Dodgy digs

 

In 1929 an archaeologist, Leslie Armstrong, oversaw an investigation of the remains of Sheffield castle. The results were published in a paper the following year. Armstrong concluded there were three layers of habitation: an Anglo-Saxon hall of 'aula', probably belonging to Earl Waltheof (executed 1076); a motte and bailey built on top of the earlier site by William de Lovetot; a stone castle built by the Furnivals c.1270.

In a section of his report, titled BURNT LAYER: REMAINS OF DE LOVETOT CASTLE, Armstrong claimed to have discovered evidence of the burning of Sheffield castle in 1266. In that year Sir John de Eyvill, a Montfortian knight, 'cum equiis et armis', destroyed 'Saffield' during the Barons' War. Armstong highlighted a 'destruction layer' 4-8 inches thick, comprising charcoal and wood ash combined with calcined rubble and fragments of dressed masonry displaying damage by fire, found 'at various points' on the site. Beneath the courtyard buildings he recorded a layer, 9 to 12 inches thick, of ashes and burnt stone. In a further report he described these layers vividly as 'crackled and burnt to a deep red tint by the action of fire'.

A licence to crenellate

The tale grew in the telling. However, a more recent survey in 2008 shot Armstrong's paper full of holes. First, the burnt layer lay below the remains of what he had assumed was the castle built by Thomas de Furnival. Armstrong had only a text-based assumption that it dated to the destruction of 1266. Nor did he have any independent means of dating the layer to the mid-13th century, never mind to a precise year. Outside of brief chronicle accounts, the actual damage committed by John de Eyvill is unknowable.

Certainly, in 1270 Furnival received a licence to 'build a stone castle and fortify and crenellate it' at Sheffield. However the extent to which the earlier castle was damaged is unknown, and the licence to crenellate in 1270 proves nothing: many of these licences were acquired long after the actual work had been completed, or even prior to work that was never undertaken.

Modern analysis failed to recover any dating evidence from the burnt layer. However, the Pipe Roll accounts for 1188 record a large payment of £66 spent on rebuilding Sheffield castle after a fire. Thus the 'Ashes 1266' layer that Armstrong identified in all likelihood refers to damage almost eighty years earlier.