Investigations beyond St Oswald’s Gate: the end of season overview

St Oswald’s Gate is written about as the entrance to the fortress of Bamburgh in the 8th century AD and very likely the route that you made your way up onto the rock plateau since the earliest settlements were constructed in the late Bronze Age. Our current work is investigating this entrance, the access routes up to the entrance and the structures and outworks built to control this access over many generations. If you want to catch up on some of the reposts on the earlier season of work, in September 2021, you can find it here.

In our second season of work on the area of the cottage and tower we continued to remove rubble from within the cottage structure as well as to further investigate the wider outworks.

The outworks beyond St Oswald’s Gate showing the main areas of investigation

We knew from both map and plan evidence, and our initial work in removing rubble from the ruins of the cottage, that they extended below the present ground level. We were less certain if this was a case of the ground level being raised considerably by the accumulation of wind blown sand or whether a basement level had been dug into the ground at the time of construction. It may well turn out to be the case that both are somewhat true. The full depth that it may reach, below the current ground level, started to become apparent when we found the top of a stone archway, having already excavated a considerable depth of fill from within the structure. This arch would have led into the small room that we believe contained the well (it is depicted there on a plan of c. 1803).

Our only depiction of the well on an early plan dates from AD 1803

In a previous report we discussed the entranceway to the cottage and how it was realigned so this time we concentrated on the room that we think contained the well and was almost certainly part of the tower mentioned in the medieval records. This small room almost certainly started out as the medieval tower of Elmund’s Well and was later incorporated into the cottage. At the end of the last season of work we had revealed the top of the third flight of steps down into the building. The first two had been comprised of two short flights of stairs each at right angles to the one before. The third flight turned again through a right-angle to align with the first and extended down from a small landing.

Investigation of the tower and well room

One of the first exciting discoveries of the new season of work was the presence of two splayed lights (open window-like features) in two of the walls. Only the base of the features survived, but their form, rather like arrow loops, strongly suggested that they were medieval features and were our best indication yet that this part of the structure was indeed the part of the building that originated as a medieval tower.

The drone photo shows the unusual shape of the tower as it emerges from the rubble. The two narrow loop windows discovered this season are marked by the black lines

Removal of the rubble was a slow process but eventually this was greatly advanced by two of the castle staff and a small machine excavator being on site for a day. As more and more of the structure emerged it became increasingly obvious that multiple phases were present, showing many different building and alteration events in the history of the structure. The archway had been very well constructed and was a very solid structure standing to a considerable height. The steps that led down to it were built from cruder stone and some had clearly been reused. They were also rather worn suggesting they were of some age as well.

The well room, as perhaps we can now call it, continues to be an odd feature. When originally uncovered it seemed broadly rectangular, with internal dimensions close to 4m north to south and nearly 2m wide east to west. There was, right at the start, the impression that there were few if any right angles at the corners where walls joined and this has been confirmed as we have excavated deeper and exposed more of the walls. The east wall, that was the continuation of the east wall of the outworks was seen to curve inwards at the north end of the room. On the other side, where the arch has now been found, the stone wall that contains the arch also curved in to narrow the room at this north end. This odd shape then appears to be a deliberate feature of the build. Fanciful thoughts that the structure might turn into a D-shaped tower were ended when more of the north wall was exposed, showing it to be without any curve. Variations in the build of the room suggest a structure much altered over the centuries and with many phases to unpick. The lower courses of the east wall (uncovered so far) are butt jointed to the south wall and so not the same building event. A missing stone in the south wall that marks a gap that seems to penetrate right through the wall also suggests this is a deliberate feature. Clearly a lot of work remains next season to properly interpret the history of the building

The outworks are also producing unexpected results

The outworks were also further explored. Last season we looked at a small section of narrow wall, but this proved to be frustratingly difficult to understand so this year we have expanded the areas we excavated to gather a better and wider picture. The new area of excavation was next to the path that extends through the postern towards the village, a narrow gap between what was thought to be the oldest stretch of standing wall and the broad wall containing the postern. This narrow wall contained an archway that is tall and quite narrow that has long intrigued us as possibly quite early in date. Some of the stonework low in this wall certainly seems to be a good match for 11th to 12th century work. This little triangle had been looked at many years before, but this time was dug and explored to as great a depth as we were able. This revealed the foundations of the early, thinner stone wall and showed that the thinner wall ended abruptly, likely cut to allow for the later broad wall to pass across it. This seemed to confirm that the broad wall was, as expected, the later of the two.

The outworks immediatly outside St Oswald’s Gate. The diference in thickness between the thin wall with it arch and the broad, later wall with the postern gate is very obvious from this drone photo. In the bottom right hand coner of the photo still covered in ivy you might just be able to make out where the two builds of the broad wall overlap.

This broad wall was also investigated at its east end where it approached the bedrock on which the castle stands. It was expected that it would simply extend to where the rock rises up, but like so much at Bamburgh turned out not to be so simple. It stopped well short of the bedrock and in what appears to be a deliberate gap, that was later in-filled with rubble and earth in a crude wall-like blocking. The wall was well above ground level at this point so it seems an unlikely spot for a second entrance so this feature remains a bit of an enigma to be further investigated.

This same broad wall was further investigated to the west of the cross-wall with the archway and found to be constructed of at least two walls that joined in an unusual overlap. Again confirming the complexity of the outworks and their numerous build phases. The foundations showed it to be a very substantial structure and a discontinuity in its exposed internal face associated with a rubble spread within hinted at the presence of a feature now lost.

Can we fit some of our new discoveries into the historical records?

As the medieval tower that formed the core of the structure had been named after a well it is possible that it had been constructed deliberately to control access to this important resource as well as to dominate the small harbour present beyond the outworks to the west. The tower was already old enough to be in need of repair in 1249-50, which suggests it had stood for some time. We think it is not unreasonable to put its construction back to the 12th century. This entry in the records (The History of the Kings Works, 1976) also describes works and repairs to the adjoining barbican before St Oswald’s Gate. It is possible that the broad wall with its postern gate leading out towards the village could be the result of this building work. Its form and style are at least consistent with work of this age.

If the stone tower did originate in the 12th century then we really would have expected it to be square or rectangular and not the odd non-parallel sided structure that is currently emerging as we empty the rubble from within it. At least its relatively small size is consistent with such an early date and is close to that of the two 12th century towers on the north wall of the Inner Ward and single, probably 12th century, tower on the south wall of the Inner Ward. All around 5m externally on their longest sides and with narrow loops for windows within the wall. There is also the issue of the name of the Well. Elmund may be a variation of Ealhmund an Old English name, which suggests that the well predates the conquest. It may be asking a lot of the investigation to find evidence of structures predating even the 12th century but we can hope.

There is a good deal of work still to be done it seems as the outworks have proved to be a complex subject whose building sequence is only just beginning to be understood. We are also excited to find what traces remain of the well itself and that is before we consider what might be found within it.

Join us this September!

If you are intrigued by any of these discoveries and would like to learn more in a hands on experience, we are running a post-excavation taster week open to all who want to learn a bit more about how archaeology uncovers evidence of the past. Info can be found here.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s