MHG8490 - Atlantic Roundhouse - Castle Corbet

Summary

No summary available.

Type and Period (4)

  • (Former Type) DUN (Iron Age - 550 BC to 560 AD)
  • (Former Type) BROCH (Iron Age - 550 BC to 560 AD)
  • (Former Type) CASTLE? (Medieval - 1058 AD? to 1559 AD?)
  • ATLANTIC ROUNDHOUSE (Iron Age - 550 BC to 560 AD)

Protected Status

  • None recorded

Full Description

Castle Corbet (NAT) (Supposed remains of)
OS 6" map, Ross-shire, 2nd ed., (1907)

Macfarlane (1906-8) in first half of18th century mentions Castle of Easter Arbol "now ruinous, belonging one to Corbat of Easter Arbol" and situated "a short mile W of church" (of Tarbat). Davidson (Davidson 1948), however, states that Castle Corbet was quite unknown to farmer on whose land it was situated.
A low, surrounding ditch was visible in 1872 (ONB 1872).
Name Book 1872; W Macfarlane 1906-8; J M Davidson 1948.

The site of this alleged castle is on a spur between a stream gully and a raised beach. All that survives are overgrown footings of an off-square stone-walled structure of uncertain origin. There is no trace of ditch seen 1872. No further information obtained locally. Resurveyed at 1:2500. Visited by OS (A A) 14 September 1972.

The wasted remains of what is probably a broch. It measures 22m in diameter over a wall 5.5m thick; a single outer facing-stone is visible on NW. The remains are obscured by dense vegetation and an enclosure of no great age.
RCAHMS 1979.

The remains of this feature, which are those of a dun rather than a broch (see NH87SW 6 and 8), are generally as described above. The spread wall, which has an internal height of 0.5m, has been crudely dug into in N arc revealing a stone and earth mixture, which confirms that wall lacks quantity and quality of stone expected from a broch. No entrance is evident. Three rough pits dug in interior reveal nothing significant. Within dun are ill-defined remains of a sub-rectangular enclosure, 11.5m by c8.5m, SW side having been destroyed. The wall is spread to 1.5m and is 0.4m high. These remains are later than the dun, and seem too weak to be those of a 'castle'.
On SW side among modern stone clearance is a possible cup-marked stone bearing three poorly-defined cups.
Surveyed at 1:2500. Visited by OS (J B) 3 March 1981.

NH98 1 CASTLE CORBET ('Cnoc Tigh')
NH/9026 8326
This possible broch or dún in Tarbat, Ross and Cromarty, was first mentioned as a castle in the first half of the 18th century [2] but was unknown to the local farmer two centuries later [3]. A suitable site was found in 1972, on a spur between a stream, gully and a raised beach and an off-square structure identified [1]. A few years after that it seemed more like a broch with a diameter of 22m over a wall 5.5m thick [1]. A single outer facing-stone was visible on the north-west. A later visit in 1981 returned the site to the dún category and noted that crude excavations into part of the wall on the north arc had revealed no sign of the core of massive stone rubble one would expect in a broch wall [1]. On the south-west side, among modern stone clearance, is a possible cup-marked stone bearing three poorly-defined cups.
Sources: 1. NMRS site no. NH 98 SW 12: 2. Macfarlane 1906-08, 215: 3. Davidson 1946, 28: 4. RCAHMS 1979a, 23, no. 184. <1>

The remains of this dun stand on very low knoll on a promontory formed by the steep face of a raised beach on the WNW, a deep stream gully on the N and a shallower one on the E; to the S there is a broad terrace the cultivation of which has resulted in the truncation of the foot of the knoll on that side.
The dun measures about 17m in diameter over a wall that has been very heavily robbed of its stone and has largely been reduced to a grass grown stony bank of up to 5.5m in thickness and 0.8m in height. The only facing stones visible on the date of visit were those comprising a short row exposed on the WNW. The interior of the dun appears to have been levelled to provide a platform on which there are the remains of an enclosure. Of irregular shape, it measures about 14m from NW to SE by 12.5m transversely within a stone wall 0.5m thick and only 0.2m high. It is depicted on the 1st edition of the OS 25-inch map (Ross-shire XLIII.1, 1880), though nothing is now visible of a second smaller enclosure shown immediately to the NNW.
At least six trenches have been excavated into the remains, probably within the past fifty years judging from the relative freshness of the cuts, one of which on the NW has been driven into the thickness of the dun wall from outside and another on the SE which slices through the enclosure wall. In addition to the truncation to the southern edge of the site by the plough, cultivation has also resulted in a considerable amount of field-cleared stones being dumped on the site, in places obscuring the detail of the underlying remains. At the edge of this clearance on the SW side of the dun there is the cupmarked boulder that was first identified by the OS in 1981.
Although the 1872 description of this site in the OS Name Book mentions a ‘low surrounding ditch’, none was depicted on the contemporary 1st edition 6-inch map (Ross-shire and Cromartyshire (Mainland) 1880, Sheet XLIII) and no trace of such a feature is visible today. Fragments of a slight bank, almost certainly of comparatively modern origin, are visible immediately E of the dun where it skirts the crest of the previously mentioned shallow stream gully.
Visited by RCAHMS (GFG, JRS, IP) 27 February 2014.

As part of the Northern Picts Project surveys and excavations have been undertaken from 2012 to 2014 on six Atlantic roundhouses on the Tarbat Peninsula to track the nature of settlement in this landscape in the first millennium BC to AD. The investigation of the thick-walled dry stone roundhouse Cnoc Tigh was undertaken in June and July 2014. Cnoc Tigh (also known as Castle Corbet) is located on a knoll at the edge of a raised beach (c12m OD), about 300m SE of the current shoreline overlooking the Dornoch Firth. The site is flanked to the NE by the deeply-cut ravine of the Allt a’ Chaoil-eag which, with the steeply sloping raised beach, forms a small promontory. The roundhouse is visible from the Pictish monastery at Portmahomack, located less than 1km to the NE.
Excavation of a 15 x 10m trench across the N side of the roundhouse identified three phases of construction and remodelling of the dry stone walls. In its primary phase the roundhouse wall was only c1.2 m thick. The house had a c13m internal diameter with a 1m wide entranceway located due W. The second phase of the house saw the construction of a boulder wall against the outer face of this primary wall, increasing the width of the walls to c2.8 m. In its final phase the walls were further widened by a rubble and earthen core laid against the Phase 2 wall, bounded by an outer face of boulders. In this final phase the roundhouse walls were c4m thick.
Internally two phases of occupation deposits were separated by a layer of windblown sand, an abandonment or re-flooring episode. A scarcement ledge was keyed into the inner face of the primary wall of the roundhouse. As the top of the ledge was located only 0.20m from the base of the wall this may have supported a timber floor at ground-level. Overlying the roundhouse was the remains of a 19th/20th-century rubble enclosure. Apart from post-medieval glass and pottery within the enclosure rubble, no finds were identified at Cnoc Tigh.
Geophysics around the southern edge of the roundhouse did not reveal any archaeological activity. A cup-marked stone slab located on the S side of the knoll is potentially the former capping stone for a Bronze Age cist located at NH 9008 8324. <2> <3>

Sources/Archives (7)

Map

Location

Grid reference Centred NH 9025 8326 (80m by 80m) (2 map features)
Map sheet NH98SW
Geographical Area ROSS AND CROMARTY
Civil Parish TARBAT

Finds (0)

Related Monuments/Buildings (2)

Related Investigations/Events (2)

External Links (1)

Comments and Feedback

Do you have any more information about this record? Please feel free to comment with information and photographs, or ask any questions, using the "Disqus" tool below. Comments are moderated, and we aim to respond/publish as soon as possible.