Real World Applications
Future Directions for Ceramic Analysis in Southeast Asian Archaeology
Inasmuch as I have in the course of this dissertation attempted to be comprehensive in my analyses, is it undeniable that the conclusions made at all levels suffer from investigative biases (even beyond those concerning the material of choice, as outlined above), which only magnify at the higher resolutions.
Each SEM-electron microprobe analysis area ranges from about 0.2mm2 to 1.4mm2. Even with four electron images per cut sherd, this still makes their total a subsample of a sample of a broken sherd, which is unto itself a subsample of (in the context of this study) an unstratified salvage deposit. While it may be hoped that intra-sherd variation remains below the error-margin of the analysis anyway (as wild fluctuations within the same fabric increase the risk of vessel failure during firing), this nevertheless remains an assumption.
How may these limitations then be resolved? It would seem unproductive to take multiple samples per sherd, at least with the current semi-quantitative regime imposed by the mechanical limitations of the SEM. There exist other options such as a specialised electron microprobe, such as that employed by RLAHA's tephrochronology laboratory (Smith, 2017), or perhaps even mass, absorption or emission-spectrometry for very fine distinctions in elemental proportions (Pollard and Heron, 2008, 21-41). Counter-arguments to using these are their cost and relative time-consumption, while the quicker SEM analysis also permits a visual study of the microstructure. Nevertheless, future pilot studies may aim to determine the utility of these other methods in characterising commonalities and differences in ceramic and glaze production.
Even then, such data would be limited in value without a similar basis for comparison; Wood (1999) cites all his oxide proportions to 1 decimal place, and increasing resolution of new measurements beyond those would not be useful. In order to make a truly instructive conclusions from such data once a preferred method is settled upon, a whole suite of material from a variety of contexts —Chinese and Southeast Asian, terrestrial and shipwreck — will have to be sampled for body and glaze compositions, perhaps finally resolving mysteries concerning ceramic traditions and transmission thereof in both knowledge and produce; the addition of data from local kilns will make considerable progress towards overturning Sinocentric models and assumptions concerning high-fired ceramics.