ISSN 1357-4442Editor: Simon Denison

Issue no 38, October 1998


Castle where they `played games in wait for battle'

Excavations by a team of British archaeologists at a castle in France, thought to be the oldest surviving secular building in western Europe, have produced strong evidence that it was built around AD900, or earlier, in the Carolingian era. A large part of the 10th century Chateau de Mayenne, at Mayenne in north-western France, survives, encased in 13th century buildings.

The excavations, by the Oxford Archaeological Unit, have also produced a large number of gaming pieces from an early form of backgammon, as well as part of a backgammon board. They are thought to date from the 10th or 11th century, making one of the largest collections of early backgammon pieces yet found in Europe (see BA, February, and July 1996).

The 10th century castle consisted of a rectangular stone hall with basement and pitched roof, with a tower at one corner rising possibly to three storeys, within a stone enclosure wall. The structure survives almost complete, missing only the top storey of the tower which was removed in the Middle Ages. The tower and hall basements were later backfilled for defensive reasons, possibly in the 12th century, leaving sealed 10th and 11th century deposits including ten 10th century coins of Charles the Simple and Louis d'Outremer, numerous items of military equipment and the gaming pieces.

The circular backgammon pieces, made of bone and ivory, were carved with animal motifs and ring-and-dot patterns. According to Rob Early, the excavation director, the discovery of a number of unfinished and broken pieces suggests some were produced in the castle, while other more decorated examples were probably carved by specialists and brought from elsewhere.

The castle was built partly out of re-used building material from the Roman fortress at Jublains, a former civitas capital nearby, including large granite blocks for the tower, and brickwork used for arches in the hall and for windows and a door in the tower. The re-use of Roman building material suggests a shift of power from Jublains to Mayenne in the post-Roman period, Mr Early said. Evidence for an earlier timber structure has been found at the castle suggesting it may have been the established seat of a local leader.

The 9th and 10th centuries were a period of repeated Viking raids in the area, and the excavators believe the castle was rebuilt in stone as an `expression of power' to warn off any potential raiders. Lying in the buffer zone between Normandy, Brittany and Anjou, it was later besieged by William the Conqueror in 1063. The absence of domestic finds from this period produces a strongly military impression. `You do picture these people waiting around, polishing their weapons, playing games, and waiting to do battle,' Mr Early said.

Return to Table of Contents | Return to CBA Homepage

Orkney cairn `let in the morning sun'

An underground Neolithic chambered cairn on Orkney has been found to have a possible `light box' cut into the roof to allow the rays of the rising sun to enter the tomb at certain times of the year. The tomb seems to have lain undisturbed for 5,000 years until a roofing stone was dislodged by a tractor earlier this year (see BA, June).

Although the light box, and any solar alignments, remain unconfirmed, the lintel over the doorway of the tomb at Crantit, near Kirkwall, has a slot cut into its top which suggests a deliberately-made aperture. The tomb faces east-south-east, and it is thought that the light of the rising sun may have entered the tomb around 21 February and 21 October.

If the light box is confirmed at Crantit, the tomb will become only the second in the British Isles known to have had this feature. The other is Newgrange, a passage grave in Co Meath, Ireland, where the light enters the tomb for 17 minutes at sunrise on 21 December, the winter solstice.

Excavations this summer by the Glasgow University field unit, GUARD, directed by Beverley Ballin Smith and Colin Richards, found that the tomb was unusual in a number of respects. Small, and consisting of three chambers arranged like a `flattened clover leaf', it was dug into the side of a hill, thus becoming invisible in the landscape. Most chambered cairns on Orkney and elsewhere, by contrast, were built on the surface as prominent mounds. It contained the skeletons of only four individuals - a mature woman, a girl, a child, and one too degraded to identify - with no pottery, artefacts or other deposits. The entrance was heavily blocked up, both inside and out, in a way that suggests the tomb was not intended for re-use in the traditional Neolithic manner.

One internal upright stone was decorated with interlocking zig-zags and parallellograms, and the stonework of the walls and ceiling was skilfully assembled. According to Ms Ballin Smith, far from suggesting a `poor example' of a tomb, Crantit's unusual characteristics probably reflect a cultural change in the treatment of death and burial, with an emphasis on going `down into the earth'. The date of the tomb, within the Neolithic, however, is at present unclear.

Return to Table of Contents | Return to CBA Homepage

Roman HQ `later became a guardhouse'

An early Roman military headquarters at Fishbourne in Sussex, which transformed historians' understanding of the Roman invasion of AD43 when it was discovered two years ago, is now thought to have survived for up to two centuries as a guardhouse in front of Fishbourne Roman Palace. The new view follows further excavations at the site this summer.

The large stone building, dated AD50-60, was identified as a principia when a sunken-floored room - probably a strong-room holding soldiers' pay - was discovered in the middle of the building's eastern wing in 1996. The discovery influenced the view that the invading Roman army may have landed in Sussex, in order to restore the ousted King Verica of the Atrebates (see BA, September), as well as (or, perhaps, instead of) at their traditional landing-site of Richborough in Kent.

The military building was assumed to have been demolished in about AD75 when Fishbourne Palace was built, possibly for Verica's successor, King Togidubnus. The assumption was based on the idea that the owner of the palace would not have tolerated such an ungainly building within 20 metres of his grand new front entrance.

Further work at Fishbourne by the Sussex Archaeological Society, however, has confirmed the relative absence of small finds from the site of the building compared to surrounding areas. Some 1,800 small finds - sherds of glass and pottery, nails, and the like - have been removed from the building over three seasons of excavation, compared, for example, to over 5,000 found this season from a limited area to the north. According to John Manley, chief executive of the society, the low numbers can best be explained by the survival and repeated cleaning of the building, perhaps until the late 3rd century, when a drainage gully through one wall suggests that the building had finally been taken down.

Togidubnus probably died before AD100, and his client kingdom was absorbed into the Roman province. It is possible that the palace then became one of the residences of the provincial governor. `We now think that the principia building remained as a formal reception centre to vet visitors to the palace, and to provide a symbolic military presence, like the Household Guard at Buckingham Palace,' Mr Manley said.

Return to Table of Contents | Return to CBA Homepage

In brief

Caracalla gem

A carved intaglio, or engraved gem, bearing the head of the Emperor Caracalla (AD211-17) has been found during field-walking at the Roman fort of Trimontium at Newstead in the Scottish borders. The fort was used in AD208 as the base for a major campaign in Scotland by the Emperor Septimius Severus, who brought his sons Caracalla and Geta with him.

The gem, used as the centre-piece of a signet ring, now suggests that Caracalla, or his troops, later returned to Scotland. It may have been presented by Caracalla himself to an important member of his retinue during a ceremony at the fort. The oval piece of jasper, found by Walter Elliot of the Trimontium Trust, is about the size of a fingernail, and is in mint condition.

A series of ditches and mounds uncovered at Inveresk Fort, near Musselburgh in East Lothian, have been interpreted as a `first line of defence' for the fort's construction workers against possible attacks from locals. The fort was built in about AD140-65, at about the same time as the Antonine Wall. The excavation, by AOC Scotland, has also produced coins, pottery, spear-heads and shield handles.

Ancient forest

A mile and a half of prehistoric forest has been discovered on the south bank of the Thames at Erith on the outskirts of southeast London. The crumbling remains of oak, ash, alder, Scots pine and yew are thought to date from the Neolithic, and to represent a wooded island between two Thames channels.

The area contains signs of possible raised trackways and fishtraps, and is scattered with flints and pottery. The woodland was probably submerged by rising waters, and Bronze Age pottery has been found overlying the forest remains. It is now being recorded by the Thames Archaeological Survey directed by Mike Webber.

An early Neolithic wooden club that resembles a cricket bat has been found by the Thames Archaeological Survey near Chelsea. Made of oak, and about 2ft 6in long, the club has a rounded handle with a knob, and has been radiocarbon dated to 3540-3360BC.

Return to the British Archaeology homepage

Return to the CBA homepage

© Council for British Archaeology, 1998