Digging into databases

Arguably, one of the biggest challenges that continually faces archaeology is accessibility. While that can be interpreted in several different ways, what I’m going to concentrate on in this month’s ‘Science Notes’ is access to data. With the advent of Open Access, as well as more excavations making it to publication, we are beginning to make headway in this area, but there are still many hurdles to overcome. One of the ways in which accessibility could continue to improve is through the creation of more online databases, connecting data currently held in museums and institutions across the country. The need for this has become even more apparent recently with the COVID-19 crisis restricting physical access to these sites. When data can be easily retrieved, new, large-scale, and more complete research can be done.

An example of some of the innovative research that has already been done with large datasets is highlighted in a recent paper by Emma Brownlee, University of Cambridge. Recently published in Antiquity (https://doi.org/10.15184/aqy.2020.153), it assesses how use of grave goods in Europe declined between the 6th and 8th centuries AD, and posits how this behaviour spread across the continent. Previous research in this area has predominately focused on local factors and has yet to look closely at the phenomenon as a whole. This is not unexpected, as to do so requires the analysis of large amounts of data – which is exactly what Emma did. For her research, she compiled data from more than 33,000 graves found in 237 cemeteries across modern-day England, France, Belgium, the Netherlands, Germany, and Switzerland.

ABOVE A relative kernel density ‘heat’ map showing the changing concentrations of grave-good use across Western Europe between AD 680 and 800.
A relative kernel density ‘heat’ map showing the changing concentrations of grave-good use across Western Europe between AD 680 and 800. IMAGE: Emma Brownlee, Antiquity Publications Ltd.

Using kernel density estimates (KDE; see CA 353), Emma was able to visualise the changing density of grave-good distributions over both time and space. To determine whether any difference seen from the KDE results was statistically significant, she then used the Getis-Ord Gi* statistic, which compares one data point with those directly around it, and then compares those points as a group to the sample as a whole.

The results indicated that, at the beginning of the 6th century, England (particularly East Anglia) and Alamannia (in modern Germany) had the highest concentration of grave goods, while West Frankia and Burgundy had the lowest. This quickly changed, however. In England, the decline of furnished burials appears to have happened in two parts: first, a decrease between AD 550 and AD 600, followed by a period of steady use until AD 650, when the practice began gradually to decline again, until it was almost completely abandoned between AD 680 and AD 690.

On the Continent, where cemeteries were longer-lived, the statistical resolution is less precise. The results suggest that use of grave goods began to decline c.AD 680, when the most richly furnished cemeteries fell out of use. Analysis of individual well-dated cemeteries from these regions, however, indicates that the decline in grave-good use also began in the mid-6th century, but is simply obscured at the larger scale. This process accelerated into the 8th century, and by AD 750 only the Lower Rhine and parts of the Low Countries and Frisia still had cemeteries with lots of grave goods (which could be due to the cemeteries in these areas having longer life-spans). By the end of the 8th century, however, the practice had almost completely ceased across all of Europe.

These results appear to fit nicely into a diffusion model. As Emma notes: ‘The patterns of change… match that of a typical diffusion curve. Early uptake of a new behaviour is usually slow, as only a few innovators experiment with the new ideas. Once the behaviour is taken up by 10-25 per cent of the population, however, its adoption by the remainder becomes increasingly likely, as the system self-generates the pressure to adopt. Finally, the rate slows due to the resistance of the late adopters, some of whom may never embrace the new behaviour.’

For this practice to have spread so quickly across Europe, there must have been high levels of connectivity across the region. Bulk trading of goods, not just high-status objects, begins to become apparent in the archaeological record starting in the 6th century AD, with foreign items even finding their way into more rural cemeteries. These trading links continued to intensify with the development of sails in the 7th and 8th centuries, allowing for smooth travel across the North Sea. As the paper concludes: ‘This growing interconnectivity led to a “global Europe”, which shared not only funerary rites but also many other aspects of culture.’