If we consider settlement and the nature of settlement then various sites and site types may help us.The site at the Brough of Birsay has been mentioned already. Here evidence suggests a Pictish high status site. John Hunter’s excavations on the south-west side of the island, which is eroding, showed cellular or curvilinear shaped Pictish buildings each set in a plot of land. These were replaced perhaps in the early ninth century with rectangular buildings of stone construction, a type usually associated with the Norse, but which respected the land divisions. The Orkneyinga Saga tells us that in the mid eleventh century the Norse Earl Thorfinn of Orkney “had his permanent residence at Birsay, where he built and dedicated to Christ a fine minster, the seat of the first bishop of Orkney.” The earldom was probably established by the late ninth century. The seat of the earls and minster at Birsay might be on the Brough, but equally it could be under the present village on the mainland.
The Brough of Birsay, a tidal island off the west coast of Orkney mainland (foreground), looking into Birsay Bay and towards the modern village of Birsay(top centre right). In the foreground is one of the rectangular stone-built buildings of the Viking Age.
However, the settlement on the Brough does seem to have been re-planned in the late ninth centry and it is tempting to associate the change with the new earldom. A central drainage system was installed and the new buildings all took the same alignment.
Brough of Birsay with a number of buildings of the Viking Age. These comprise simple rectangular stone-founded buildings with a single entrance interspersed with larger rectangular buildings some with central hearths and side benches. Perhaps they were houses with outbuildings.
In the late 10th or early 11th century the whole site was cleared, levelled and a new plan adopted, suggesting an authority behind the changes. Older excavations showed hall houses up to 20m in length with central hearths and low benches along the long walls.
Brough of Birsay showing details of one of the Viking Age houses excavated in the 1930's. Note the stone walls with rounded corners, the entrances - one in the short end and two opposing entrances in the long walls (one to the left of the figure)- and the raised side benches along the inner long walls (visible as low grass mounds to right and left of the entrance in the forground) .
Brough of Birsay - detail of a side bench lined with stone in one house.
Brough of Birsay: shorter building with central drain. The stone lined drain, with stone slabs covering it, emerges through a gap in the end wall (centre foreground).
Shorter outbuildings were included in the complex, which unfortunately cannot be dated. Metalworking continued on the site. Steatite moulds for ingots of precious metal and crucibles have been found. Bone pins of Pictish type have been found in Viking levels, which have also yielded combs of both Pictish and Viking types.
Another site of potential high status was at Skaill, on the eastern coast of the main Orkney island. Once again the clue is in the Orkneyinga Saga which mentions that Thorkel Amundason, foster father of Earl Thorfinn lived at Skaill. However, skaill, derived from old Norse skali or hall is a fairly common name and the location is not secure.The site lies above a sheltered sandy bay, one of the few safe landing places on this rocky coast.
Excavation at Skaill has revealed a long sequence of settlement in the area. It includes a house 5m x 7.6m.in dimension constructed of large facing stones with an earth core. It had two rooms, with a central hearth in the northern room. Construction was neat and careful and a midden associated with the house yielded Pictish bone combs and needles. A period of abandonment followed which ended when a flimsy structure was built partially over the earlier house, re-using walls and materials from the previous period. The excavator, Peter Gelling, suggested this marked the arrival of a Viking war-band building a shelter for the winter. A steatite bowl may be associated with this phase.
A sand blow marked a second period of abandonment after which a substantial longhouse was built on the site. This had a raised hearth and side benches, though neither feature could be claimed as diagnostically Viking. A possible bath-house, which would be seen as a Viking innovation, lay nearby. This had hearths at either end with benches and a sump in one corner. A complicated building on the same site, below a much later structure, might be the remains of Thorkel’s house.
The site at Earl’s Bu, Orphir is also mentioned in the Saga in a description of Earl Paul’s Christmas in 1135. “There was a great drinking hall at Orphir, with a door in the south wall near the eastern gable, and in the front of the hall, just a few places down from it, stood a fine church.”
Remains of the Earl of Orkney's hall at Earl's Bu, Orphir with the remains of the twelfth century church behind the modern cemetery wall above left.
Beneath the middens associated with this hall, which lies behind the modern building in the top right hand corner of the picture, a small water mill of 10th or 11th century date has been excavated recently. It shows that by this time the earls were probably able to insist on centralised milling of corn. The head race, which brought the water into the mill in a wooden chute set in a stone lined channel is pictured here.
Horizontal mills are known from Ireland as early as the 6th century AD and could have been in use in Pictish Scotland. Here is a reconstructed click mill on the island of Harris, Hebrides.
The scale of the enterprise suggests centralised milling under the earl’s control. Close by archaeology has also revealed iron and copper alloy waste from metalworking. The buildings on the site, excavated some time ago, are more difficult to interpret, but a ruined church still stands nearby. It has a semi-circular chancel and circular nave and is in the tradition of centrally planned churches which began in Europe with The Emperor Charlemagne's chapel at Aachen. The Orphir church is 12th century.
The site at Cunningsburh, Shetland is of great interest in understanding the Norse takeover of Pictish high status sites. Here, there are profilic steatite (soapstone) deposits, which were worked in the Viking Age. The shapes of bowls cut from the deposits can still be seen. This was a site of high status activity in the Pictish period, which was apparently taken over by Viking rulers. The steatite would have been a major attraction.
Here, excavation in the 1940’s revealed a rectangular aisled house 22m by 5m with one wall slightly bowed. The internal timber posts supporting the weight of the roof may have been imported from Norway since no such timber was available in the Northern Isles. The walls were of stone alternating with turves, an unusual construction. Internal benches were inserted on each long side and the remains of a central raised hearth were discovered. The dating of this first Viking Age phase was difficult, as it depended almost entirely on the find of an Irish bronze mount, of late 8th or early 9th century manufacture.
Interior of the main house at Jarlshof as displayed on site. The house was reconstructed several times in its history. Here the paved byre with central drainage channel of the longhouse phases is clear.
Outbuildings included a barn or byre, a smithy and a possible bath-house or corn drier. By the 11th century the house had been extended into a long-house, by the addition of a byre at the end of the house. At about the same time a second longhouse was added at right angles to the first. The site was occupied until at least the thirteenth century, although it is doubtful if more than two houses were occupied contemporaneously. Carvings incised onto slate and sandstone depict ships and people and appear to be in Pictish style, suggesting the survival of Picts in the area into the Viking Age, although their status is unknown.
At Underhoull on the island of Unst, most northerly of the Shetland Islands another house complex has been excavated. Again this is a typical site above a sandy bay, where there are traces of occupation in earlier periods.
The site at Underhoull on the island of Unst, northernmost of the Shetland islands. The remains of the house are centre right. Other possibly contemporoary remains with a midden occur above the bay. The sandy bay would have been attractive to settlement. Pre-historic ritual sites exist at the top of the hill above the site.
The house, some 17m by 4.6m had a byre at its eastern end and a single line of internal posts supporting the roof. There were at least two phases of construction, as rooms had been added on its south side. Steatite pottery was found in the excavation and a source of steatite exists on Unst, which appears to have been worked in the Viking Age. A midden 200m distant from the house strongly suggested that other houses existed in the area at this time. Midden material included bones of domesticated animals and fish. Together with the rotary querns found on the site this evidence suggests mixed farming was practised. Dating evidence generally was poor. The Underhoull building might have been in use within the period 9th to 11th centuries.
Few farming settlements of the Viking Age have been identified and investigated in Scotland, perhaps no more than about ten. Most of these are atypical in that they have been revealed by sea erosion and have been excavated in rescue conditions. Finds make it difficult to assert that the buildings were constructed and occupied by Vikings. Features which have been taken to be diagnostic of Viking presence include the rectangular shape of the house - compared with the cellular Pictish houses - side benches and raised central hearth, although these latter elements were not unknown in Pictish houses. The presence of steatite on sites may suggest Scandinavians present since there is no evidence that this material was used for containers in the Pictish period. However, the importance of the site at Cunningsburh in the Pictish period, referred to earlier, should not be overlooked, since this may suggest that Pictish leaders were aware of the steatite.
Place-name evidence will be dealt with in a separate section, but here we should note that Scandinavian place-name elements survive in very large numbers in the Northern Isles and are frequent too in Caithness and in the northern half of the Western Isles. They occur also further south on the eastern and western mainland coasts and in the Western Isles. A detailed discussion of this material follows in a separate section, but we should be aware that the place-name evidence does suggest at least the very widespread influence of the Old Norse language in Scotland, especially in the Northern Isles. We could go further and suggest that in some areas we may well be seeing Scandinavian colonisation on some scale. What were the inland settlements as indicated by place-names like? Are the few sites exposed by sea erosion typical? There are however, as you will see, major problems in interpretation.
A whole landscape approach, which studies the total palimpsest of settlement over time, has been adopted for a project run by the Universities of Sheffield and Cardiff on the island of South Uist in the Hebrides. Here, their studies have revealed the boundaries of the 32 ‘townships’ or local political divisions which existed into modern times. These straddle the island from east to west allowing each township its share of the hilly terrain to the east, the machair to the west and the peat in between.
Typical vegetation on the machair in the Western Isles. This was light, well-drained soil composed mainly of broken shells and which attracted settlement from the Bronze Age.
Typical peat deposits in the central part of several of the Western Isles with the land rising to hills in the east
The boundaries appear to go back at least to the Iron Age and occupation, on the machair, was continuous from that period.
Hay fields on the peat in Lewis.
At Upper Bornish two settlement mounds have been excavated revealing Viking Age longhouses, but with floors below ground level. Sunken features were characteristic of buildings on South Uist since the Bronze Age, so it appears that native building tradition here influenced the Scandinavian newcomers.This, plus the use and maintenance of existing township boundaries show elements of continuity in the landscape of South Uist.
Deserted settlement on Lewis. The site has not been excavated, but desertion was probably in modern times - Highland clearances - but earlier phases on the site may include the Viking Age. Note its position above the loch.
Compare these results with the situation on the island of Coll, southern Hebrides, where Johnston has shown that 50% of the names on the island are of Norse origin and seven out of ten of the primary estate names are Norse.
These units run across the island from the machair in the west to the thin rocky acidic soils in the east, thus ensuring that each estate had a share of different types of land. Each estate was eventually known as an ounceland, that is the rent for the estate was one ounce of silver per annum. The evidence is insufficient to show whether this organisation was already established on Coll before the arrival of the Vikings or whether it was established by Scandinavian overlords, conceivably even the Earls of Orkney. But such studies do serve to show that land settlement and the relationships between natives, colonisers and lords may be more complicated than we have yet realised and that there may be considerable differences from one area of Scotland to another.
Questions for you to think about
- What clues exist at Cunningsburh to suggest that it was a high status site?
- Which term best fits the settlement at Jarlshof - family farmstead, hamlet, town or village or something else?
- What emerges from a ‘whole landscape’ approach in the Western Isles? Is this a better approach than excavating individual sites?
Quiz
Now click here to try a quiz which tests your knowledge of this unit.