The Dunster Castle Mosaics

Dunster Castle in west Somerset, is one of three Wessex Norman motte and bailey castles now owned by the National Trust. Their 11th century designers all used natural hills. Each was a strategic location but history changed them.. only Dunster has remained a residence through 1000 years.. a grand mansion house, impressive in scale and outline, high above the road into Exmoor.

1754 painting2

1754 painting of Dunster’s dramatic setting on display in the Castle

In south Somerset, Montacute Castle, on St Michael’s Hill , is now only visible as earthworks under trees. It ended its military life in the 12th century when the land was given to Montacute Priory.

Corfe Castle thrived as a royal castle, particularly in the 13th century, but had become old fashioned by Tudor times. Elizabeth I sold Corfe and it became a rich family’s trophy house.. They backed the King (the losing side) and so in 1646 it was made uninhabitable. Now it’s a craggy ruin.

Dunster is different.. It survived the turbulent years of the English Civil War. It progressed.. and was developed during the 18th and 19th centuries.. complete with stables, outbuildings designed parkland, gardens and summerhouses.

And so it was… that last August I took the long and winding road from Taunton to Minehead in search of a Dunster mosaic.

Don’t get me wrong… these are pebble mosaics not Roman ones .. but they are intricate designs, hidden and poorly understood.

The thing about Dunster Tor is that it’s got unstable slopes. The paths and access road, spiraling up the steep hill to the Castle’s front door, keep slipping away.

I arrived at the right time, morning tea-break in the bothy, and then Robin the Head Gardener guided me up the hill with drawing board, camera, notebook and measuring tapes.

IMG_4898

Starting to clear the overgown path below the Castle. 

A busy summer day, many visitors enjoying the sunshine but I was shown down a lost path. Closed because of health and safety. It doesn’t go anywhere now. After about 30m, it stops abruptly at a steep slope, where the old route has tumbled down the hill.

Robin found the spot and pulled some creeper plants which had grown across the abandoned path. There, was a pattern of pebbles set in a hard white mortar.

He wished me well and left me to it ..and that was my home for the day.. shaded by the bushes and tall plants and all around me the voices of happy holiday people walking along other paths. Nearby but out of sight.

The path had been cut into the hillside. On the uphill side, I pulled back the greenery and found the red sandstone blocks of the revetment wall. Where the path met the wall there was a heap of soil and roots. I moved the vegetation… and just above the mosaic surface were fragments of plaster and pieces of brick and slate.

IMG_4908

The pebble mosaic running under the revetment wall.

There were also two blocks of stone joined together and forming an 120 degree angle as though they once formed the corner of a polygonal building. The revetment wall had been built above this corner and the mosaic ran up to it….The archaeological sequence .. first the stone corner, then the pebble floor built against it and then, at a later date, the revetment wall for the path built above them.

IMG_4904

Now it was time to clean back from the wall and reveal the pattern of the white pebbles. It was edged with a curving fan of long, pitched, red-brown stones. Then there were zig-zag patterns of long grey stones among the white pebbles. In the centre of each zig and zag, was a rosette of long stones with a pebble in the middle. Beyond that and further downslope there were interlocking arcs of grey stones dividing up the white pebbles…but then I ran out of path.

IMG_4916

The stone rosettes 

Slabs of the mosaic had  fractured and tipped down slope and then had been covered and resurfaced in the 1970s to repair the path and make it horizontal again.

Really good mortar… it held the pebbles fast as the floor cracked and slipped away down the hill.

By the end of the day I’d uncovered about half the surviving semi-circular design. Originally, it must have been about 5m in diameter but ….how old was it and what period in the Castle’s long history did it belong to?

I’ve been writing up the report and the answers are not easy to find.. definitely 18th or 19th century but surely we can do better than that.

There are two known Dunster mosaics. The other one, on the north side of the castle, was built against the 15th century gatehouse. This floor design is a series of concentric pebble petals and was carefully uncovered and drawn in the 1990s. Robert the excavator concluded that the mortar used in the floor was a kind of ‘Roman’ cement and was therefore at least earlier 19th century in date.

The one I had revealed was on the south side of the Castle and although it had a different design, the mortar and types of stone were similar. There is no reason to doubt that they are contemporary and part of the same period of garden design.

Dunster Castle has such a dramatic scenic profile: it has been drawn, painted and mapped many times since the early 18th century.

Changes usually take place when there is money and the Luttrell family (the owner occupiers of Dunster from the 1404-1976) didn’t always have large amounts of spare money.

In the early 18th century, Dorothy Luttrell had cash to spend and used it to redesign the gardens. A drawing of Dunster in 1735 shows a white building in the area where I drew the mosaic. There is a painting dated 1754 which also shows the building. Is this the building which covered the mosaic. There’s no similar structure for the north pebble floor and the the type of mortar doesn’t work for such an early date. ‘Roman cement’ was invented by James Parker in 1798 and is unlikely to have been used at Dunster until the early 19th century.

18th century

The early 18th century painting at Dunster showing a little white building on the left side of the Castle in the area of the pebble floor.

Henry Fownes Luttrell 1747-1780 had money and lived at Dunster much of the time as did his son John 1780-1816 but the next owners lived mainly in London and the Castle went into decline.  Then, in 1867, George Luttrell inherited and took the place in hand. He commissioned fashionable architect, Anthony Salvin, to design a gothic revamp for the place.

The surviving later 19th century photos maps and plans give no hint that the mosaics were created at this time.

However, they may have been designed and seen for just a few years and any covering pavilion or summer house building may have been a light timber framed structure quickly removed.

My best bet… given the type of mortar …and the occupation history of the Luttrell family, is that the floors were commissioned by John Luttrell before 1816… can’t prove it though.

Unfortunately William Turner’s painting of 1811 shows nothing and neither does the tithe map of 1840. But they were  not created to show garden detail….

1840 Dunster

Dunster’s Tithe Map 1840

so I must hope for a future researcher who one day.. at Taunton.. at a table in the Somerset County Record Office…working through deep pile of papers in the Dunster Archive, will suddenly alight on the conclusive document ….I hope he or she spots it.

 

It’s all in the name..

Close up detail of plaster work around the top of the ceiling above the marble staircase

Once again I headed for Kingston Lacy with a mission to check under the floorboards in the house. A condition survey was being carried out by Clivedon Conservation on the plaster ceiling above the marble staircase.

Douglas and Tina (National Trust paintings conservator) surveying the painted plaster ceiling

It was while looking under the floor in the third Tented Room above the ceiling that Douglas from Clivedon Conservation spotted some writing on one of the joists of the superstructure, but he had not had time follow it up further.

“James” written in pencil on the wooden joist

So as well as looking between the joists for objects lost down the cracks between the boards or hidden on purpose, I had a look at the faces of the joists to see if I could find more writing. It was difficult to get the right lighting and angle to make out the words, especially as not all the boards had been lifted. But with the help of torches and various settings on my camera I could make out one full name, a part name and a date!

The surname “Game” to go with the first name James

The complete name was James Game, followed by the name Isaac and something illegible, presumably a surname, and then the date November 25th 1837. William John Bankes commissioned Charles Barry in 1835 to remodel Kingston Hall. This work was completed circa 1841, so the 1837 date fits with work being carried out in the house.

November 25th 1837

With access to the 1840 census I thought I would look up James Game to see if I could find him in the area or on the estate. It was exciting to find someone of this name living at Hillbutts, a small group of dwellings beside the boundary of the parkland around Kingston Lacy house. But best of all, his occupation was listed as a joiner!

I think the second name of Isaac starts with an N? All ideas and suggestions welcome, then we’ll see if we can find Isaac on the census as well!

I think the surname of Isaac starts with an N, or perhaps M

The name Isaac written in pencil

 

All I want for Christmas……

It’s always exciting when I am handed bags of finds from work done by archaeological contractors in and on our properties.

Box of delights

Box of delights

This week it was a few objects found by Ian, while doing a building survey, they were under the bedroom floor of a farmhouse in North Somerset. The main part of the house dates to the 18th century but it looks like it could go back  to the 16th or 17th centuries and was at times the home farm for a bigger estate.

I took out the bags and noticed it said wooden animal on all of them, so not the usual nails, fragments of wall paper, cigarette packets or chewed up paper from rat nests!

I took them out one by one, they were a bit nibbled but still recognizable as animals. But apart from the piggy they did not look like ordinary farm animals.

The wooden animals a pig, a Deer/Lama and a Sheep/?

The wooden animals a pig,  a Deer/Lama/?  and a long  legged Bear/?

I wondered if they could be from a set of Noah’s Ark animals, I remembered seeing one at one of our properties, so I searched our collections database and found quite a few images of very similar animals to the ones Ian found.

Wooden animal from the collection at Erddig, Wrexham

Wooden animal from the collection at Erddig, Wrexham

Wooden animals for Noah's Ark, from Felbrigg, Norfolk

Wooden animals for Noah’s Ark, from Felbrigg, Norfolk

Our animals have the remnants of paint on them so would probably have looked a little bit like the set below.

Wooden toy figures of Noah and his wife, and pairs of animals, next to the Ark, at Scotney Castle, Kent.

Wooden toy figures of Noah and his wife, and pairs of animals, next to the Ark, at Scotney Castle, Kent.

Close-up of the Pig showing evidence of paint

Close-up of the Pig showing evidence of paint

Close-up of the possible Bears head

Close-up of the possible Bears head

Probably more like this set from  Snowshill Manor

Close view of the wooden Noah's Ark with model animals made in the mid-C19th in the Black Forest area of Germany, collected by Charles Wade and displayed with other toys in Seventh Heaven, Snowshill Manor.

Close view of the wooden Noah’s Ark with model animals made in the mid-C19th in the Black Forest area of Germany, collected by Charles Wade and displayed with other toys in Seventh Heaven, Snowshill Manor.

So the rest of  the title would be ….. the rest of the Noah’s Ark animals, oh! and an Ark to put them in.

 

 

 

 

 

 

 

Quern quest

Looking east from Seatown, West Dorset

Looking east from Seatown, West Dorset

As Martin so eloquently puts it ‘the cliffs are leaking archaeology’ especially in West Dorset, with its soft geology and erosion by the sea. Luckily for us there are keen-eyed locals who walk the same routes and notice changes and strange objects laying on the beach or sticking out of a fresh landslip.

A few weeks ago I found a message on my desk to ring a Mr Bickford who had found what he was sure were parts of a quern stone used for grinding corn and some clay loom weights, near Seatown in West Dorset. I felt a little jolt of excitement, as regular readers of this blog will recognize Seatown as the place where we excavated a Bronze Age burnt mound and two Iron Age ovens. (see 20/07/2015 burnt mound the story so far). Could we have more evidence to fill out the story of the Iron Age at this site, or was this a new place to investigate further along the cliff?

The layer of burnt flint and stone can be seen in the middle of the picture

The layer of burnt flint and stone of the ‘burnt mound’ can be seen in the middle of the picture

I rang and arranged to pop over to Seatown and look at what he had found and record were they came from. So it was that I headed west on a bright and sunny morning, deep blue sky above and spirits high. I was not disappointed!

I met Humphrey in the car park and we walked up the hill to his house, round the corner and into the garden. What I saw took the last of the breath away that the climb up the hill had left me. On the garden table were three large pieces of quern, both upper and lower stones, and next to them what looked like one and a half very large triangular clay loom weights!

“Wow! Oh yes they are exactly what you thought they were”

The top and bottom stones together as used

The top and bottom stones together as used

The pieces of quern stone

The pieces of quern stone

 

 

 

 

 

 

The stone the quern is made from is not local to the immediate area. We have had a few geologists look at images and one suggestion is that it may be continental! But they need to see it in the flesh, so to speak, so they can see every mineral and inclusion.

The loom weights are very large and have more holes than necessary so may not be loom weights. If they were they would have been used on a warp weighted loom, to make cloth by keeping tension on the warp(fixed thread)

The loom weights

The loom weights? probably something else but what? 

Hopefully my hand gives a scale to the size of the weights

My hand gives a scale to the size of the weights

Both the quern and the possible loom weights are probably Iron Age and the small piece of pottery found with them looks very like the Iron Age pottery from the ovens found when excavating the ‘burnt mound’ site nearby.

A reconstruction of a warp weighted loom, the weights are along the bottom behind the lowest bar

A reconstruction of a warp weighted loom, the weights are along the bottom behind the lowest bar

A roman hand quern very similar technique to an iron age quern

A Roman hand quern, using a very similar technique to an Iron Age quern

 

 

 

 

 

 

Once again we are on the trail of more information about a site. Try to solve the mystery of the weights and it’s a trip to the geologist first to see if we can track down the origin of the quern stone, who knows what stories we can then tell about the people who lived at Seatown over two thousand years ago.

Day 11 – 21 today, 21 today…..

The last full day of digging as dawned and its all hands to the pump to get to the bottom of the bath house and gather the last ounce of information from all the trenches.

Eileen set to work in the Buckeye tree trench next to the cross passage ‘buttress’ her task was to find out what was happening next to the wall were the soil changed colour.

Eileen cleaning the last of the upper dark soil from the trench

Eileen cleaning the last of the upper dark soil from the trench

Eileen soon popped up to alert us to the answer,  a very large stone in the corner where the ‘buttress’ meets the corridor wall of the north range.

A very happy Eileen

A very happy Eileen

The top of the large stone on the right of the picture, with the 'buttress' wall to the right of the stone

The top of the large stone on the right of the picture, with the ‘buttress’ wall to the right of the stone

Carol was joined by John and Les in the mosaic trench to finish checking if the mosaics were in good condition.

Les enjoying revealing mosaic after a couple of days in the sticky clay Nymphaeum tench

Les enjoying revealing mosaic after a couple of days in the sticky clay Nymphaeum trench

Once again we travel past the bath house trench (more later :-)) up to the Nymphaeum  trench and Peter who has been gallantly digging through the sticky clay to find the  probably roman culvert from the Nymphaeum  spring. It looks like all his hard work has been successful, under the three metal pipes the wall continues and seems to be forming the sides of a stone culvert.

Peter determined to reach the roman culvert

Peter determined to reach the roman culvert

The probable roman stone culvert wall to the right of the pipes

The probable roman stone culvert wall to the right of the pipes

Now back down the steps to the bath house trench, were Fay and Rob are working hard to get to the bottom of the hypocaust pilae (the pillars that the floor sat on, so the hot air could circulate around)  and find what kind of floor is under them.

Rob and Fay working round the pilae to find the floor

Rob and Fay working round the pilae to find the floor

They were joined in the trench by our colleague Claudine, a National Trust archaeologist from Wales.

Claudine happy to be back in a trench digging

Claudine happy to be back in a trench digging

Steve and Max returned to give us a hand back filling. We were not ready to do any, so Claudine who needed to stretch her legs offered to let them dig the bit she was doing. Steve went first and within minutes had found a wonderful object under a piece of flue tile.

A roman Key next to another piece of flue tile

A roman Key next to another piece of flue tile

 

A well deserved find Steve, please don't feel guilty Claudine has forgiven you I am sure :-)

A well deserved find Steve, please don’t feel guilty Claudine has forgiven you I am sure 🙂

So the day ended with a fantastic find and Rob and Fay are poised above the floor. Half a digging day left, so it’s an early night for all.

Day 8 and 9 – Plaster, plaster everywhere and some iron and mosaic!

Due to technical difficulties it’s a bumper edition of the blog 🙂 Day 8 turned into a day of recording, with walls and sections to draw in some of the smaller trenches and the other trenches that are still being dug, had a lovely clean up for photos. Not a lot of fresh digging was done and that that was involved more plaster and ‘little cubes of loveliness’ aka tesserae.

Lovely colourful wall plaster

Lovely colourful wall plaster

Day 9 was an early start as some filming was being done for a documentary about the National Trust. We also had a small section of turf to start lifting to see if there was any mosaic under the turf next to the main north range corridor. We started with a couple of turfs being removed to see what depth the hoped for mosaic was at. Hurray it was there, large whitish and smaller white tesserae of the border of the main entrance room. It did not survive across the whole piece we did but hopefully we will have time to check a larger area.

The small area of mosaic next to the fresh hold stones of the north range corridor

The small area of mosaic next to the threshold stones of the north range corridor

In the north bath house trench loose tesserae hindered the digging, we ended the day with three seed trays piled high with them. The painted wall plaster is still being found, but with no time to check each piece we are waiting for our finds cleaners to have the eureka! moment when they clean of the mud and a face or animal stares back.

A lump of mortar with the ghost lines of the tesserae that have fallen off and lie in the hole it came from

A lump of mortar with the ghost lines of the tesserae that have fallen off and lie in the hole it came from

We have had our first metal objects from the bath house trench a couple if T shaped and L shaped nails/brackets, there are also very small fragments of probable knife blades as well.

L shaped iron object 'a very fine example of its type' as we say when not very sure of what it is!

L shaped iron object ‘a very fine example of its type’ as we say when not very sure of what it is!

More metal was found in the Nymphaeum in the form of water pipes, there were three next to each other, two iron ones and a lead one. Sadly all look to be 20th century. But tomorrow the guys in the trench will be digging around and down to find if the original roman culvert is under these pipes.

Harry cleaning the pipes so we can record them

Harry cleaning the pipes so we can record them

Carol has been slogging away often on her own in the trench below the Nymphaeum one, just behind the wall of the bath house. She has been looking for walls and may have a new one to record. Her best find today was part of an iImbrex which is the curved tile that sat on the join between tegula, the large clay tiles of the roof.

Carol happy with her large piece of imbrex

Carol happy with her large piece of imbrex

We had a treat for lunch, Sue the historic en-actor set up her roman kitchen and we were able to sample her roman creations from bread salad to sweet toast and something with the fish sauce they fermented called garum all very tasty 🙂

Sue and her yummy tempations

Sue and her yummy temptations

Tune in tomorrow to see if we finally find the floor in the north bath house ….

 

 

 

 

Day Six – Saturday sunshine and showers

After such a wet day yesterday, we were hoping that the site would dry out so some trenches could be cleaned up and recorded. As the day progressed it became clear that this was not an option for all!

Fay testing the mud to see of its workable, conclusion time for coffee!

Fay testing the mud to see of its workable, conclusion – time for coffee!

Thankfully the north bath house trench was dry enough to work in so all spare hands were moved to this trench and the top two behind the bath house.

Fay in the dry even the waterproofs are removed

Fay in the dry even the waterproofs are removed

We managed to get some work done between the showers. Lots more plaster is appearing in larger and larger chunks, I think I have underestimated how many boxes we need to transport it for post excavation work when we finish, I may need to hire a large truck!