The economics of Late Bronze age gold mining by the Egyptian New Kingdom in Nubia

This paper provides the first quantitative estimates of the profitability (in the sense of reward for effort) of gold mining in Antiquity. We have developed and assessed four models for New Kingdom gold mining in Nubia, during this period of peak gold production. These are: mining from oxidised lodes (OLM), proximal alluvial placer deposits in desert wadis (PAP), alluvial placer deposits along the Nile (NAP), and alluvial clast mining (ACM). The last refers to selective extraction of gold-bearing clasts (“quartz chunks”) from wadi alluvium and has become an archaeological focus in Nubia. We examine two variants: proximal to the source lodes (ACM-P) and distal in larger catchments (ACM-D). Daily worker productivity has been estimated for each step in the chaîne opératoire using parameters drawn from: records of 19th century miners in isolated locations and reliant on simple technologies; modern artisanal miners; recent observations in Nubia and along the Nile; and limited experimentation. Gold production in each model is standardized to a 50-day expedition of 50 people. Under our base-case parameters all models, except for distal clast mining (ACM-D), are profitable: they produce more gold than required to pay the workers. The main factor is the low cost of labour in New Kingdom Egypt, when priced in gold. However, only the placer models (PAP and NAP) can generate large returns within realistic modelling parameters, because their free gold particles can be simply recovered by screening and washing. In contrast, lode and clast mining methods (OLM and ACM) must allocate most of the workforce to crushing and grinding ore to liberate gold before recovery. Clast mining is further disadvantaged by barren sediment dilution and is only viable in small catchments close to source lodes. New Kingdom Egypt expanded into previously unmined areas of Nubia and along the Nile and therefore enjoyed “first-mover” status over large areas of richly endowed goldfields. By analogy with virgin goldfields in Australia and California, where substantial placer gold dominated earliest production, they had access to numerous unworked deposits with grades perhaps significantly higher than assumed in our models. A surge in New Kingdom gold production was likely, followed by an inevitable decline after depletion of best resources. • Quantitative models constructed for four methods of gold mining in Antiquity. • Inputs include estimates for daily production at each step of the Chaîne opératoire. • Outputs demonstrate most methods yield positive returns (are profitable). • Labour costs and particularly fine grinding control profitability. • Placer models provide best returns with realistic parameters owing to simple processing.