Publication Account
Date 2002
Event ID 575387
Category Descriptive Accounts
Type Publication Account
Permalink http://canmore.org.uk/event/575387
HU41 5 LEVENWICK ('Broch of Levenwick', 'Helm')
HU 415197
Visited 5/6/63 and 30/6/87.
Broch, probably solid-based, on flat ground at the edge of the cliffs bounding the east coast of the Dunrossness peninsula; it is at the foot of fertile land which slopes gradually down to the coast, ending as a more or less flat strip about half a mile across next to the sea (Frontispiece and Ills. 4.50 and 4.60). There are many modern settlements nearby and the site has a good view out to sea.
The earliest reference to this site is by Sibbald [5, 41] who describes '…ruins of two forts, built by the Picts upon a Rock, close by the sea.' Levenwick was cleared out by Goudie in 1869 [3, 4] (frontispiece) but much of the interior has filled up again with debris in the interval so that the lower features are obscured. Further destruction to the masonry was caused by waves being blown up to the site by the great gale of February 1900 [2] (Ills. 4.54-57).
Removal of rubble began in the centre and revealed a thick secondary wall built against the inner face of the broch wall and with three openings in it, on the north, east and south, all apparently at floor level and leading to corresponding openings in the broch wall. Goudie describes this secondary wall as a 'scarcement' -- a common habit at that time -- but the term has long been reserved for the narrow ledge on the inner face of the primary broch wall.
According to Goudie the main entrance was the passage on the east side but it can no longer be seen. It was c. 90cm (3 ft.) wide throughout, but no trace of door-checks in its walls are mentioned in the contemporary accounts and the plan of the site shows none. Two lintels of the long passage were found in situ, one at the inner end of the primary passage and another at the inner end of its secondary inward extension.
On the north side Goudie found the door to the intra-mural stair, with a wide, curiously triangular gap through the secondary wall leading to it; this stair door is still visible, as is the passage turning right from its inner end. The primary door-way itself was found to be filled with blocking masonry so the curious opening in the secondary wall may have been planned as a cell. Nevertheless it seems unlikely that such a secondary cell would have been placed in front of the stair door if access to this was not still needed; the blocking may belong a to a tertiary stage of occupation, when the upper parts of the broch wall were in ruins and the stair covered with rubble. On the other hand there is no sign of the stairway in the foreground of Goudie's drawing (Frontispiece).
The stair rose to the right from floor level for 2.4 or 3.0m (8 or 10 ft.) and the diggers in 1870 were evidently following a roofed passage upwards; 'eventually 'day-light appeared above'. The stair is described as being roofed with 'overlapping slabs' which term is not entirely clear [4, 16]. It led to an upper mural gallery which is now obscured (not marked on Goudie's plan) and this ran right round the east half of the wallhead, presumably over the entrance passage, to a raised void at 9 o'clock which can still be seen.
Here 15 steps of a second flight of stairs led upwards to the right of the void but these seem to have been destroyed or buried. In 1963 the possible outlines of an oval mural cell were observed at 12 o'clock but this is not referred to in the earlier reports.
Also not mentioned in the earlier reports is a large unexcavated doorway on the south side, under the void mentioned; one side of the latter may have been rebuilt in modern times. This door may have been exposed relatively recently by dilapidation, but it is hard to understand how Goudie did not find it, especially when his drawing (Frontispiece) clearly shows the void or chamber over it; it also shows a secondary doorway leading towards the old main entrance and giving access to a long cell in the wheelhouses wall, so presumably the primary entrance was blocked up. There is also a clear scarcement in the primary broch wall face here, about 30 cm above the primary passage lintel (Ill. 4.62); from Goudie's drawing it seems that the secondary wall came up level with this.
The thick secondary wall which can still be seen running round the interior belongs to an inserted wheelhouse (Frontispiece and Ill. 4.49); it was about 1.8 m (6 ft.) thick, and of a similar height when exposed, and had five stone piers bonded to it and projecting radially into the central area. The length of these piers varied from 0.75 - 1.2 m (2.5 - 4 ft.) so the central area was 3.1 - 3.6 m (10 - 12 ft.) in diameter. Three slabs on edge seemed to have formed a fireplace on the north side of the central area. All these internal features except the circumferential wall are now buried under debris.
The broch is surrounded on the landward side by a mass of 'out-buildings', now reduced to turf-grown rubble; one of them looks like a celled hut ('courtyard house') of Shetland type. There are two concentric stone walls beyond these and partly surrounding the broch [1, plan]; they now run round from north to south around the west arc, and the missing segment has presumably been destroyed by the sea.
Structural analysis: in the light of our modern understanding of broch architecture Goudie's account is reasonably clear and is made more so by his excellent sketch (Frontispiece); Ill. 4.50 shows a reconstructed plan of Levenwick at Levels 1 and 2. There is little doubt that the structure was well preserved in 1869, standing at least as high as the top of Level 2 (perhaps 3.7 m or higher) and perhaps including a small part of Level 3 (the second floor gallery). There is no mention of any systematic exploration of the outer face of the broch wall which is evidently still buried under huge quantities of stone debris.
Level 1: The ground plan makes much better sense if it assumed that this solid-based broch had a secondary doorway forced through the wall on the east, presumably after the high, hollow wall had been largely pulled down. The absence of any mention of door-checks in this passage, and the omission of such from the ground plan (Ill. 4.49), makes this fairly clear; similar secondary passages without door checks are found in at least two of the brochs near Keiss in Caithness (ND36 5 and ND36 6) [7].
The primary entrance, not observed by Goudie, seems to be the opening on the south side (its innermost lintel is now visible) and it is clear that the wall here was standing sufficiently high in 1870 for this to be preserved intact, together with most of the gallery above it and the void to the interior to remain. Clearance of rubble at the outside of the wall would quickly show whether this hypothesis is correct; a massive outermost lintel should appear and the passage should still be completely roofed. At 9 o'clock there may well be an oval mural cell inside the solid wall base.
At 12 o'clock, according to this assumption about the main entrance, is the intra-mural stair which was covered over in 1870; since this rises through Level 1 through a solid wall base one may suppose that stone lintels rather than corbelled masonry roofed it, leaving a gap of about 6 ft for people to walk up the stair.
Level 2: Not mentioned in earlier reports is the scarcement on the inside wall face which can be seen beside and just above what is assumed here to be the primary entrance (Ill. 4.62). It is difficult to know how high it is above the primary floor level, but, as usual, it can be taken to mark the start of level 2 (perhaps slightly above the lintels of the entrance passage).
This first flight of the intra-mural stair led to a raised intra-mural gallery which ran round the wallhead clockwise from 12 to 6 o'clock; in other words this must have been a long landing (the longest known in any broch), the floor of which, presumably about 6 ft above the ground, was the rubble of the solid wall core of Level 1. Goudie states that this gallery was continuous so, if the entrance at 3 o'clock was a secondary one, this must have been pushed through under this gallery.
One might assume that the raised void on the south -- at 6 o'clock and above what is assumed here to be the main entrance -- evidently marks the point at which this long landing gives access to the interior, and presumably at one time to a ring-shaped raised wooden floor resting on a scarcement in the primary broch inner wall face (marked on the Level 2 plan: Ill. 4.50). Immediately west of this raised door, which still has a lintel in position (Ill. 4.56*), fifteen steps of the next flight of the intra mural stair were preserved in 1869, running clockwise from the opening and up to Level 3. Judging from other visible broch stairs it probably ran all the way up to the top of the wall in one flight. Assuming that a broch stairway has about three steps per foot this would take Level 2 up another 5 ft at least.
The only problem with this interpretation is that Goudie's drawing (Frontispiece) does not seem to show the gallery entering the chamber over the entrance from the left, though the details of the masonry are not absolutely clear.
Finds. Goudie seems to have found nothing but a few fragments of wood but this is perhaps not so surprising since his operations seem mostly to have been devoted to clearing rubble; no floor levels are described clearly.
Dimensions. The broch has an external diameter of 54.5 ft and walls from 12 - 16 ft. thick: its wall proportion is thus c. 51.4%.
Sources: 1. OS card HU 41 NW 3 (with plan): 2. RCAHMS 1946, vol. 3, no. 1144, 24-5 and figs. 486 and 499: 3. Goudie 1872, 212-19: 4. Goudie, 1904, 14-18 and ground plan: 5. Sibbald 1829
E W MacKie 2002
