forced to deviate from a straight trajectory by one or more paths due to localized non-uniformities in the medium through which they pass. In conventional use, this also includes deviation of reflected radiation from the angle predicted by the law of reflection25.
Beam scattering may in some cases lead to a ghost-like haze surrounding an object that makes it difficult to define the exact borders of an object or distinguish transitional zones between materials types within the same object. In addition, beam hardening may also result in unwanted visual anomalies. Brooks and Di Chiro (1976, 391) explain that if x-ray pass through matter:
24
myVGL is a free-to use software package developed by Volume Graphics and is available to researchers at https://www.volumegraphics.com/en/download-viewer.html [Accessed on 07/09/2018].
25 https://en.wikipedia.org/wiki/Scattering [Accessed on 09/09/2018]. Also refer to the following page for an
illustration of electromagnetic scattering: https://cso.kmi.open.ac.uk/topics/electromagnetic%20scattering [Accessed 01/11/2018].
33
Low energy photons are preferentially absorbed, and the (logarithmic) attenuation is no longer a linear function of absorber thickness. This leads to various artifacts in reconstructive tomography, [but fortunately] artifacts can be remedied by the additional prefiltering of the beam and by applying a linearization correction to the detector outputs.
With these factors considered, it is of utmost importance that researchers who are new to nuclear imaging consult with trained professionals in the field who are able to distinguish between visual anomalies and active diagnostic features. In all instances, the MIXRAD team assisted the researcher with basic visual interpretations during project compilation.