Figure 65: The coinage 'fingerprint' for the VASLE National dataset
In order to understand what the presence of coinage on a site may imply, it is imperative to be aware of the nature of its circulation. Variations in the numbers of coins found on a site may be more indicative of changes in the circulation of coinage than changes in activity on that site (Reece 1987, 71-80). Therefore by dividing coin issues into groups by date, a better understanding of the chronological evolution of coin use can be ascertained. This section follows on from work on Iron Age (Haselgrove 1993), Roman (Reece 1987), and Early Medieval assemblages (Blackburn 2003; Naylor 2004; 2007).
The method used here closely follows Naylor (2007), with the date groups used there employed here with the exception of the period 840-870. In Naylor (2007) this is divided 840-55 and 855-70 as this marks a large change in coin circulation peculiar to Northumbria. Here the period is kept as a single group. In addition, in Naylor (2007) the analyses end at c. 867 with the take-over of York by the Vikings. Here, analysis extends to the Norman Conquest. New date groups have been produced in the same way as previously by looking at a best fit for the coinages issued in the period. Nine further date groups were developed: 870-99; 899-920; 920-40; 940-59; 959-78; 978-97; 997-1020; 1020-1040; 1040-66. Blackburn's (2003, 21) method of placing a third of coins in the succeeding group to reflect coin loss dates rather than striking dates (except where known re-coinages were undertaken), is not used here. It assumes that all coinage behaved in the same way over 500 years and is problematic for earlier coinages (sceattas especially) for which our chronologies are quite broad anyway (with the addition that the data we have for them come from hoard evidence and archaeology – surely these would indicate dates of use/loss not just dates of issue). Once again, 'fingerprints' have been produced for each region as well as the entire national assemblage.
Figure 65 illustrates the general pattern of coinage as described by Blackburn (2003, 31-35). The early 8th-century peak represents the sceattas, probably produced with an in-flow of silver from Germany. This is followed by a general contraction in the currency by the mid-8th century, apart from in northern England where the regal series Y was produced from c. 740-90 (Naylor 2007). During the mid-8th century small numbers of coins were lost until Offa's broad flan pennies were introduced c. 765 and start to feature in the chart (Chick 1997). The mid-9th-century peak is simply due to the Northumbrian stycas (Naylor 2004, 47), which had relatively little circulation outside of the kingdom, mostly only occurring in any number in North Lincolnshire, plus a scatter down the east coast. Blackburn (2003, 35) notes the lack of silver in mid-9th-century Europe and the rarity of finds in southern England, the Low Countries or Germany. This was probably due to the Viking threat and payments of tribute in silver. From the late 10th century and into the 11th century the chart shows a steady increase in coin loss. However, it is worth noting that while most Middle Saxon sites have been subject to metal-detecting, by the 10th and 11th centuries many markets and other settlements were located where there is ongoing occupation. As a result, most finds may only be found by excavation or chance. Therefore, we should consider that actual coin loss in the 10th/11th centuries was probably higher than represented here (Blackburn 2003, 33-4).
In conclusion, like the artefact 'fingerprint', the coinage 'fingerprint' is a useful device for examining the way coinage, in general, circulated in Anglo-Saxon England, and illustrates what we might expect to find on an average site. Deviations, where they exist, need to be explained by recourse to levels of activity, occupation and communication.
© Internet Archaeology/Author(s)
URL: http://intarch.ac.uk/journal/issue25/2/3.2.2.html
Last updated: Tues Apr 21 2009