Abstract

By the latter part of the first millennium BCE, glassmakers on the Levantine coast were making their glass from a mixture of coastal sand and naturally occurring soda from Egypt.  This raw glass was made in large tank furnaces on a scale of many tonnes and shipped in lumps around the ancient world, where it was re-melted and converted into artefacts.  This extremely efficient mode of production allowed a significant up-scaling of the industry, accommodating developments in shaping processes such as slumping and blowing, and allowing the furnishing of individual large Roman public buildings with many tens of tonnes of glass window and mosaic.  This so-called Roman glass dominated production until the eighth-ninth centuries, when new types of glass based upon the use of the ashes of burnt plants and trees instead of natron become dominant in the Mediterranean and in northwestern Europe.

Our understanding of the structure of the glass industry has moved forward mainly due to evidence from excavation of production sites combined with the compositional analysis of glass using geochemical techniques.  It is possible to distinguish at least nine compositional groups representing different primary production centres in the eastern Mediterranean and their durations range from several hundred years to as little as a few decades.  Evidence for the production of raw glass in Europe is at best ambiguous, suggesting that this may not have occurred at all. Some regions appear to have preferentially received specific types of glass, potentially providing insights into contact and trade.  Evidence for recycling is frequent and by the ninth century CE the stripping of glass from large buildings appears to have occurred. On a different scale, individual production events (“sets” or “batches”) can be identified, allowing intentionality and use to be inferred.

Coupled with rigorous typological and contextual analysis, the analysis of major elements, trace elements and isotopes has the potential to transform the role of glass as a tool in archaeological interpretation, but the resources needed are expensive.