In 2018, as part of our ongoing work for the Bath Abbey Footprint project, Wessex Archaeology excavated a sequence of the remarkably well-preserved archaeological remains in the vaults and cellars beneath Kingston Buildings and Abbey Chambers. During this phase of works, the team identified the remains of two well-built stone apsidal (semi-circular) structures sealed below later medieval archaeology, but overlying Romano-British deposits. Both structures had plaster on their internal walls and appeared to be the east ends of separate buildings.


Saxon Structures?

As we tentatively suggested in a blog post last year, detailing the historical and archaeological evidence for the Anglo-Saxon monastery at Bath, these structures could potentially be of Saxon date.

Apses are not uncommon within Roman architecture, but for later periods they are almost exclusively associated with chapels and churches. We knew these buildings occupied a space which would have been open-ground from the 12th century onwards, but we could not rule out the possibility that they could be of late Roman or early Norman origin.

A further note of caution was struck as to date, no Saxon stone buildings had been positively identified in Bath. This is partially due to the nature of the archaeology – although remains of this period are deeply buried, they are often more ephemeral than Roman and medieval remains, and are more vulnerable to disturbance by later activity. Saxon and medieval remains were also largely overlooked by 18th- and 19th-century antiquarians in their haste to expose the more dramatic remains of the Roman baths and temple complex.

Despite the suggestive stratigraphic evidence, we knew we would need to use scientific dating techniques if we were to confirm the date of these structures. Fortunately, within the internal plaster rendering of the southern structure there were fragments of charcoal. With the agreement of the Abbey, we sent two of these charcoal fragments to the 14CHRONO lab at Queen’s University, Belfast, to see if it was possible to obtain radiocarbon dates.

A late Saxon apse found during the Bath Abbey Footprint Project
Radiocarbon Dating

The lab successfully obtained dates from both samples. The dates were AD 780-970 and AD 670-770.

The charcoal was derived from oak. As Inés López-Dóriga, Environmental Archaeologist at Wessex Archaeology, explains, oak can be a problematic species to obtain dates from;

“Radiocarbon dating on wood charcoal provides a date for the growth of the rings of a tree and not for the time of burning of the wood. Oak is a long-lived tree species and the difference in radiocarbon age between a ring in the centre and a ring in the periphery of the tree trunk can be considerable (some hundreds of years in the case of veteran trees): this is because the radiocarbon age of a ring in the centre dates to when that particular ring formed, whilst the radiocarbon age of the last ring dates to when the tree was felled or died. In addition to this problem, wood may be used as fuel long after a tree was felled or died (for example, an old timber used as a beam in a building can be re-used as fuel after repairs or demolition works are undertaken) and even then, there may be a gap between the time of burning and the time of the charcoal becoming incorporated into plaster.

The radiocarbon results have a small uncertainty associated, but can give a certain date after which (terminus post-quem) the plaster must have been applied: sometime between AD 780 and AD 970.”

An undated but probably late Saxon Apse
Bath’s lost Saxon monastery

The positive dating of these structures as Anglo-Saxon is extremely significant as Cai Mason, Senior Project Officer, explains;

"When you find something unusual, you have to think 'what is the most mundane explanation for what we've found?', and most of the time that will be the explanation, but sometimes that doesn't work, which makes you wonder 'have we found something genuinely unusual?’

 Anglo-Saxon stone buildings are rare, and to date completely unknown in Bath, so when we found the two apsidal structures we were, despite the circumstantial evidence, reluctant to say they dated from this period.

 In a post-Roman context, the most likely place to find this type of structure is at the east end of an ecclesiastical building, such as a church or chapel, and given the fact that the excavated structures are surrounded by late Saxon burials, this is the most likely explanation for their use. This, together with the late Saxon stonework and burials found at the Abbey, provides increasingly strong evidence that we have indeed found part of Bath's lost Anglo-Saxon monastery." 

Excavating a Late Saxon apse

Given that the potential date of these structures spans some 200 years, there are several possible contexts for their construction. King Offa of Mercia acquired the monastery in AD 781 and is credited by William of Malmesbury for building the famous Church of St Peter, probably utilising the ready supply of worked stone from the nearby collapsing Roman baths complex. Extensive building work within this period is further attested to by Offa’s successor Ecgfrith having the infrastructure in place to hold court at ‘the monastery in Bath’ in AD 796. This phase of energetic building activity does fit neatly with our earliest possible date for the plasterwork, but it is certainly not our only candidate.

King Alfred of Wessex annexed Bath from Mercia following his defeat of the Danes in AD 878. We know he improved the town’s defences. Is it possible that a man of such deeply held religious conviction sought to improve the monastery as well? Certainly, his later successors Æthelstan and Edmund both granted land to the monastery which could have generated the wealth needed for further construction work, and it seems very likely that the monastery would have seen further improvements prior to King Edgar’s coronation there as the first King of England in AD 973.

It is also likely that the two buildings represent two separate phases of construction at the monastery. We may not be able to refine the dating for either structure much beyond late 8th- to late 10th-century, but what is certain is that they constitute an incredibly rare and important discovery.

By Bruce Eaton, Project Manager