Iron Age and Roman copper alloys from northern Britain

Abstract

This paper presents the chemical analysis of over 1500 Iron Age and Roman copper alloy artefacts. The data are considered as a whole and in the light of a number of key archaeological variables: chronology, typology, provenance. Patterns evident from the analyses carried out have enabled the nature of alloys to be reliably characterised. Iron Age alloys are almost exclusively of one type only - a tin bronze which often contains a small amount of arsenic (up to 1%). This alloy is quite distinctive when compared with the range of alloys used in the Roman period (brass, bronze and gunmetal, all with varying amounts of lead). The contrast between Iron Age and Roman alloys allows a reconsideration of many 'Celtic' items. It is now clear that the majority of 'Celtic' metalwork which survives dates to the end of the Iron Age or the Roman period (despite the traditional equation of the Iron Age with 'Celtic' material). The most successful means of representing the range of Roman alloys used has been the three-dimensional plot. This examines the relationship between two elements (in this case zinc and tin) and shows the relative frequency of the different alloy types. This 3-D plot also illustrates some general features of Roman alloying and possible recycling. There are a number of peaks in the distribution of zinc and tin contents that represent specific alloy types which were commonly produced. The largest peak relates to the commonest alloy: bronze. The second peak relates to brass and the third to copper. All the remaining analyses fall into a diffuse area between the bronze and the brass peaks, and these are referred to by the modern term gunmetal. It is clear that copper alloys were recycled and that some care was taken over the ways in which this occurred. The lack of low zinc brasses shows that this alloy was rarely recycled on its own. If brass was recycled then it was always mixed with some bronze. The proportions of bronze and brass that were mixed varied widely as there is no distinct peak within the distribution of the zinc and tin contents of gunmetals. The paper uses an overall view of Roman copper metallurgy (and that derived from previous work) to examine changes in alloy production and use. The results (interpreted in the light of metallurgical theory and practice, and of wider archaeological theory and data) challenge many of the traditional accounts of chronological and cultural change, and of the deposition processes operating in the Roman period. Access to the data is provided by means of a clickable map, a typological catalogue, a site catalogue or by an individual record.