Castle Studies Trust Blog

Delayed by the pandemic, at long last our 2020 project to see if the Wirk, in the Orkneys, was a castle is now complete. Project leads Sarah Jane Gibbon & Dan Lee look at what they found.

A programme of archaeological fieldwork funded by Castle Studies Trust was undertaken at The Wirk, on the island of Rousay, Orkney by the Orkney Research Centre for Archaeology during 2020-21 (UHI Archaeology Institute). The Wirk comprises the remains of an upstanding stone-built tower with the site of an adjoining hall range to the east, which are stylistically considered to be 12^th century in date. The site was part of a high-status Norse settlement with a medieval chapel nearby (below the present St Mary’s kirk) and Norse hall further to the south at Skaill (Research projects – Landscapes of Change: Archaeologies of the Rousay Clearances & the Westness Estate (uhi.ac.uk)). The project aimed to characterise The Wirk with geophysical survey, undertake evaluation excavation to assess the hall and recover material suitable for radiocarbon dating.

Subscribe to our quarterly newsletter

The Wirk was excavated in the 1920s by J. S. Clouston, who had excavated at other Norse castle sites in Orkney. Later dates have since been suggested for the The Wirk (16^th century), however the date of the buildings, their function and the relationship between the tower and range remained open to debate. Clouston’s excavations focused on the tower and exposed the remains of a hall and ancillary buildings upslope to the east. He made a detailed plan of the site showing the tower and hall. Few finds were recovered but they did include a highly decorated stone finial now on display in Tankerness House Museum, Kirkwall. The impressive tower survives today, but the hall is not currently visible in the neighbouring field.

Donate to support our work

Geophysical survey in 2020 was used to characterise the buried remains of the hall range and identified additional features in the vicinity (Geophysical survey at The Wirk reveals buried walls of the hall – Castle Studies Trust Blog). Evaluation excavation in two trenches in 2021 targeted the eastern hall range and located substantial wall footings just below the ground surface (part of an ancillary building in Trench 1 and the southern external wall of the hall in Trench 2). Excavations concluded that the tower and hall range were built at the same time (contra some earlier interpretations by Clouston). Clouston just exposed the footings of the hall walls, without fully excavating around them, but his plan of the site was proved to be very accurate. The hall may have had raised wooden internal floors, certainly in the lower western part, perhaps supported by an internal scarcement (although excavations were too limited to be conclusive).

A significant assemblage of c.13^th century worked and moulded red sandstone was recovered from Trench 2 nearest the tower. The red sandstone is interpreted as ecclesiastical in origin and likely to have originated from the former medieval kirk nearby, rather than the hall. The Wirk was unlikely to have been an ecclesiastical building due to the raised floors and association with the tower. Radiocarbon dates from material (charred grain) found in deposits abutting the southern external wall of the hall returned Late Iron Age dates (Pictish period mid-7^th to mid-8^th century AD). This material is most likely derived from the disturbance of earlier activity at the site during the construction of the hall, hinted at by the surrounding geophysical anomalies, rather than dating the hall itself. It was concluded that the tower and hall are contemporary and likely to be 12th century in date, based on architectural style and stratigraphic relationships observed during the excavations.

Subscribe to our quarterly newsletter

Donate to support our work

Featured image: View of The Wirk during excavations (foreground) looking south to St Mary’ Kirk and Skaill (Credit: Bobby Friel @TakeTheHighView)

In the second of two articles on the Castle Studies Trust / Historic Environment Scotland co-funded project to date the timber left in a wall socket at Old Wick, Dr Will Wyeth offers an explanation for the surprising date of the timber.

Past investigations of Castle of Old Wick provide a context for the most recent research on this enigmatic Caithness castle. The archaeological evidence combined with historical details give sharp insight into an episode of violence and destruction at the castle in the life of Christian Sutherland, the Lady of Berriedale.

Subscribe to our quarterly newsletter

Based on some similarities with Cubbie Roo’s Castle in Orkney, Old Wick’s standing fabric – a unornamented stone tower with small windows – has been dated to the 12^th century. A survey in 2016 led by Dr Piers Dixon of Historic Environment Scotland (HES) was the first comprehensive assessment of its standing buildings and earthworks since the publication of MacGibbon and Ross’s Castellated and Domestic Architecture in the late 19^th century. Dixon’s study queried the consensus of the castle’s high medieval origins, pointing to regional comparators whose documented history sat more comfortably in a date range beginning in the 14^th century. My review of archaeological and historical evidence for Castle of Old Wick in 2019 substantiated the conclusions of the 2016 survey.

Nevertheless, the simple stone towers of Caithness are poorly understood. They are fairly numerous in the county but our understanding of them relies on an unproductive mixture of simplistic architectural study and a reliance on references in historical sources.

Donate to support our work

Dr Coralie Mills’ and Hamish Darrah’s research gives scope to uphold Dixon’s assertion, and challenge a 12^th century date for Castle of Old Wick. Their analysis of the fragment of alder has given the first substantive dating evidence for the castle with a felling date range of 1515-50 (95% probability).

The slot in which the timber was recovered, located on an internal wall face within the tower, was argued by Dixon to be part of a hanging lum. This is a form of fireplace common to buildings of middling and high status in late medieval Britain, also helpful for dating the construction of the tower at Old Wick. A hanging lum is a fireplace whose hearth and flue are built against, not within, a wall.

Mills and Darrah suggest that the alder was a replacement for an earlier timber used for the same purpose, i.e. to support a hanging lum, therefore, the felling date corresponds with a period of repair, restoration or improvement of the interiors of Castle of Old Wick in the early 16^th century.

Looting and legitimacy

The historical context is one where violence both within and between kin groups is a feature of elite society in late medieval Britain. Typically, these disputes centred on rights of succession to property and titles. Those held by women were the most precarious. In 1517 two parties from the extended Sutherland of Duffus family met to settle a violent succession dispute at Drumminor Castle in Aberdeenshire. William Sutherland of Duffus agreed to an arbitration on the matter of assisthment (compensation for loss) and kynbut (compensation for manslaughter) with Christian Sutherland (the Lady of Berriedale) and her son and heir, Andrew Oliphant. William and his accomplices were held responsible for the murder of Christian’s elder son Charles. Duffus was also accused of seizing and looting two of her properties: Berriedale Castle and Castle of Old Wick.

Subscribe to our quarterly newsletter

The family dispute which led to the murder of Charles Sutherland originated in the legitimacy of Christian’s inheritance of several estates on the death of her father, Alexander Sutherland (d. 1451×1471), including those in Caithness but also Duffus and elsewhere. William Sutherland’s father, also William Sutherland of the fittingly named Quarrelwood, contended that Christian was illegitimate. The court of the Bishop of Aberdeen had found in favour of Christian in 1494, but two years later Quarrelwood violently seized Castle of Old Wick. This was very likely not the same occupation mentioned in the 1517 document. Still unsatisfied, Quarrelwood pursued his case in the court in Rome for several years, until a settlement of sorts around 1507, when Christian surrendered her father’s Duffus lands.

We don’t know exactly why she reached this settlement but it may be telling that her husband’s kin, the Oliphants, had spent substantial sums (not entirely selflessly) on supporting her legal case and accommodating Christian and her children during the difficult years of legal wrangling. We also can’t be sure if the 1517 document references this settlement, or another outburst of violence.

It is tempting to connect the episode of refurbishment at Castle of Old Wick implied by the radiocarbon dating and the documented evidence of looting at the castle which took place before the  1517 settlement, with the implication of subsequent repairs implied by that settlement. I think this is the best conclusion, but others are possible. Between 1515-50 the castle was held by at least seven different parties, but evidence suggests that they were either in financial difficulty or held the castle to generate money from its lands, not as a family seat. Only when the senior branch of the Oliphants take over after 1548 is there a compelling reason to think that the castle was systematically renovated: this is the best alternative scenario to that suggested above.

Archaeologists’ efforts over the last six years have drastically altered our understanding of the Castle of Old Wick. But they have also shed light on the story of Christian Sutherland and violence and upheaval occasioned by her kinsman’s legal contestation. This research demonstrates the value of revisiting the smaller castles of the world, for the potential to challenge an existing consensus as well as shed light on lesser-told stories from the medieval past. 

Subscribe to our quarterly newsletter

Donate to support our work

To read the first blog by Dr Coralie Mills and Hamish Darrah click here: