Archaeology Notes

Event ID 641826

Category Descriptive Accounts

Type Archaeology Notes

Permalink http://canmore.org.uk/event/641826

HU32SE 4 3685 2090.

(HU 3685 2090) St. Ninian's Kirk and Burial Ground (LB) (Site of)

(HU 3667 2068) St. Ninian's Well (LB)

OS 6" map, Shetland, 2nd ed. (1903)

St Ninian's Church was abandoned at the Reformation, but was venerated locally until its destruction in 1750. The graveyard continued in use until c.1850. The site was recovered by excavation by Prof. O'Dell between 1955 and 1959, revealing the foundations of a 12th century church, overlying a pre-Norse(?) church, which in turn overlay an Iron Age site (possibly a wheel house) yielding pottery datable to about the 2nd century AD. Each level was accompanied by contemporary graves, those of the Iron Age being centred on a crouched burial of late Bronze Age type, which lay SW of the chancel of the medieval church.

A C O'Dell 1958

By far the most important discovery on the site was that of the 'St. Ninian's Isle Treasure:- a hoard of possibly ecclesiastical silver-ware, found on 4 July 1958 - which lay under a cross-marked slab in the floor of the (?)earlier church; and had been contained in a larch box. This is supposed to have been hastily hidden on the approach of Norse raiders. It was donated to the National Museum of Antiquities of Scotland (NMAS) in 1965-6.

A C O'Dell 1959; D McRoberts 1963; Proc Soc Antiq Scot 1967

The sculptured posts of a slab shrine also dateable to c.800 AD were recovered south of the chancel of the mediaeval church, and to the north of the church and running parrallel with it was what may have been a garth wall which was traced for about 30'.

Ogham - inscribed stones, one of which is in NMAS had previously been found in the graveyard.

G Goudie 1878

St. Ninian's Well is described as an ancient well.

Name Book 1878; F T Wainwright 1962

This complex is as described and planned by the previous authorities except that the Iron Age site lies to the south of the church. St Ninian's Well is a natural spring.

Visited by OS(RL) 10 May 1968.

177. Sword-pommel. Silver-gilt, length 5.5cm, height 3.1cm. Pictish or ?Anglo-Saxon, late 8th century. NMS FC 278.

Hollow one-piece carving in 'cocked-hat' form with internal corroded concretion. Each side has crisply incised decoration with punched detail, done to an identical pattern: an arched central panel has two confronted animals with shovel-shaped snouts and speckled bodies, whose sinuous and extravagantly interlaced trunks eventually find their way to opposed pairs of haunches. Each flanking panel has a motif of two animals constructed on similar principles, placed one above the other and loosely interconnected.

This object and the two scabbard-chapes (nos. 178a and b) formed part of a large hoard of Pictish silver brooches, bowls and other artefacts excavated in 1958. The 'cocked hat' shape of the pommel clearly derives from late seventh-century one-piece silver Anglo-Saxon pommels such as that from Crundale. Its dominant central panel, however, is certainly an eighth-century development which is lossely paralleled on the embellished iron pommels from Windsor and Chiswick Eyot and, less directly, on the very elaborate hilt from Fetter Lane. Its decoration, like that of the chapes, though usually acknowledged to be Pictish in execution, reveals at least very strong Anglo-Saxon influence. The use of gilding and the interlacing speckled bodies are readily paralleled on Anglo-Saxon objects such as the Chelsea ring, while the spiral element on the forequarters of the flanking animals is also widespread in Insular art, including some metalwork thought to have been made in southern England. Likewise, the shovel-shaped muzzles seen also on the chapes are most closely paralleled on the Fetter Lane hilt and the Gloucester key. Only the striking spiral convolution of the hindquarters and interlocking limbs recall the contortions of Pictish animals as seen, foe example, on the St Ninian's Isle bowl 3 and the Aberlemno churchyard stone. But this, too, is a motif familiar from Northumbrian art, the Lindisfarne gospels for instance. Thus, both in form and style, the pommel may reflect not simply an acknowledged Pictish facility in adapting an Anglo-Saxon artistic vocabulary, but could be an actual import of the later eighth century. At the very least, this and the associated chapes show the strength of Anglo-Saxon influence in eighth-century Pictland.

178 (a, b). Two scabbard-chapes. Silver, gilding and blue glass; maximum width (a) 8.1cm, (b) 8.2cm. Pictish or ?Anglo-Saxon, late 8th century. NMS: (a) FC 282 and (b) FC 283.

178 (a) Three-piece U-shaped casting. Composed of two side-pieces and a binding strip which caps the join. The binding strip is held by rivets and terminates at each end in a stylised animal head. The chape terminates in two shovel-snouted open-jawed animals which differ slightly on front and back.: on the back their teeth are clenched in a jagged grin while on the front paired openwork fangs trap a tiny fish. The eyes were probably once set with blue glass and a trumpet scroll is produced behind each: the muzzles are ribbed. On the front, three brambled dummy rivets divide an inscription in insular script which reads INNOMINEDS ('in the name of God'). On the back, a continuation of this reads RESADFILISPUSSCIO ('property of the son of the holy spirit'). The crest on the front has pairs of carved incisions, on the back a frieze of gaping broad-muzzled beast heads.

178 (b) Similar casting composed of two side pieces riveted together. It terminates in two shovel-snouted fanged animals with gaping jaws, ribbed muzzles and protruding tongues. The lentoid eyes are set with blue glass studs and the face is decorated with incised triskele scrolls. A key-pattern collar on both sides separates the head from the main panel, which on the front has two fields of interlocking speckled animal pairs, seperated by a central scroll pattern, and on the back a triangular spiral pattern. The crest has key-pattern and spiral motifs.

These two chapes come from the same hoard as the pommel. They have no surviving Insular parallel but there can be no doubt that they are fittings from scabbards. Both in the script of (a) and in their decoration they draw on a range of Insular motifs. In particular, aspects of the fanged animal heads with their broad ribbed snouts, trumpet-spiralled eyes and openwork jaws recall Northumbrian parallels, the runic mount from the Thames and the more recent find of the crest terminal on the York helmet. The ribbed spine, blue glass eyes and openwork fanged jaws of the former, and the ribbed shovel-shaped muzzle and spiralled eyes of the latter are strikingly similar. The helmet has two further points of similarity, its ribbed crest and eyebrows and the Christian invocation in Insular script which runs along the crest. Like this, the inscription on chape (a) invokes divine protection for the owner. The interlocking animals on chape (b) cannot be clearly paralleled in Anglo-Saxon art although, like the ornament on the pommmel, they owe a clear debt to Anglo-Saxon sources. The Shovel-snouted frieze of animal heads on chape (a) likewise recalls the Fetter Lane pommel and the Gloucester key.

L E Webster and J Backhouse 1991.

HU 368 209 The site is one of the most renowned in Scotland, having produced the St Ninian's Isle hoard - the best example of Pictish silverware yet discovered. However, little is unfortunately known of the archaeology of the site, other than its considerable potential. The site was excavated by Prof O'Dell between 1955-59, revealing the foundations of a 12th-century church overlying an earlier building (NMRS HU32SE 4). Sculptured stones, including those from Early Christian corner-post shrines, and later Norse crosses, were also found. Below and to the S of the church an Iron Age site was identified. There are burials at the site associated with each period, including as late as the mid-19th century AD. The archive from the 1950s excavation is very limited due mainly to the untimely death of the excavator before he was able to publish.

The fieldwork undertaken in 1999 comprised an initial desk-based study followed by topographic and geophysical surveys of the site and immediate surrounding area. Four small trial trenches were then opened to test the results of the survey and assess the condition of the archaeology at the site. The site is being very badly eroded by rabbits and is in some state of disrepair. Iron Age short cists and Early Christian long cists first uncovered by O'Dell to the S of the chapel were found to be largely intact in the small area investigated. Both later burials and earlier Iron Age midden deposits were also found to survive in parts of the trench. Evidence of the survival of later burials was recovered to the N of the chapel, together with indications of the extent of O'Dell's work to the E. Cultivated soils dating to a period of post-medieval settlement were identified to the N of the site. A plane table survey and photographic record of the chapel were also completed.

Sponsors: Historic Scotland, University of Glasgow, Shetland Amenity Trust.

R Harry 1999

Excavations carried out July 2000 by Glasgow University on behalf of Historic Scotland.

see NMRS MS/1069/13 (Glasgow University, 2000)

HU 368 209 Following a desk-based study, topographic and geophysical surveys and excavation of four small trial trenches in 1999 to define the extent and assess the condition and nature of the archaeology at the site (DES 1999, 81), two larger excavations were undertaken in July 2000.

Trench 1 was opened to the S of the chapel (NMRS HU32SE 4), within the fenced area, to establish a chronological sequence for the burial and structural remains, and to distinguish between those features excavated, and possibly reconstructed, in the 1950s by Prof Andrew O'Dell, and those undisturbed by his excavations.

The earliest phase on the site was of an Iron Age structure, the beginnings of which were uncovered. Overlying this were two areas of midden, one containing much pottery and animal bone, the other comprising mainly shells. Three crouched burials were buried, rather haphazardly, into these midden deposits. Above these, stone cists were constructed, which had already been excavated by O'Dell. At least one of these had contained a crouched burial, aligned N-S. The cists were also associated with a wall, aligned N-S, across the area. The wall had largely been excavated and then reconstructed in the 1950s. Schist, the bedrock of the island, was used to good effect through the Iron Age features, used as the end slabs for each cist and as part of the wall. Adjacent to these, and undisturbed by the previous excavations, a kerbed feature, filled with quartz pebbles and stones, and separated into five 'compartments' by upright stones, was excavated. The compartments were aligned E-W, and at the W end of each compartment was a small upright stone, like a headstone. On the surface of two of these stones were inscribed crosses, both of the same form - a double cross with possible interlacing of the lines and the beginnings of an interlace design at the base. Beneath the pebbles and stones of this complex, six infant burials were excavated, buried into the midden below. Slightly above these burials two long cists were excavated, each of which was empty. These had also been filled with quartz pebbles, although it is possible they were disturbed in the 1950s. Adjacent to one of these was a smaller cist, which appears to have been truncated. This was covered by a flat schist slab which had been inscribed with a simple cross. It seems likely that these burials all date to the Early Christian period, about which little is known in Shetland. DNA work is planned on the infant burials to try to determine matrilineal descent, and dating will be carried out on all the human remains to determine a data sequence. There are no Iron Age cemeteries excavated in Scotland or the Northern Isles with which to compare these, and only one other (Pictish kerbed cairn) Early Christian burial excavated in Shetland. Finds from this trench include bone beads, large amounts of Iron Age pottery, a broken gaming piece, and worked whalebone.

Trench 2 was excavated to the N of the chapel, outwith the fenced area, to attempt to define the extent of the medieval and post-medieval burials at the site. This area of St Ninian's Isle was used continuously as a burial ground from the Iron Age to the mid-19th century AD. O'Dell excavated probably Late Norse long cists S of the chapel, and removed 'thousands' of medieval and later burials from around and above the chapel, buried after the chapel had gone out of use in the 16th century. Unfortunately these remains have been lost, and although there are local stories of the graveyard, the extent has never been known. The geophysics and trial trenching in this area in 1999 showed the presence of burials to the N of the chapel, and so a larger trench was opened in 2000. Three post-medieval burials were excavated from Trench 2, and evidence was found of a curved wall aligned roughly N-S, across the trench.

Sponsors: Historic Scotland, Glasgow University, Shetland Amenity Trust.

R Harry 2001