The Trench 3 ‘doughnut’ has been lifted

An unusual worked stone that we have been referring to as the ‘doughnut’ has been a familiar feature in the southern part of Trench 3 for the last few seasons at Bamburgh Castle. It was first revealed in 2011, but it took two further seasons to lower the ground surface around it sufficiently for its unusual form to become apparent.


The metalworking building roughly outlined by the stones used for its foundations with the top of the ‘doughnut’ just visible 1m outside the west wall, in the lower- left of the photo.

It lay immediately west of the metalworking building and is broadly contemporary with that structure, which would make it middle 9th century AD in date. The wear pattern on the side suggests that the top third of the stone was uncovered above the surface for sufficient time for it to wear and erode more than the base which appears to have been set into the ground. We believe therefore that it was inserted into a cut which we have struggled to see.  The presence of a small number of what are likely packing stones reinforces this interpretation.


As removed sitting right side up. The unusual form and perforation is visible (scale 0.5m demarked with 10cm units).

The stone is difficult to interpret and has been suggested as a socket for a timber pivot, a socket for a standing cross, which would be exciting but sadly is rather unlikely give its circular shape. It could also to have been set as a soakaway, given that it is perforated all the way through. Now we have lifted it and examined it seems likely that it started life as a mortar as the upper part of the perforation is smooth and the lower crudely cut through. Our best guess is that it was a large mortar re-used as a pivot stone.

Stone 2

In the side view the more eroded upper third of the stone is seen to the right (scale 0.5m)


The base of the stone. It is just possible to see tha the last part of the perforations is quite crudely chiselled unlike the upper part which is finished or worn to be relatively smooth. Also just visible are a series of shallow dimples unevenly arranged around the hole (scale 0.5m).

Visit to the Heugh Excavation on Lindisfarne

We recently visited the ongoing excavation on Lindisfarne that is being undertaken by the Archaeological Practice as part of the Peregrini Lindisfarne Landscape Project. It’s a fascinating site and should be familiar to some, as it has been the subject of a number of news reports. The team have opened up a series of trenches on the Heugh, which is the long, narrow, dolerite rock promontory above and to the south of the medieval priory site in Lindisfarne Village. The Heugh has long been speculated to have been part of the early medieval monastery founded around 634 during the reign of King Oswald, as a daughter house of Iona. It quickly rose to be a site of great importance and remains famous for its association with Saint Aidan and Saint Cuthbert as well as being the place where the wonderful Lindisfarne Gospels were made.


The foundations of the potential new church on the Heugh

Here the excavation team have found the almost complete foundation of a stone building, that all current evidence suggests is early medieval in date. Direct dating evidence is scant, but the near complete absence of later medieval and modern pottery from the structure, despite a considerable volume of material being excavated, suggests a time of construction when pottery was not in use. This, together with the absence of mortar bonding and the rather crude-tooled finish to the stones, adds up to a quite compelling argument that they have discovered a building from the early monastery. In addition, the ground plan, with what appear to be a chancel and nave, is very suggestive of a church which greatly adds to the excitement.


The finish of the stones is mostly quite crude and no mortar bonds them together. It was a substantial structure though with wide foundations made of large blocks.

We know from textual evidence, particularly the writings of Bede, that Aidan’s successor Finan built a timber church that was later covered in lead. We also know from a later text that this church was removed to Norham as a relic when the monastery was partly abandoned in the 9th century. We can be certain then that this structure is not that church, but the site would have likely held several churches during its lifetime. The crude working of the stones, particularly of some sculpted stones that appear to form a trough or bowl, and part of a possible window, are very interesting as they may suggest builders that are beginning to come to terms with a new construction medium- stone instead of wood. As a consequence it is tempting to imagine this building as particularly early, but it is perhaps also possible that it could be later. In the Viking age many monasteries were abandoned, but the continued use of stone in the construction of monuments at Lindisfarne suggests that the site remained important, though the sculpted fragments of the 10th century and later often appear to be cruder and derivative. It is therefore possible that this structure could date from this later time, when the working of stone was not done with the same confidence or competence as the 8th and 9th centuries.


Bamburgh Castle, Lindisfarne’s near neighbour, just 8.5km over the water!

It is tempting also to see the location on the height of the Heugh as meaning that the building was meant to be seen from a distance. It has a clear sight line to Bamburgh, the great secular palace site, and this may be no coincidence. We have evidence of pre-conquest stone architecture at Bamburgh and it is likely that the use of this medium was intended in both instances to reference Rome. In the case of a monastery, this would be the Catholic Church as the successor of Rome, and at the palace as legitimising rule through being the heirs of the Romans. This is a good lesson in why it’s important to study how a site fits into its wider world in order to properly understand it.

Interpreting the Lower pavement in Trench 3

Blog image

A Trench 3 phase plan for the middle 9th century AD with some earlier features depicted in the northern Hope-Taylor area, including the ‘lower pavement’. The pit and socket were yet to be identified when this was drawn but it lies in the west part of the baulk, close to where the baulk joins our main excavation.

During the last week we have been excavating two pits in the area of the baulk through the Hope-Taylor excavation. One of these has proved to be quite substantial and associated with a broken stone socket used as a pivot, which we have now lifted. Working out what the socket was used for is difficult as we have few other features that we can associate with it at the moment. It likely held a door post that rotated in the socket so we should be looking for traces of a building in the vicinity. Hope-Taylor excavated to both north and south of the feature and there is what appears to be a construction cut for a timber structure in his records to the north of the stone. If we can prove that these two features were broadly contemporary we might be closer to solving the problem and identifying a new building.


The socket lies on its side and with a visible crack at the centre of this photo. It lies within a pit that cuts a mortar surface to the right and stones to the left.

West of the pits we are currently working on one of the more interesting features in the Hope-Taylor excavation. It’s a linear stone structure that lies along the west extent of his trench. Hope-Taylor called it his ‘lower pavement’ and it has been a recognisable part of of his excavation since we first uncovered it. We touched on it in the last blog on Trench 3, as we were looking to see if we could join this ‘pavement’ through the baulk to a series of stones visible in section. The ‘pavement’ is important because it is a long linear feature that extends for many metres along the west side of the trench.  Importantly it represents a stratigraphic signpost linking Hope-Taylor’s excavated material back into our sequence via the section we drew of the baulk.


Lower pvement

‘Lower pavement’ extending along the west extent of Hope-Taylor’s excavation level

We have been looking at the ‘pavement’ feature itself and some characteristics seem to be apparent. Is this stone structure really a pavement/road or a structural foundation for a timber building. The fortress wall must lie within 6 to 8m of our trench edge, judging by the position of the current wall and the edge of the dolerite plateau, which certainly leaves space for a building or a road.

The lower pavement extends for 11m (assuming it is seen emerging from the baulk) but does not extend all the way to the north end of the trench or further to the south into the southern Hope-Taylor trench. This alone would seem to make the road interpretation problematic. Looking at the surviving Hope-Taylor records, and the feature in the ground, it seems that it is a discreet structure that does not quite extend to the western trench limit. This makes us think that it must surely be a structural foundation, not a road. It also contains a variety of stone from dolerite boulders to occasional dressed blocks of masonry and these must have been re-used from an earlier building. If our further investigation can define it turning right angles at each end then we will have proved it is a building. Frustratingly as the majority of the structure lies beyond our trench and under the standing castle wall we will have only a limited chance to go further and define is role.


Aerial view of Trench 3 with a rather speculative guess at to how a building might lie between the trench and the fortress wall


Great news about our Inner Ward publication

The blogs describing our excavations at Bamburgh Castle naturally tend to be concerned with the excavations in the West Ward, where the majority of our work, like Brian Hope-Taylor’s before us, has been concentrated, but now and again we have had the opportunity to do some investigations at the heart of the fortress, the Inner Ward. This is the summit of the hill where the Keep now stands alongside the re-constructed medieval buildings and where we know from the pages of Bede that the church that held the relics of St Oswald had had formerly been built. We can also imagine the great hall of the kings of Northumbria here, once the focus of the palace complex.

BC04 Chapel 12

The chapel under excavation in 2004. Two trial trenches were located to investigate features identified by geophysical survey.

The main body of work was undertaken in 2004 and 2008, but we also conducted geophysical survey in advance of that as well as a modest excavation in the centre of the ward with Time Team in 2010. We have blogged about some of this work in the past: Now we are delighted to have a substantial paper published detailing all of this work in the Archaeological Journal, which is the journal of the Royal Archaeological Society. It’s terrific to have had this opportunity, aided by a grant from the RAI as well as the HLF and the Mick Aston Fund. It has taken a lot of work, not just by the BRP authors Jo Kirton and Graeme Young, but also by the editor Howard Williams and our anonymous referees.

It is the largest publication on Bamburgh so far (you can check out the others here) and intended to be the first of many as we increasingly concentrate on publishing our work. It will be in libraries free to access next year but until then you will need to be a member or have institutional access (Archaeological Journal). Long before then we aim to publish more popular articles on the results, and we will keep you posted on progress.




All good things must come to an end…. An end of season wrap up of Trench 3!

As week eight of the 2015 excavation season comes to its end, let’s reflect back on what we achieved and discovered this season. This blog will come to you in bite-size, area-by-area portions! How handy!

Timber building

One of the first things on our to-do list this season was to continue excavating the timber building. We started exploring the beam-slots to this building back in BC13. Beam-slots, for the confused reader, are trenches dug to contain either a horizontal timber beam to form the foundation of a building, or more likely a series of upright timbers in a construction style called ‘post-in-trench’ by archaeologists.

Understanding the western beam-slot sufficiently to be confident that we had associated the correct linear cut with our building, proved to be difficult, and took a great deal of time.  By the time the 2015 season started, we were finally confident of it, an only the southern part of the eastern beam-slot and the southern beam-slot remained un-excavated. Both these beam-slots could be seen in section thanks to the presence of three of the WWI test latrine pits (which we are pretty sure, were not used for their intended purpose, but do make good archaeological test pits).

The middle section of the southern beam-slot offered a pleasant surprise; two teminals to the construction cut marked a central doorway into the building. Fascinatingly this lay directly in the line of a pebble path in the south east corner of the trench, that ran parallel to our metalworking building (we excavated this a few seasons ago). The association of the two buildings through the presence of the path does not seem too optimistic. This means we can date the new building to the middle 9th century AD, as we a great deal of dating evidence for the metalworking building. .

This discovery motivated us to take some hammerscale samples within the boundaries of the timber building, to see if it had, at some point, been used for metalworking purposes. The low finds recovery from within the building though does make us wonder if it could have had a timber floor, which would make thethe layers within it part of an aerlier phase. Something to work out next year. Either way the sampling will hopefully be informative and not be wasted.

The outline of our new building and the pebble path that assciates it with the metalworing building. Coincidently its entrance lies in the same area as the stone theshold of the overlying large timber hall of probable 10th century date.

The outline of our new building and the pebble path that assciates it with the metalworing building. Coincidently its entrance lies in the same area as the stone theshold of the overlying large timber hall of probable 10th century date (also shown).

Elusive paving feature

You may remember from earlier blogs this season that we excavated a rectangular paved feature, consisting of 3-4 large flat slabs with two vertical flat stones to the north and south. When we originally uncovered this feature in BC13 we briefly interpreted this as a drain-feature, but quickly discarded this theory, as we could trace it no further. Sadly, excavation produced no clear-cut “Eureka!” moment, and it remains enigmatic. Our best guess, based on the realtively central location within the building, is that the horizontal slabs were put into support a large post or beam that was one of the roof supports.


The north-west corner was mostly left alone until the last two weeks of excavation. When we did return there, our main idea was to dig down two large and deep contexts, not expecting to encounter any features in the process. But as always, whenever you don’t count on finding anything, you do.

Firstly, we finally took down our large ‘jaggedy rock’ context, which took up most of the north side of the NW-corner of the trench. As expected, most of this context went down onto bedrock. However, we also found a small, circular feature consisting of small flat stones, with a diameter of about 45cm. We are choosing to excavate this feature next season, but our current interpretation is that it served as a way to level off the ground with the bedrock to the west.

Also, to the south of the NW-area, we dug down onto a large rectangular stone that looked suspiciously like it was part of a structure. Further excavation showed a line of large stones going east and then making a turn to the south of the trench. This shows that we may be dealing with the corner of a building, that might be largely extending westward outside of the trench.

Image 3 - NW

The stone linear in the NW-corner of Trench 3.

Brian Hope Taylor

This season has also dealt with the partial uncovering of Brian Hope-Taylor’s tarped area. Brian Hope-Taylor excavated at Bamburgh Castle in the early 60’s and early 70’s, and since we are almost down to the same level as the northern part of his trench, we are close to joining up the two excavation areas! BHT left us with some exciting features to excavate next season, including a drain feature and a hearth area.

Eastern side of the trench

In the eastern side of the trench, next to the eastern beam-slot to our timber building, we uncovered a pebbly surface’. As we continued excavation, this narrow strip of pebbles expanded into what we now think is either a pebble path or pebbled surface which slopes down as it extends more eastward. We are still working on finding its southern extent and finding out if it extends west of the eastern beam-slot (which would make it earlier in date than the timber building).

Image 2 - pebble path

Pebble path in the eastern side of the trench.

Lastly, to the north of our pebble context we have located what seems to be a high-medieval pit-feature. We found several shards of pottery from this period, together with two styca’s, finds that shouldn’t be in the same context under normal circumstances. This means that a pit was dug down from above (probably in the 11th-12th centuries), through 9th century layers (explaining the styca’s), which was then filled up with both the Anglo Saxon and high-medieval material. We will finish this pit next season.

So there you have it! Right now we are in the process of finishing off excavation on all the features we are currently working on; namely the pebble path/surface in the east and the potential new structure in the north-west. We will use the last two days of the excavation season to plan and level the most recently excavated areas, finish off all the necessary paperwork, and then, finally, tarp up the trench to protect it from weathering.

A confused person’s guide to Trench 1

We thought it might be helpful, for regular users of the blog, to put up annotated photographs of our two trenches, as I am sure at times it is difficult to imagine just where the individual buildings and features lie. In this blog we will start with Trench 1.

Trench 1, labelled to identify the key features, facing north

Trench 1, labelled to identify the key features, facing north

As regular readers will be aware, Trench 1 lies at the northern tip of the fortress, at the lowest point of the bedrock plateau. Here we have unearthed evidence of the early phases of defensive structures built in timber together with a rather substantial timber hall. On the photograph you can see this as a shaded outline with the outline of a later stone hall superimposed on top of it. It is perhaps only when you outline it so clearly that its full scale becomes apparent. As we have described before it completely dominates the gate cleft (in the bedrock) to its south-west, which is the earliest known entrance to the fortress.

The stone building has been assumed to be the later of the two, but it is only this season, whilst investigating the area where the two structures come close to each other in the north-east corner, that we have proved that this is indeed the case. The date of the stone structure’s construction is uncertain, but it appears to have been robbed out before the Norman Conquest.

We are on less certain ground on the western side of the trench where we have a massive laid stone boulder foundation, for what we believe to be a timber wall, that we are interpreting as part of an early phase of defences. This is based on its general alignment with the break of slope of the bedrock, and the presence of a large timber post-setting that could have carried an archway across the gate cleft itself.

The later medieval defences are much better understood, as we have written records surviving from the 12th century to help in our interpretation. The later medieval gate is built in two phases, the first dating from the later 12th century with a 13th century widening, presumably to carry a breteche (an extension like a balcony built over a gate, with openings in its base to shoot projectiles or drop objects through). The gate widening and breteche are likely to be contemporary with the glacis built in front of the gate that we have no direct dating evidence for.

The tall stack of surviving medieval curtain wall that survives on the northern wall line would once have extended all the way around the seaward side of the West Ward, just as the 20th century wall does today.

Hammerscale Sampling and Thin Sectioning in Trench 3

In the past two weeks, apart from the usual excavation and recording process, Trench 3 did a number of hammerscale samples and two thin-sections, which is quite exciting! Today’s blog is all about elaborating on these processes, so that we may get more insight into different elements of archaeology.

Hammerscale Sampling

Hammerscale sampling is used to find out if a certain area of the trench was used for metalworking purposes. Last week we carried out a small number of hammerscale samples within the borders of our large timber building to see it it has been used for any metalworking.

Hammerscale is created when a blacksmith hammers iron; the sparks that fly off are actually tiny pieces of iron oxide that are only a couple of millimeters in size.

When we take hammerscale samples, we take a planning frame of 1x1m, which is divided into 25 smaller squares of 20x20cm. We then take small samples from each of these little squares. The hammerscale, hard to see with the naked eye, is magnetic, so when we run a magnet through each individual sample, any present hammerscale material will stick to the magnet, which is exactly what we want.

Hammerscale sampling

Hammerscale sampling

When labeling the bags we need to be mindful of the coordinates of each individual sample. We gather this information so that when the samples are processed we can relate them to their position in the trench to see how the concentrations of hammerscale fluctuate throughout the area. In some cases plotting out the hammerscale concentrations can result in a so-called ‘blacksmith’s shadow’; a negative shadow where the concentration of hammerscale is low because that is where the blacksmith was working from, sweeping the hammerscale off the anvil, resulting in a high-concentrated cone-shaped ‘shadow’ forming around the blacksmith.

In our case, we dare not hope to find anything halfway as exciting as a blacksmith’s shadow (though we can always dream!), but finding evidence of hammerscale in our samples would give us some well-needed clues as to what the function of our timber-building might have been.

Thin sections

At the end of week 5, Graeme Young led Trench 3’s staff and students alike in putting in two thin-sections in one of our World War I test latrine pits.

Trench 3 has three WWI practice latrine pits. While it is quite sad that the archaeology in these parts of the trench has been destroyed, it does give us a nice little sneak peek of what we will be excavating in future seasons.

While we as archaeologists can quite clearly see differences in contexts and different occupation layers, there is only so much we can see with the naked eye. While a context can seem fairly homogeneous to us, it may potentially consist of many single events that we are unaware of.

This is where the thin-section comes in. We take a simple metal box, and after preparing the section by cutting it completely vertical, hammer the box into the section.

Preparing the section for the thin-section sample, showing off some beautiful stratigraphy

Preparing the section for the thin-section sample, showing off some beautiful stratigraphy

Removing the thin-section box from the section.

Removing the thin-section box from the section.

After getting the X, Y, and Z coordinates using the Total Station, we then carefully spade it out and wrap the metal box, now containing a nice sample of the vertical stratigraphy, into several layers of cling film, securing the sample into place and making sure it doesn’t get contaminated.

Two succesful thin-section samples

Two successful thin-section samples

The sample is then sent off to the laboratory, where they set the sample with resin, and cut off a thin slice of the sample, which they then study under a microscope. An ash layer which up until then might have been interpreted as a single event may turn out to be dozens of smaller single events!

In Trench 3 we took two samples, slightly overlapping, so that we end up with around 20cm of continuous stratigraphy. The section we have sampled is from the industrial occupation of the trench, potentially spanning a time frame of around a hundred years.

If all goes well, both of these excavation techniques should provide us with some exciting new insights into the function of our timber building and the yet-to-be excavated archaeology in our trench!

Castle Trench One Update

For our latest trench 1 update we hear from Assistant Supervisor Sam.

The highlight of the week, at least for me, came when I found a small shell-button. Although it appears relatively late, made in the last 200 years, it was counted as the first small find of the Trench and was the talk of the week, at least for my part.

Screen Shot 2015-07-10 at 1.44.57 PM

The past couple of weeks have been brilliant in terms of weather and sunshine, cheering for the staff as well as the students. Though we were happy to work under such conditions, it is very important, at least for the trench, to get a little rain to add some moisture, from time to time. This is because when its damp the soil reveals more vivid colours and contexts are easier to see. It often even reveals new features.

Our main aim for the past few weeks was to uncover as many post-holes or post-pits as we could in order to make sense of the relationships between the floor surfaces we have and possible buildings. In such matters it is worth noting that although we, in Trench1, may not find as many small finds in comparison to Trench 3, we have many interesting building features. In particular our timber and stone buildings, cut again and again by the considerable number of post-pits and post-holes, relate quite a different story about the architecture of the area.


Post hole in trench 1

As a last note on this blog, we are excited to announce that we found the base of a what appears to be a Bamburgh Ware pot. For those of you who don’t know, Bamburgh Ware is the locally produced ceramic pots which were in existence around the 11th to 12th centuries. This was found within a pit feature we had previously thought we had completely excavated. Remember how I told about how a little rain can really help us see the features!


Trench 3 Week 3: Rain foiled all our plans

While we had many exciting plans for trench 3 this week, we were only able to execute a few, due to rainy weather conditions. Two of the tasks that were completed were the removal of a large vertical stone from the ‘porch’ feature in the central part of the trench and the uncovering of the northern-most portion of Brian Hope Taylor’s excavations from 1970.

As we mentioned last week we suspected that the entranceway to the timber building lay along the southern beam-slot as there was a portion that contained uninterrupted stratigraphy where the beam-slot should have cut through. This interpretation was strengthened when we removed the vertical stone that rested where we suspected our ‘entranceway’ to be, revealing more intact stratigraphy. We proceeded to excavate the western portion of the southern beam-slot in order to find the extent of the ‘entranceway’.

Picture 1

Showing the outline of the timber building

Picture 2

Entranceway to timber building. See below for closeups of 1,2, and 3.

Picture 3

(1) Showing stratigraphy in western cut of southern beam-slot

Picture 4

(2) Showing stratigraphy cut of vertical stone

Picture 5

(3) Showing stratigraphy in eastern cut of southern beam-slot

The other major development from this week was de-tarping a portion of the trench that covered Brian Hope Taylor’s original excavation back in 1970, as we are now close to the same level of stratigraphy and are getting ready to join up our own excavation with his. Years and years of the area being tarped had resulted in a buildup of blown in and trampled material, so we broke out our mattocks and spades (something that is usually only reserved for the Kaims) to tackle this plant-infested area. We reached our goal for the week, but plan to peel back the tarp even further south to unveil more of Brian Hope Taylor’s excavation extent and marry it up with our own.

Some wet day speculation, or how to read a little too much into some burning

It’s probably only fair to start this blog with a warning that in places it contains more than a little speculation! Hopefully I will make the areas where I am reaching a little, apparent in the text.

This last week, within Trench 1 at the castle, we have been removing the last fill from our middle Saxon timber hall. This structure is clearly traced, cut through boulder clay subsoil, and also in places, through the bedrock. So we have, quite literally, hard evidence for the footprint of this building. Its height is less certain, but even if it was a fairly normal single storey structure, its position immediately above the cleft in the rock that leads down to St Oswald’s Gate, would mean that it would tower above anyone entering the fortress. Our best interpretation for it function is as the gate wardens hall. This is based on its location and impressive siting, so represents our first bit of speculation. To its immediate west and very close to the edge of the bedrock, where it falls away to the external ground surface (outside the castle), a heavily constructed rubble foundation extends, parallel to the bedrock edge. We identified this several years ago and have described it ever since as the foundation for the inner wall of a box rampart. Part of a timber phase of the fortress’ defences. We believe this to be early medieval, though can only date it to before the 12th century AD, with any certainty.

So far we have looked at the archaeological evidence and have made some quite reasonable extrapolations from the structures and material that we have unearthed. During the last few days we have identified some patches of discoloured subsoil that are almost certainly the result of some pretty intense burning. Intense enough to penetrate to the subsoil and chemically alter it. This burning lies in the narrow gap between the foundation for the wall of the building the gate cleft. One possible explanation for this would be that the workmen, who cut the bedrock for the building foundations, used a technique of heating and rapid cooling with water, to fracture the bedrock. This does not seem to be likely though as we have identified a number of examples of foundations cut through the bedrock and do not see this elsewhere.

The corner of our construction slot for the timber building as it approaches the rock cleft at St Oswald's Gate

The corner of our construction slot for the timber building, as it approaches the rock cleft at St Oswald’s Gate

The discouloured subsoil is easier to see with the naked eye, but perhaps is just discernable.

The discoloured subsoil is easier to see with the naked eye, but perhaps is just discernible.

This leaves us with another exciting possibility. The burning is very close to the gate, and our inner line of timber defensive wall, which raises the intriguing possibility that this represents an attempt, by an enemy force, to burn their way into fortress through its most vulnerable point. In fact we have a record, in the pages of Bede, to just such an event at Bamburgh in the later 7th century, when Penda, King of Mercia, made a great heap of timber against Bamburgh’s timber wall and set it on fire. In Bede’s story he goes on to relate how the prayers of St Aidan caused the wind to change direction and blow the fire and smoke back in the direction of the attackers, foiling their plans. Could we have evidence of this very attack, burned into the subsoil at the fortress’ weakest and most vulnerable point? Its certainly not impossible, but our speculation metre may now be close to off the scale, so we should perhaps leave it there. After all there are lots of reason why things catch fire and burn.