There has been a longstanding debate among archaeologists over the purpose of Bronze Age (c.2500-800 BC) hoards, particularly those including objects that appear to have been deliberately broken. early theories suggested that they were purely functional: created either for temporary storage, recycling, or for actual ‘hoarding’ in times of strife. More recently, though, many archaeologists have ascribed a more ritualistic meaning to the practice: perhaps buried as religious offerings or as displays of social status. until now these ideas have mostly been based on the descriptive analyses of hoards’ characteristics. A new study, however, addresses this debate using mathematical modelling.
Rob Wiseman from Oxford Archaeology created a model that describes groups formed through the random accumulation of objects, as well as the outcome of randomly breaking objects into pieces. He then ran the model against the quantitative characteristics of 174 Bronze Age hoards discovered in England and Wales since 1995 – all of which had been recorded through the Portable Antiquities Scheme. The results showed that the model was a good fit for three-quarters of the hoards, particularly those from the Late Bronze Age. This suggests that the majority were the result of unintentional formation and not through deliberate selection.
Rob also used the model to estimate the accumulation rate needed to produce the distribution of hoard sizes. The findings indicate that approximately 85%-95% of all bronze objects in circulation during the Late Bronze Age would eventually have been broken and deposited in scrap hoards. This high rate suggests that they were unlikely to have been created for permanent disposal, as this would have quickly consumed Britain’s entire supply of bronze. Instead, the results suggest a process of deposition and reuse – more likely serving a functional purpose than a ritualistic one.
Rob stresses that this model does not explain every type of hoard, only collections of broken scrap metal. There are other, quite different hoards that the model does not explain, such as those of unbroken weapons, or objects deposited in rivers and fens from causeways or in funerary contexts. nonetheless, the model does address the bulk of hoards found in the past 20 years, and provides a new way of analysing groupings of objects in the past.
This article appeared in CA 339.